Training programs for Road Safety Professional

Exam (Level 1):

RSP1-SA

Phase One
21-25/02/2021

1
 Module 1: Road Safety Fundamentals
Course 1: Basics of Road Safety

Overview

Unit Modules Topics in this Module

1-Basics of Road Safety The Importance of Road Safety

Evidence-Based Road Safety

2-Effects of Human Behavior on Road Safety

Contributing Factors of Traffic Crashes

3-Data and Its Role in Measuring Road Safety

Principles of Road Safety

4-Methodologies of Solving Road Safety Problems Road Safety Partners

Road Safety Management

5-Traffic Safety Strategic Plans and Programs

Safety Versus Other Transportation Goals

Road Safety Culture

Road Safety Future

2
 Basics of Road Safety

This module is concerned with the fundamentals of road safety. It highlights the definition of road safety
and describes the impact of the principles of road safety on—and the contributing factors to—traffic
crashes. It also provides an overview of the main terminologies related to road safety. The following topics
are covered:

• Define road safety using an approved reference source.

• Explain evidence-based road safety, including the distinction between nominal and substantive
safety, using road safety literature.

• Illustrate the contributing factors to traffic crashes as well as the influence of traffic crash types
and multidisciplinary characteristics on crash severity.

• Understand the principles of road safety related to driver, vehicle, and the road and their
influences on the selection of safety countermeasures.

• List road safety partners and their related disciplines.

• Discuss road safety management and its integrated role of improving road safety.

• Explain the balance between satisfying safety and other transportation goals (e.g., mobility and
environment) based on the scientific evaluation of safety benefits and on costs for better
decisions.

• Demonstrate the elements of a culture that enhance road safety within local communities or
organizations.

• Discuss developments in policy and technology that would affect future decisions and actions in
road safety.

3
The Importance of Road Safety

Basic Definition of Road Safety

The basic definition of safety is the absence of danger or risk. In terms of transportation background, road
safety can be defined as the ability of a person to travel without risk of serious injury or death. Several
transportation agencies have a goal of zero deaths on the road. Transportation also plays a vital role in the
movement of people and goods (freight) over space, and in undertaking their economic, social, and cultural
activities. However, travelling has always been associated with some level of risk in order to participate in
a timely manner in some desired activity. To reflect the reality of risky behavior, road safety can be
considered as the data, i.e., the performance measures and decision-making tools that can be employed to
mitigate fatalities and serious injuries within the road environment. This risk varies among road user groups.
For example, what may be considered as safe for drivers may present an increased risk or decreased level
of safety for other groups of users such as vulnerable road users (VRUs) or non-motorized road users (e.g.,
pedestrians and cyclists).

The definition of road safety varies among disciplines and the number of organizations depends on the
context and the goal of each organization which can be found in their publications. Most of the organizations
including the Institute of Transportation Engineers (ITE), Federal Highway Administration (FHWA),
National Highway Traffic Safety Administration (NHTSA), and the American Association of State
Highway and Transportation Officials (AASHTO) have a comprehensive definition of road safety which is
similarly defined among them but may focus on differing elements or emphasize differing aspect of safety
relative to some of the definition. For instance, the Highway Safety Manual (HSM) which is published by
the AASHTO defines safety as the crash frequency or crash severity, or both, and collision type for a
specific time period, a given location, and a given set of geometric and operational conditions.

Exercise:

Write down a definition of safety or road safety from your own perspective. As you engage in certain
aspects of road safety, summarize your thoughts on the definition of road safety for anyone who has heard
this term.

In the literature, the term “traffic accident” commonly refers to a transport collision. Nevertheless, recently,
“traffic crash” is becoming preferred over “traffic accident”. The reason for using the word “crash” is that

4
"accidents" has connotations of being unavoidable, without cause, and thus unpreventable (Hauer,
2015). The use of a particular word does not make any difference to the main concept. So, both terms are
used interchangeably to indicate an unintentional transport collision.

Traffic Crash Consequences

Traffic crashes are considered one of the top ten contributors to the total death toll in the world. By 2030,
it is expected to be one of the top five causes of death. A report published by the World Health Organization
(WHO,2015) envisages that the road traffic crash death toll will rise from the 1.3 million reported in 2004
to 2.4 million by 2030. Such a huge increase is believed to be a result of the increasing numbers of vehicle
ownership and the vehicle use associated with economic growth in developing countries.

Furthermore, more than one-third of road traffic fatalities in developing countries were mostly pedestrian
and cyclist related. Due to the ever-increasing demand for both travel- and transportation-related
developmental infrastructures, ensuring the efficient movement and safety of road users has proven to be
challenging for governments. In addition, the aftermath of increasing vehicular operations, such as traffic
congestion, air pollution, and oil dependency have worsened the situation.

Nearly half of registered motorized vehicles are in high-income countries (HICs), and only 10% of road
traffic deaths occur there (WHO, 2015) as shown the Figure 1;

Figure 1: Registered Motorized Vehicles Vs. Road Traffic Deaths for Countries Based on Inome (WHO,
2015)

The road traffic death burden in the Kingdom of Saudi Arabia (KSA) was 65% higher than the average of
the world's road traffic deaths and 3.5 times the average of HICs in 2015. However, in 2018 the fatality rate

5
per 100,000 population was equal to 18.03. Figure 2 displays this rate for various countries including
Saudi Arabia in 2018*.

35

30

25 Saudi Arabia = 18.03

20

15

10

0
Austria

Italy

Qatar

Uruguay
Luxembourg
Portugal

Bangladesh
Netherlands

Spain

Kazakhstan

China

Iran
Iraq

Mozambique
Switzerland

Kuwait

Lebanon

South Africa
Singapore

Finland

Australia

Greece
Egypt

Indonesia
Poland

Turkey
USA

Tajikistan

Malaysia
Ghana
Eritrea

Libya
Ethiopia
Ireland
Japan

New Zealand

Oman

Russian Federation
Saudi Arabia

Morocco
Brazil

Nigeria

Sudan
Norway

France
Sweden

Denmark
Germany

Iceland
Hungary

South Korea

Mexico
Ukraine
Pakistan

Tunisia

Madagascar
Canada

India

Jordan
Sri Lanka

Gambia
Thailand
Zimbabwe
United Kingdom

Philippines

United Arab Emirate

Figure 2: Fatality Rates Per 100,000 Population for Various Countries, (WHO,2018).*Based on Saudi
Officials

In Saudi Arabia, rapid population and economic growth has led to an increase in travel demands which, in
turn, have led to a high rate of traffic crashes. According to the Central Department of Statistics and
Information, KSA’s population in 1970 was rounded up to approximately 5,745,000 yet had risen to
33,413,660 by 2018 with 38% being non-Saudi. In addition, KSA's main roads had an estimated road
network length of 71,500 km in 2018. Figure 3 shows the trend of many national measures such as total
population, registered vehicles, and traffic crashes in KSA over the last five decades. During this period,
there was an increase in the total number of crashes, fatalities, and injury crashes. However, the total number
of fatalities has been relatively low within the regions, particularly over the last few years.

6
 100,000,000

10,000,000

1,000,000

100,000

10,000

1,000

100
1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
Year (AD)

Population Registered Vehicles Total Crashes Total Fatalities Total Injuries

Figure 3: Trend of Total Population, Registered Vehicles, and Traffic Crashes in KSA from 1970 to 2019,
(Al-Shammari et al., 2019), GaStat.

Based on the General Directorate of Traffic’s statistics, there were 5,754 fatalities and 32,910 injuries in
KSA due to traffic crashes in 2019, which means 16 persons killed per day and 92 injured, including
serious injuries that resulted in permanent disability. In comparison, there have been 6,397 COVID-19
epidemic fatalities in KSA over the last 12 months. The number of road fatalities is close to the number of
fatalities due to COVID-19 but public outrage is absent.

Exercise:

What are your thoughts on these numbers? Discuss these results of the toll number and trend of fatalities
and injuries. Compare with other countries. What are the main effects/factors (e.g., motorization
development, human behavior, economic growth.)

One of the most important aspects of safety is understanding its combination action and multidisciplinary
perspectives. Over the two decades, there have been many improvements compared to the previous period
in terms of developments of better vehicles, good roads and infrastructure, safety policies and practices,
advanced engineering solutions, as well as medical and emergency services. In addition, developments of
safety standards and tools such as airbags (invented in 1952 and becoming mandatory in 1998), seat belts
(invented in 1959 and becoming mandatory in 1968), anti-lock braking system (ABS), etc., have

7
definitely improved safety. Certainly, the issue of traffic crashes still exists but the probability of
saving any person involved in a crash (crash worthiness) has improved drastically over the last two
decades. Nevertheless, it is vital to understand the external factors that can contribute to increasing or
decreasing the number of. It is a challenge for traffic engineers and safety engineers to capture the effect
of interventions out of what is happing for many reasons. For example, crash frequency might be affected
by the economy, where a drop in the number of available jobs or higher gas prices can result in less
driving, which may improve statistics. So, more caution is required for safety-related agencies to evaluate
the exact effect of specific countermeasures and to measure the percent reduction in total (and fatal)
crashes after implementing the treatment. In this case, the evaluation should account for the external
factors that may contribute to a reduction/increase in crash frequency in order to determine the benefits of
treatments and to repeat and improve them further in other locations.

Cost of Road Crashes

As discussed in the previous section there are a number of losses due to traffic crashes and we can convert
these numbers to monetary value in order to determine the cost-benefit analysis (BCA). Road traffic crashes
have a substantial social and economic impact on the individual, on families, and on society in general. The
social costs of road crashes in HICs varies from 0.5% to 6.0% of the GDP with an average of 2.7%. For
lower- and middle-income countries (LMICs) it ranges from 1.1% to 2.9% of the GDP as shown in Figure
4. With respect to the situation in KSA, the total economic and social costs of road traffic crashes is around
S.R. 76 billion (59.25% of which is direct costs), and this is equivalent to 2.88% of the Gross Domestic
Production in 2018 based on a study conducted by the national road safety center (NRSC).

4.6

3.2 3.3
2.9
2.6 2.7
2.1 2.2 2 2.1
1.7 1.7 1.8
1.4
1.1
0.5
USA

Vietnam
Australia

UK

Austria
Netherlands

Thailand

Philippines

Indonesia
Lao PDR
Germany

Myanmar
New Zealand

Switzerland

Belgium
Singapore

Cambodia

Figure 4: Total Costs of Road Crashes and Share in GDP for Various HICs and LMICs, Wijnen &
Stipdonk, 2016

8
These costs of crashes to society can be used to estimate both the direct cost (materialistic/tangible values)
based on various components including the loss of productivity, medical and emergency costs, insurance
costs, property damage costs, and administration costs. In addition, there may be indirect costs (intangible
costs) where there is a need to determine the cost of pain, grief and suffering due to the actions of the
individuals involved in fatal or injury crashes and their family members. An example of a framework for
estimating costs of road crashes is shown in Figure 5.

Human Production
Capital cost loss

Direct costs Medical cost

Restitution Property
cost damage
Approach
Admistrative
cost

Willingness-
Indirect costs Human cost
To-Pay (WTP)

Figure 5: The Framework for Estimating Costs of Road Crashes

Causalities Due to Traffic Crashes

Fatalities

In general, there are at least three outcomes when a crash occurs. The first case is when at least one person
is killed (fatal crash). The person can be the driver, passenger, pedestrian etc., in any type of crash, e.g.,
single vehicle, multi vehicle, rear-end, head-on. The second case is when no fatality results from the crash
but at least one person is injured (injury crash). There are differing levels of injuries: serious, moderate, and
minor. In the last case when there is no injury, there is only property damage (PDO crash). Usually there is
considerable emphasis on fatalities as death is the most serious consequence of a traffic crash. In addition,
fatal crash data is the more complete data compared to data of crashes at other levels.

9
For a death to be included in any fatality data set, it must occur within a specified period after the road
vehicle crash in which the injury occurred. A one-year criterion is used by the National Safety Council. It
is mainly from the medical perspective and not engineering. One of the drawbacks of using such a period
is that the data file is not complete for an entire year after the crash was recorded. However, a 30-day period
is commonly the recognized choice. This gives a cutoff point to close the crash report and complete the
data set. The World Health Organization (WHO) and Fatality Analysis Reporting System (FARS) data by
the National Highway Traffic Safety Administration, (NHTSA) define a death due to traffic crash as one
that occurs within 30 days of the crash.

Worldwide Deaths Due to Traffic Crashes

Around 1.35 million people are killed annually in traffic crashes, according to the WHO. It is the leading
cause of death for children and young adults aged 5-29 years. As discussed earlier, Figure 2 illustrating the
fatality rates per 100,000 for various countries, shows the impact of traffic crashes around the world. Thus,
there were several road safety action plans including "The Decade of Action for Road Safety 2011–2020,
2021-2030" which was officially declared by the UN General Assembly in March 2010, and August 2020,
respectively. Its goal is to “Stabilize and mitigate the forecast level of road traffic deaths around the world
by building road safety management capacity; improving the safety of road infrastructure; further
developing the safety of vehicles; enhancing the behavior of road users; and improving post-crash
response", what known as "the five pillars", (WHO, 2018).

Although the traffic crash issue is commonly considered as an engineering problem, it is also a public health
problem affecting the population of the world just like any disease that causes death such as heart disease,
and diabetes.

It can be seen from the

Table 1 that the number of fatalities due to traffic crashes in itself does not give any sense when comparing
over various years or between countries. Therefore, statistics can play a vital role in understanding these
numbers by normalization or crash rate, or by using appropriate exposure such as number of registered
vehicles (e.g., number of deaths per 1000 vehicles), population (fatalities per 100,000 people), etc. The
most common and very useful measure of exposure is vehicle-kilometers traveled (VKT) (e.g., fatalities
per billion vehicle kilometers driven). It is possible that differences between countries are even greater than
the data indicated, because under-reporting of fatalities is likely to be greater when a country is less
motorized.

10
Table 1: Fatality Rates for Various Countries From OECD Data

Fatalities Vehicles per Fatalities per Fatalities per

Country Data year
per year 1000 capita 10,0000 vehicles 100,0000 capita
Argentina 5,619 2013 559.67 24.3 13.6
Australia 1,351 2016 756.76 7.4 5.6
Austria 455 2016 885.25 6.1 5.4
Brazil 46,935 2013 406.96 57.5 23.4
Canada 2,118 2016 651.69 8.9 5.8
China 261,367 2016 179.90 104.5 18.8
Egypt 10,466 2013 86.08 148.7 12.8
Ethiopia 27,326 2016 69.22 385.7 26.7
Germany 3,327 2019 578.13 6.4 3.7
Ghana 6,789 2013 59.13 443.1 26.2
Hungary 756 2019 299.52 20.7 6.2
India 207,551 2016 127.59 130.1 16.6
Indonesia 38,279 2016 416.89 36.7 15.3
Ireland 194 2019 386.67 7.5 2.9
Japan 5,224 2016 719.30 5.7 4.1
Kenya 12,891 2013 45.42 640.7 29.1
Malaysia 7,374 2016 791.95 29.8 23.6
Mozambique 8,173 2013 20.97 1507 31.6
NZ 364 2016 780.00 10 7.8
Nigeria 35,621 2013 33.31 615.4 20.5
Norway 110 2019 666.67 3 2
Russia 27,025 2016 378.00 50 18.9
Saudi Arabia 7,898 2013 228.91 119.7 27.4
Singapore 197 2013 178.22 20.2 3.6
Slovakia 330 2019 354.33 12.7 4.5
South Africa 13,273 2013 187.45 133.9 25.1
SK 3,349 2019 500.00 13 6.5
Spain 1,922 2019 637.93 5.8 3.7
Sweden 223 2019 478.26 4.6 2.2
Tanzania 16,211 2013 30.64 1073.7 32.9
Turkey 9,782 2016 265.09 46.4 12.3
UAE 1,678 2013 288.68 62.7 18.1
UK 2,026 2019 508.77 5.7 2.9
US 39,888 2018 873.24 14.2 12.4
Zimbabwe 3,985 2013 65.61 429.8 28.2

In addition, the same data can be used to understand the relationship between fatalities and number of
vehicles, as shown in Figure 6. It can be shown that as a country has more motorized vehicles, the safety
record seems better and that may be due to the degree of development of the country. So, several lessons

11
can be learned from the countries that have more motorization and are safer. For example, Sweden
introduced a "zero vision" policy in 1997. As motorization increases, not only does the number of fatalities
change, but the types of fatalities also change. Because the problem for each type may differ, the treatment
may also differ.

100

Mozambique
Tanzania
10
Nigeria Kenya

Ghana Zimbabwe

Ethiopia
fatality per 1000 vehicles

Egypt South Africa

India

Saudi Arabia
1 China
UAE
Russia
Brazil
Turkey Malaysia
IndonesiaArgentina
Singapore Hungary

SK US
Slovakia
NZ
Canada
0.1 Ireland
Australia
Germany Austria
UK Japan
Spain
Sweden

Norway

0.01
10.00 100.00 1000.00
vehicles per 1000 population

Figure 6: Fatality Rates Vs. Motorization for Various Countries

Fatalities Among Road User Groups

Transportation professionals must understand the mobility and safety needs of various user groups and how
they interact with one another to gain a better understanding of safety problems, along with their potential
solutions. Understanding should include: 1) Passenger vehicle drivers and occupants (the highest
percentage of total traffic fatalities, 2) Drivers of trucks and other large vehicles, 3) Motorcyclists, 4)
Pedestrians, and 5) Bicyclists. Fatalities among road user groups are shown in Table 2 (Carter et al., 2017).

12
Table 2: Fatalities Among Road User Groups in USA,

Number of
No Road user Number of trips Fatalities Year
vehicles

Drivers of trucks and other 4% of registered

1 9% of VMT* 12% 2013
large vehicles vehicles

3% of registered
2 Motorcyclists 0.7% of VMT 15% 2012
vehicles

11% of trips in US (of

3 Pedestrians - 14.1% 2012
which third < 1 mile),

4 Bicyclists - 1% of trips in US 2% 2012

* VMT: Vehicle-Miles-Traveled

In recent decades, an active mode of transportation (e.g., walking and cycling) has been encouraged because
they are sustainable, inexpensive, environmentally friendly, and most importantly they curb the alarming
transportation-related problems. Promotion of public transport is also important for reducing road crashes.
A recent study by Marshall (2018) shows that some inter-related factors play a role in improving road safety
as in Australia where there is a more urban population and multimodal infrastructure. Australia tends to
discourage driving mileage and exposure while encouraging safer modes of transportation, such as transit,
compared to USA, as shown in Figure 7.

13
 A) Fatalities per 100,000 Population

B) Fatalities per 10,000 Vehicles

Figure 7: Fatalities Rates in Australia and USA

14
Pedestrian Fatalities

Exercise:

Discuss Table 2 in terms of each type of road user’s fatalities and relate it to the other provided attributes.
How can you see the numbers if they are from a developing country? For example, the main issue in
developing countries is mix of modes of transportation (motorized and non-motorized) and fewer vehicles
which means more walking and interaction with vehicles in road segments and at intersections.

Non-Fatal Injuries

Although fatal crashes may form only around 1% of the total crashes, it is important to consider other
severities of crashes and property damage to gains a better understanding of the crash situation. The less
severe the injury, the greater the frequency of occurrence. The data of non-fatal injuries is not as complete
as fatal crash data. It is vital to determine injury—serious, moderate or minor. Serious injuries require more
medical costs as well as rehabilitation and other costs compared to minor injuries.

Property Damage Only (PDO) Crashes

The PDO crash is the most common type of crash. It involves property damage only, without any individual
injury. Estimates of the total number of vehicles involved in crashes are particularly uncertain, because as
the value of the property loss decreases, so does the probability that the authorities will know that the crash
has occurred.

Recently, elements of PDOs have been collected by a private company: NAJM. Two conditions are required
in order for NAJM to handle and report a PDO crash: 1) at least one of the vehicles or individuals that were
involved in the crash has insurance, and 2) there is no injury caused by the crash. NAJM was established
in 2007 and covers 27 cities in the KSA.

Economic Losses:

As discussed earlier, the largest cost of traffic crashes is dues to property damage Table 3. Other costs may
include production loss, medical, and administrative. Production loss refers to the loss of the productive
capacity from individuals who died or were injured due to road crashes. The cost of the production loss
varies depending on the crash severity levels, the ages of the persons involved in the crash and their
incomes, as well as their number of years of work. For example, for fatal crashes, the production loss can
be estimated based on the average years of production loss which is the average ages of those who died in

15
crashes subtracted from the average ages of retirees. It also accounts for the average income, growth
rate and discount rate to be adapted to the present values.

Table 3: Estimates of the Distribution of the Direct Cost of Crashes in KSA, 2018

Category Cost in S.R. Cost in US$ Percent %

Production Loss Cost 9,698,718,895 2,586,325,039 21.59%
Medical Cost 6,145,225,048 1,638,726,679 13.68%
Property Damage Cost 28,646,329,783 7,639,021,275 63.75%
Administrative Cost 441,925,648 117,846,840 0.98%
Total 44,932,199,374 11,981,919,833 100%

Motor Vehicle Crash Avoidance and Crash Worthiness

Crashes are complex events that result from the interaction of driver behavior, the driving environment
(e.g., weather, time of day, type of road) and vehicle design. Driver error or inappropriate driver behavior
(distracted driving and/or reckless driving, etc.) is considered as the dominant factor affecting the likelihood
of being involved in a crash. Vehicle features affect safety in helping the driver to avoid a crash or to recover
from a driving error (crash avoidance) such as braking performance, vehicle stability, and visibility. They
also help in providing protection from harm during a crash (crash worthiness). Characteristics such as
vehicle stability and braking performance affect the probability of being in a crash, assuming all else is
equal. Once a crash occurs, vehicle safety characteristics such as weight and size play a critical role in
determining the protection afforded to vehicle occupants. The extent of injury is directly affected by the
crash energy and the manner in which vehicle occupants experience the associated forces. Heavier vehicles
typically have a larger interior space, thus providing a longer distance for the occupants to decelerate to a
stop and reducing the likelihood of injury. Larger vehicles, with more external energy-absorbing structures,
do a better job of preventing intrusion into the occupant compartment and increasing the time the crash
forces take to reach the occupants. Occupant protection features have been developed to reduce what are
known as the “Second” collision, this is the collision of the vehicle’s occupants against the dash board
and/or windshield of the vehicle. For example, collapsible steering columns and padded dashboards help
reflect or cushion collision impacts. Safety belt systems, which are required in all vehicles, help avoid the
second impact. Airbags further protect the occupant’s upper body in a severe frontal crash by providing an
energy-absorbing cushion. Side airbags protect from angle crashes, etc. Frontal airbags, however, provide
no protection in rollovers, rear impacts, or side impact crashes; only safety belt systems offer this protection.
Door mounted air bags are now being introduced to provide better protection in side-impact crashes.

16
In this regard, it is important for transportation professionals to consider the fundamental characteristics
of the motion of vehicles. That includes the basic kinematic and kinetic equations of motion. These two
branches of dynamics are related: the former is the study of motion irrespective of the forces that cause it,
whereas the latter accounts for these forces. The motion of a vehicle can be rectilinear or curvilinear
(horizontal or vertical curves) (Papacostas & Prevedouros, 2000). The more common rectilinear motion is
the braking distance of a vehicle where it is affected by the speed of the vehicle, the friction between the
pavement surface and the tire, and the grade of the road. These facets all contribute to determining the
stopping sight distance. In the case of horizontal curves, there are two equilibriums that should be taken
into account: 1) the sliding equilibrium, where if the centrifugal force exceeds the centripetal force the
vehicle will slide outward; it is affected by the speed of the vehicle, the radius of the curve, the rate of
superelevation and the side friction, and 2) overturning equilibrium, where vehicles that have a low center
of gravity relative to their track width have a less of a propensity to roll over if the vehicle runs off the road
or collides with a barrier or another vehicle. Crash prevention also involves measures of behavioral nature,
training, licensing, enforcement, engineering, education, etc.

Crash Causation

Factors Rather than Cause

Vehicle crashes are complex events involving driver behavior, vehicle characteristics, and environmental
conditions (including the roadway). Several studies have attempted to assign causality to each of the major
factors contributing to crash likelihood. It is rare to assign direct or clear cause and effect to a crash,
normally a set of contributing factors is involved. It was found that driver errors and inappropriate driving
behavior were the major contributing factors in 60 to 90 percent of motor vehicle crashes. Environmental
factors (e.g., weather, road conditions, signing, and lighting) played a major role in 12 to 35 percent of the
crashes. Vehicle-related factors (e.g., brake failures) were dominant in only 5 to 20 percent of the crashes
(Evans, 1991).

Goal of Safety Analysis

The goal in safety analysis is to examine factors associated with crashes with the aim of identifying those
that can be changed by countermeasures to improve future safety.

Exposure
Knowledge about the number of individuals injured at a specific level is rarely sufficient to answer specific
traffic safety questions (e.g., number of fatalities) without some measures of exposure (normalizing factor),
or of the number of people exposed to the risk of being involved in crashes. As discussed earlier that number
of crashes can be normalized or expressed as a crash rate by dividing by the appropriate exposure measure.

17
Population can be used as an exposure, but it is not preferable compared to using a measure related
more to transportation such as VKT (e.g., fatalities per billion vehicle kilometers driven). If a person drove
10,000 kilometers in a specific year and was involved in two crashes, it cannot be said that he is a less
capable driver than another who drove 2,000 kilometers and was involved in one car crash in the same year.
The former person drives five times further than the latter one, therefore he is more exposed to the likelihood
of being involved in traffic crashes. The rates of crashes per 1,000 kilometers driven for the two drivers are
0.2 and 0.5, respectively. In order for crash data to be meaningful, they must be compared with the
experience of the non-crash population. Another example is to look at the ages of drivers. Middle-aged
people drive more, hence they are exposed more to traffic and are expected to be involved in more crashes
compared to drivers of other age groups (younger and older ages). However, if normalized by their
population weight (e.g., driver license) of each group it seems that younger and older drives may be the
risker compared to the middle-aged group. There is no single normalizing factor that can be applied for
each safety data. It is based on the purpose of the study. With respect to intersection-related crashes,
commonly used data is the total entering vehicles (TEV) to the intersection, as an exposure measure. The
typical unit for a crash rate is crashes per 100 million vehicle miles traveled (VMT) for road segment/section
or crashes per million entering vehicles for intersections.

Total Fatalities Fatality Rate per 100 Million VMT

60,000 6.00

Fatality Rate per 100 Million VMT

50,000 5.00
Total Fatalities

40,000 4.00

30,000 3.00

20,000 2.00

10,000 1.00

0 0.00
1959 1966 1973 1980 1987 1994 2001 2008 2015
Year

Figure 8: Fatalities and Fatality Rate, USA, from FARS Data

18
Exercise:

What is the difference between crashes and involvement (persons and vehicles)?

Evidence-Based Road Safety

Road safety has two dimensions: 1) nominal safety and 2) substantive safety. In the first dimension, road
safety can be considered on the basis of nominal design consideration of the road based on design standards.
However, substantive safety is based on crash data where safety information can be gathered to quantify
the actual performance of the road (HSIP, 2010). Nominal safety can be defined as an absolute statement
about the safety of a particular network, corridor, or intersection based only on compliance with standards,
warrants, and guidelines for design. Yet, substantive safety is the historical and long-term objective safety
of a highway or an intersection based on crash data (i.e., the actual crash frequency, crash severity, crash
type, crash rate). The difference between the two dimensions is mainly based on the evidence of crashes or
other measures of road safety for a given location. In various cases, nominal safety is an essential but not
sufficient characteristic of a safe road, which means even if the road meets the design criteria based on
approved standards and guidance documents, it may still have higher than expected number of severe
crashes. Likewise, a road may be considered as substantively safe when it has experienced a lower number
of severe crashes, nevertheless it may not be nominally safe.

The following Figure 9 illustrates the relationship between nominal and substantive safety. It can be seen
that the line for nominal safety is an absolute and reveals more binary outcomes so the location either meets
or does not meet the current practices and design standards. Substantive safety is more of a continuum to
quantify in a function that accounts for the characteristics of the road and the behavior of the drivers with
respect to the crash risk. Substantive safety requires an evidence-based approach to estimate the expected
safety of a location through data and analysis rather than focusing solely on standards. Therefore,
transportation professionals should strive for both aspects of the road safety dimension (i.e., nominal and
substantive safety) in order to have a nominal safe road as it was designed and then continue to monitor the
road for crashes and other measures of safety to make sure that the road is also substantively safe.

19
Figure 9: Comparison of Nominal and Substantive Concepts of Safety, (Carter et al., 2017)

Crash Causation and Severity

A crash is very rarely caused by a single factor or event. It is statistically discrete, having random events
and non-negative integers. “Rare” in this case means that the majority of daily interactions and conflicts of
the events of the transportation system do not result in a crash. Events are referred to the movement of one
or more vehicles and or pedestrians and cyclists on the transportation network. There is also a strong
element of randomness associated with the occurrence of crashes. “Random” in this case means that crashes
happen as a function of a set of events influenced by multiple factors. Circumstances that lead to a crash in
one event will not necessarily lead to a crash in a similar event. The drivers and vehicles have differing
characteristics (e.g., reaction times, braking efficiencies, visual acuity), that will determine whether or not
a crash occurs. There are two ways to consider contributing factors to a crash: there are the deterministic
factors which can be controllable or predictable, and there are the stochastic ones that are the random and
unpredictable factors that can contribute to a crash.

Figure 10: Crashes are Rare and Random Events, (Part, 2010)

20
The continuum of events that may lead to crashes and the conceptual proportion of crash events to non-
crash events are shown in Figure 10. This inverted pyramid of crash causation and severity relationship
figure starts with the fact that the majority of events in a roadway environment occur with the low risk of a
crash. Following this stage, these events continue to narrow down further and the potential risk of a crash
occurring increases. As an example, when an unexpected change in the traffic flow occurs or an object is
on the road. In this case, the potential for a crash can mostly be avoided by a driver’s reaction or it becomes
riskier but in fewer events. That is can be shown in the bottom of the pyramid. So, it is only a very small
subset of events and circumstances that together contribute to cause crashes.

Crash Severity Scale

Various scales are commonly used for crash injury severity categorization, one of which is the maximum
abbreviated injury scale (MAIS). It was developed by the Association for the Advancement of Automotive
Medicine where the injuries for each body region are placed into a scale of 0 to 6, which represents the
threat to life associated with the injury rather than the comprehensive assessment of the severity of the
injury. It is also an approximate of the percentage probability of death associated with injury. It varies
between no injury at scale zero, which is typically considered as a PDO crash (approximate percentage
probability of death = 0%). Then the scale of severity increases until scale 6, which is the maximum/fatal
injury (percentage probability of death = 100%). Between them are minor, moderate, serious, severe, and
critical, where the percentage probability of death is equal to 0, 0.1%, 0.8%, 7.9%, and 58.4%, respectively.

In order to calculate the MAIS for a crash, the injury reading is taken as the highest individual injury score
that occurs during the crash. Therefore, if there is any fatality resulted in a crash the MAIS should be
recorded as a category 6 and the crash itself scores 6 on the scale. One of the drawbacks of the injury scale
categories is the difficulty in estimating the injury severities, especially at the crash scene and subsequent
probability of death (i.e., the results from those injuries).

The most commonly used scale is the KABCO scale for injury severity. It is mainly from an engineering
or enforcement perspective to be used in the crash report. It also has slightly lower number of
categorizations ranging from (k) fatal injury, down to (o) no injury (i.e., five levels):

K: Fatal crash.

A: Incapacitating/major/serious injury where the person requires hospital treatment, and where
it prevents the injured person from walking, driving, or any normal activities that the person
was capable of performing before the injury occurred.

21
 B: Non-incapacitating injury e.g., hurt, arm broken.

C: Possible injury.

O: Property damage only (no injury).

Crash Types

Contributing factors to road crashes can be investigated based on crash type. That is because contributing
factor sets for the various severity levels of crashes mostly differ according to the type of crash. There are
several crash type categorizations. Crash data can be disaggregated based on the occurrence location; it can
be either roadway segment or intersection related. This will be indicated by the police officer using his
judgment when reporting the crash. The other way involves using the coordinates of the crash and then
specifying the crash location from the map of the GIS of the road network. Recently, the HSM has been
classifying a traffic crash as intersection related if it occurs within 250 feet (76 meters) of an intersection.
An intersection-related crash has a higher risk category of crash, particularly a right-angle one.

The other way of classifying crash types is based on motorized crashes (i.e., single-vehicle, multi-vehicle
crashes) and non-motorized crashes (i.e., pedestrians and cyclists crashes). Crashes are more likely to be
fatal or have serious injuries when they involve vulnerable road users. Other categories are based on crashes
related to roadway departure, i.e., crashes with fixed objects and rollovers. The run-off-road (RoR) or
roadway-departure crash which is a type of single vehicle crash is considered as one of the most lethal types
of crashes. Lastly, crash categories can be based on the collision place, such as head-on, rear-end, sideswipe,
etc. Head-on crashes tend to be lethal on high-speed roadways and in particular on rural roads. Rear-end
crashes may occur frequently at intersections.

Road User Characteristics

Various types of roadway users should be considered when designing and operating a safe roadway
environment. There are five main classifications of roadways users: 1) passenger vehicle drivers and
occupants, 2) large vehicle/truck operators, 3) motorcyclists, 4) bicyclists, 5) pedestrians. All of these
groups have individual behaviors on the road and separate factors and hence must be taken into
consideration. One of the most important factors is the geometric design of the road. As discussed earlier,
this will involve the characteristics related to braking distance, as well as sliding and overturning
equilibrium on horizontal curves. The road features, such as curve radii, lane width and more of the design
features of the roadway itself can also largely be impacted by the type of user that is operating on the road.
Moreover, the traffic control, particularly traffic control devices, should be considered. For example, when

22
expecting a high number of pedestrians at a specific location, such as signalized intersections, an
adequate crossing time must be set.

Another road user characteristic is related to differing visibility considerations with these groups of users,
specifically pedestrians and cyclists. They tend to be at a higher risk of low visibility and have a greater
number of fatalities, not only as a result of automobile drivers who are not being able to see them easily,
but also as a result of the visual and visibility characteristics of the users themselves. The level of visibility
for cyclists on the road differs compared to automobile drivers or even large truck operators. Lastly, the
risk perception is a significant factor related to these groups, especially pedestrians and bicyclists, as they
are more adaptable to the road, which may lead them to engage in certain risky behaviors, such as crossing
outside of crosswalks.

Another aspect of road user characteristics is related to the subset of user groups within the overall driving
and operating population. Usually, the road is built and operated for motorized and non-motorized users
with a design driver in consideration. For example, the 85th percentile approach or speed limit setting and
visual acuity. Two groups of this subset of users are especially considered as a high-risk group: elderly
drivers and young and/or novice drivers. Several studies have analyzed driver death rates in relation to age
and revealed that driver death rates form a “U-shaped” curve where both young and old drivers have high
death rates, as shown in Figure 11 (Tefft, 2008).

Figure 11:Road User Rates of Deaths by Target Driver Age and Type of Road User, (Tefft, 2008)

Each group comes with their own risks and attributes that may contribute to high rates of fatalities and
serious injuries. For example, elderly drivers tend to have a higher rate of failure to yield once they merge
into traffic stream, and that may be due to their limited reaction time. In addition, the extra time for judging
the speed or distance of oncoming vehicles or becoming distracted or overwhelmed by large numbers of
visual stimuli present on the road can lead to confusion and consequently may cause higher risk driving
behavior. However, young and novice drivers tend to generally engage in more dangerous behaviors while

23
driving. They are likely to be influenced by the presence of additional passengers in the vehicle and
perhaps be distracted.

There are also a number of characteristics and activities that can lead to safer or unsafe driving depending
on how they are handled by the road user. Perception reaction time (PRT) is perhaps one of the most
important characteristics. It can be defined as the time needed before actual vehicular
deceleration/acceleration takes place. There are four main elements/components of the perception reaction
time (i.e., PIJR): 1) perception or recognition or detection, 2) identification, 3) judgment or decision, and
4) reaction or response. Driver response is related to driver characteristics and conditions. Several factors
contribute to determining how long it will take a vehicle to stop for an object or obstacle in the roadway,
for example, driver age, medical conditions, fatigue, sleep deprivation, and emotional conditions. Other
factors include the vehicle and roadway characteristics, such as braking and friction. Lastly, PRT depends
on the complexity of the stimulus, driver expectancy, and the complexity of the required response. The
recommended values of PRT for setting the stopping sight distance on highways and signals are 2.5 and 1
second, respectively.

Regarding driver expectancy, the time to react to unexpected information is obviously longer than the time
to react to expected information. The complexity of the given information also has a positive relationship
with reaction time (and conceivably a positive relationship with accident risk) (Papacostas & Prevedouros,
2000). Therefore, the larger the quantity and complexity of the information, the more time drivers need to
comprehend the information and react accordingly, as shown in Figure 12.

24
 Figure 12: Eighty-Fifth Percentile Driver Reaction Time to Expected and Unexpected Information,
(Papacostas & Prevedouros, 2000)

Vision and visual behavior are also important characteristics of road users. This refers to the visual acuity,
peripheral vision, and contrast sensitivity of the road user and his ability to adapt to light conditions and the
effect of glare on the driving environment. Regarding the visual acuity, the clarity of vision is fairly good
up to approximately 10o which can be used for traffic sign placement, Figure 13.

Figure 13:Cones of Vision, (Papacostas & Prevedouros, 2000)

25
Road Safety Partners

It is essential for transportation professionals to know the key players and road safety partners in order to
collaborate with them in developing safety plans and in designing and operating the roads to improve road
safety. Part of that process is to engage with a multidisciplinary stakeholder team. This team can assist in
developing and understanding overall community goals; for example, assisting in having safer roads,
encouraging walkability and biking, and being friendly for the pedestrians and cyclists particular areas, at
the same time ensuring the safety of these road users.

In addition, the stakeholder teams can assist with initially identifying problems, which includes responding
to any public concern, such as reporting excessive speeding or sharing related data. Each agency works on
individual focus areas at individual levels related to road safety, such as crash data, highway geometric
characteristics, traffic volumes and any other types of information that can help to identify potential safety
problems. This will help in reducing the time for collecting such data by each agency and in preserving
financial resources. Road safety can be improved significantly when transportation agencies work together
rather than tackling problems individually.

The stakeholder team can also assist with the selection of the potential countermeasures and evaluating and
prioritizing their effectiveness. They can help in implementing, installing and deploying these treatments
or policies. It is not at the level of monitoring and doing qualitative evaluation of the countermeasure, but
it is the continuing process of data collection. If there is enforcement aspect of countermeasure—
particularly law enforcement—safety partners can assess to ensure the countermeasure is achieving its
intended effect. All these areas should be accounted for when planning for safety countermeasures, both
strategically and programmatically, as well as implementing them for ongoing monitoring and evaluation.
The road safety partners can vary based on their responsibility—either governmental and/or private or on
their areas of expertise such as engineering, education, enforcement, and emergency services.

The dimension of road safety partners is based on various focus areas of roadway safety including road
design and environment, road user behavior, vehicle design and technology, law enforcement and transit
safety. In USA, each dimension is represented by at least one agency over all levels (Federal, State, and
local) Figure 14.

26
 Figure 14: Transportation Agencies by Focus Area in the USA, (Carter et al., 2017)

Approaches to Road Safety Management

Road safety management refers to the process of identifying current and potential safety problems,
particularly those resulting in fatalities and serious injuries, selecting appropriate countermeasures for
issuing, implementing and evaluating the use and effectiveness of these countermeasures. It is a
comprehensive approach to ensuring roadways are substantively safe based on the use of many types of
crash data roadway information and performance measures collected by the road operator.

There are various approaches to road safety management. There is site-level safety management, where the
focus is on high-priority locations, such as intersections, curves and corridors. At the system level or
network level, safety management focuses on issues affecting the broad transportation system which can
be based on specific geographic units, such as traffic analysis zones (TAZs).

The most common approach to road safety management is the 4 E's approach which has been taught in
transportation classes for decades: 1) engineering, 2) education, 3) enforcement, and 4) emergency
response.

Another approach to road safety management is known as the Haddon’s matrix. This is a framework to
identify possible contributing factors related to the driver, the vehicle and the roadway environment which
are cross referenced against possible crash conditions before, during and after crashes, to identify possible
reasons for the events.

Lastly, the safe system approach is a recent approach to road safety management. It is strongly linked to
the “Zero Vision” concept which basically aims to have zero fatalities due to traffic crashes. The “Safe

27
System” approach is based on understanding that human error is inevitable as long as vehicles are driven
by human beings. That does not mean that humans drivers have no responsibility to operate safely.
However, when the error occurs it should not result in death or serious injury to the roadway user or other
users. Therefore, the safe system approach requires that the road should be designed, built, and operated to
make crashes avoidable or at least survivable. Further analysis and explanation of the mechanism of the
tools and techniques in road safety management will be covered in another module.

Balance Between Safety and Other Transportation Goals

Economic analysis of safety

There are several examples of trade-offs associated with safety in a road environment to meet society’s
needs for mobility. These may include cost, convenience, accessibility, and environment protection.
Regarding the environment, protection of flora and fauna or any historical building and other community
values that may be in conflict in some cases with safe operation of the roadway. Thus, this is where more
qualitative analysis of safety impacts can be a strength and a useful tool for the transportation professional.
Readers are referred to the (Carter et al., 2017) for more trade-off examples.

One of the first steps in conducting an economic analysis of safety is quantifying safety impact of safety
countermeasures. Two common tools that will be adapted and used together in such analysis are safety
performance function (SPF) and crash modification factor (CMF). The SPF is an equation used to predict
the average number of crashes per year at a location as a function of exposure and roadway characteristics.
The CMF is A multiplicative factor to compute expected number of crashes after implementing a
countermeasure. First, developing a SPF to estimate the total number of crashes on a particular road segment
or intersection without safety treatments and then, based on that predicted number of crashes, the CMF
would give an estimate of the percentage of crashes that would be reduced through the implementation of
the safety countermeasure.

The following step is to assess the crash costs for each severity level. There are two categories of crash
cost: the economic costs, which refers more to the direct monetary impact of a crash, such as property
damage, loss of productivity as well as medical costs associated with crash occurrence. There is also the
harder-to-define quality-of-life cost, which mainly refers to quality-adjusted life years (QALY). All the
previous steps and values will be used in calculating the benefit-cost analysis (BCA). The indirect cost can
be estimated using the willingness-to-pay approach (WTP) and estimating the value of statistical life
(VoSL).

28
Road safety culture

Road safety culture refers to each road user's choices and the attitudes of society to transportation safety.
In order to achieve the goals of road safety, a culture of safety needs to be adopted where individual and
officials who govern transportation safety agree to have no tolerance for fatalities due to vehicle crashes as
a price for mobility. It consists of the beliefs, values, norms, and things people use, which guide their social
interactions in everyday life (Hedlund, 2007). There are five points that can explain the safety culture:
culture is never naturally given, never singular, never neutral, always an effect of power and is best modified
through changes in social practice (Mitchell, 2004). Safety culture is considered as a component of an
organization or groups, and at the community or societal level it can be continually improved rather than
simply instilled, yet it varies among cities or countries (Wiegmann et al., 2007).

There are several strategies for implementing a positive agency safety culture, based on FHWA:

• Improving the accessibility to safety data and safety analysis tools for agency employees.
• Building and maintaining organizational knowledge of road safety.
• Building and maintaining communications tools across an organization.
• Implementing comprehensive, formalized road safety management policies and procedures.
• Working internally and with external partners to facilitate ongoing road safety research.
• Ongoing evaluation of program effectiveness.

Future trends in safety policy and technology

In the future, new technologies will be under development as part of the intelligent transportation system
(ITS). This will include heads-up instrument display panels, enhanced night vision systems and collision
avoidance systems, which may all enhance driver ability to avoid crashes. In addition, there will be
autonomous and connected vehicles (ACV) to improve safety and mobility. The CAV involves four
different types of technologies: 1) Vehicle-to-infrastructure (V2I) technology which enable vehicles to
communicate with roadside equipment unit. The second is Vehicle-to-Vehicle (V2V) technology that allow
two adjacent vehicles to communicate with each other. In addition, the Vehicle-to-Pedestrian (V2P)
technology. Lastly, the Autonomous/driverless/self-driving Vehicle (AV) technology. However, there are
still many challenges regarding the technology development and also the ethics that required a lot of work.

There are several dimensions related to the trend in safety policy and technology, including the promotion
of the active transportation modes, i.e., walking and cycling. Other dimensions include graduated driver
licensing, driver impairment and distraction, strategic agency reorganization and dilemma zone protection.

29
 Module 2
Human behavior and Road Safety
Overview

Unit Modules Topics in this Module

1-Basics of Road Safety Key characteristics and limitations of human

behavior

Multidisciplinary safety strategies

2-Effects of Human Behavior on Road Safety

Key characteristics of effective educational

strategies
3-Data and Its Role in Measuring Road Safety
Key characteristics of effective enforcement
campaigns

4-Methodologies of Solving Road Safety Problems Roadway infrastructure features and elements
affect human behavior

Human factors in the process of planning, design

5-Traffic Safety Strategic Plans and Programs and operations

Applying positive guidance principles to road

elements

Applying the driving task model

30
 Outline
1 Human Behavior in Road Safety
(Corresponds to Module 2.1 and 2.6)

1.1 Introduction
1.2 Why human factor is critical? Let's explore who is involved in the road accidents
1.3 Personality traits
1.4 The Driver Distraction factor
1.5 The Speed factor
1.6 The Age factor (Old-Young)
1.7 The Fatigue factor
2 Human Behavior Modification - Principles to affect human behavior (Corresponds to
Module 2.2 and 2.7)

2.1 Public empowerment

2.2 Creation of road safety culture
2.3 Safety System Approach
2.4 Road Safety Culture
2.5 Improving the safety of vulnerable road users
3 Educational Strategies
(Corresponds to Module 2.3)

4 Road Safety Campaigns

(Corresponds to Module 2.2)

5 Road Safety Enforcement

(Corresponds to Module 2.4)

6 Roadway Features and Human Behavior

(Corresponds to Module 2.5)

6.1 External distraction - advertising signs, billboards etc.

6.2 Road safety engineering strategies
6.3 Infrastructure and Pedestrian Safety
7 The driving task model approach
(Corresponds to Module 2.8)

8 Conclusion

31
1. Human behavior in road safety
1.1 Introduction
Human factors research examines the way people interact with various aspects of the world and
aims to make these interactions safer, healthier, and more efficient. This interdisciplinary field of
research has a wide scope of application, spanning road safety, healthcare delivery, physical,
cognitive, and technological systems. In the context of safety features and road safety, human
factors research aims to understand the driver’s role in the safe operation of his or her vehicle.
Various factors contribute to how a person behaves in the driver’s seat, including environmental,
psychological, and vehicle design factors. The goal of human factors research is to expose these
factors, determine the extent of their influence on driver performance, and modify road or vehicle
design to reduce unsafe behavior and improve driver performance. Vehicle safety features are part
of a safe-driving system that includes human factors: the various ways that drivers interact with
these features will help determine both how safe the driver is and how effective those safety
features are.

The Traffic Injury Research Foundation’s (TIRF) nation-wide 2012 survey revealed that some
drivers admit that they would engage in unsafe driving practices like distracted driving and
speeding if they knew that their vehicle was fully equipped with modern safety features. This is a
clear example of human factors affecting the overall amount of safety benefit that these drivers
can expect to see as a result of driving a vehicle with safety features. However, not all human
factors are this obvious. A general lack of familiarity with how safety features work may have a
subtle yet negative influence on a driver’s ability to benefit from safety features.

Human factors are not necessarily limited to unsafe driving behavior like excessive speeding and
tailgating. Age, driving experience, attention level, and vehicle maintenance can all have an effect
downstream on the performance of safety features. Drivers are an indispensable part of the road
safety system, so factors that affect drivers accordingly affect road safety in general.

• Examination of the operator plays a large role in transportation psychology. While many
external factors influence traffic safety, internal factors are also significant. Some factors
include:
• Decision-making
• Demographics
• Distraction
• Detection Thresholds
• Drugs and alcohol
• Driving training and experience
• Familiarity with vehicle and environment
• Fatigue
• Inattention

32
 • Perception-reaction time- response to the unexpected
• Risky behaviors
• Stress and panic

The contributing factors for a road accident are inside the Human – Environment – Vehicle system.
The Accident contributing factors inside the Human – Environment – Vehicle system are presented
below:

Human behavior is largely guided by two different systems: a deliberate, rational system
(deliberative) and an implicit, unconscious system (intuitive). Driver expectancy is defined as
“driver’s readiness to respond to situations, events, and information in predictable and successful
ways”. If drivers fail to recognize the violation that the geometric alignment is not consistent with
their expectancies, the likelihood of a crash may increase. A successful highway design would
make a driver’s mental workload level high enough to keep the attention needed in driving
performance, but would not exceed the driver’s processing capacities

33
Table 1: List of Accident Contributing Factors - Human

before the accident during the accident after the accident

human Physical condition Physical condition Physical condition

• Fatigue, illness, • reflex • resistance to

medication impact
Error
• handicaps: sight, Physiological
hearing, etc. • poor mental image condition
of the road
Physiological condition • emotional
• poor evaluation of shock
• stress, inattention, distances and
distraction, attitude speeds Experience and skill

Socio-demographic profile • inappropriate • safety first

manoeuvres
• age, sex, professional • protection of
occupation, level of Action accident-
education scene
• speed
Experience and skill • raising the
• braking alarm
• driving experience,
knowledge of vehicle • positioning Action
and itinerary, knowledge
• warning • manoeuvres
of regulations
after
Action collision

• manoeuvres before
collision self-protection

• seatbelt, helmet

Studying the human error by laboratory studies, field trials and data can be controversial as it is
unclear which is the ‘correct’ action in any given driving situation. According to the literature,
human error accounts for 50% to 70% of aviation crashes, 70% of marine accidents, but 90% of
car crashes. However, although human error can be defined as a contributory cause to an accident,
usually the error is one of a lengthy and complex chain breakdowns (some mechanical) that caused
the system to break down.
Reason proposed what is referred to as the “Swiss Cheese Model” of system failure. Every step in
a process has the potential for failure, to varying degrees. The ideal system is analogous to a stack
of slices of Swiss cheese. Consider the holes to be opportunities for a process to fail, and each of
the slices as “defensive layers” in the process. An error may allow a problem to pass through a
hole in one layer, but in the next layer the holes are in different places, and the problem should be
caught. Each layer is a defense against potential error impacting the outcome.

34
For a catastrophic error to occur, the holes need to align for each step in the process allowing
all defenses to be defeated and resulting in an error. If the layers are set up with all the holes lined
up, this is an inherently flawed system that will allow a problem at the beginning to progress all
the way through to adversely affect the outcome. Each slice of cheese is an opportunity to stop an
error. The more defenses you put up, the better. Also the fewer the holes and the smaller the holes,
the more likely you are to catch/stop errors that may occur.
Human factors involve a large number of specific factors that may be considered as accident
causes, including:
• Driver injudicious action (speeding, traffic violations etc.)
• Driver error or reaction (loss of control, failure to keep safe distances, sudden braking etc.)
• Behavior or inexperience (aggressive driving, nervousness, uncertainty etc.)
• Driver distraction (mobile phone use, conversation with passenger etc.)
• Driver impairment (cerebral diseases, fatigue etc.)

1.2 Why human factor is critical? Let's explore who is involved in the road accidents
Young novice drivers (8.5% of the population), are involved in 16% of all crashes and 13% of
fatal crashes. Motor vehicle crashes are the leading cause of death for this age group.
Older drivers make up 12 % of the population but are involved in 14% of traffic fatalities. Older
people drive less, are less likely to drink and drive or speed, and more likely to buckle up.
Males are about three times more likely than females to be involved in fatal crashes of all types,
e.g. motorcycle, passenger vehicle, pedalcyclist, etc.
Racial and ethnic groups are overrepresented in fatal crashes. They are less likely than whites to
use safety belts or child car seats and more likely to drive and walk impaired. Fatally injured Native
American and Hispanic drivers are less likely to hold valid licenses than White, Asian and Pacific
Islander or African American drivers.
Aggressive driving involves speeding, following to closely, aggressive weaving in and out of
traffic, and other factors.
Speeding is cited as a factor in about 1/3 of all fatal crashes. Although not a subject of study to
date, it is likely aggressive drivers distract and unnerve other drivers which may also contribute to
driver error and further incidents.
Driving while intoxicated, under the influence of drugs (illegal, over the counter, or prescription),
or fatigued all are known to contribute to crashes.
Drivers and passengers who avoid using safety restraints pose a higher risk, as crashes tend to be
more severe.
Driving with a revoked license, without a license, and without insurance tends to be associated
with high-risk driving. About 20% of drivers involved in fatal crashes are not properly licensed.

35
Drivers, who are distracted and do not give sufficient attention to the driving task, are high risk
drivers.
Human factors including basic demographic factors that may increase the likelihood of accident
involvement including for example, older versus younger drivers. However individual behavior
either whilst driving or before driving can also influence the likelihood of accident involvement.
1.3 Personality traits
There have been some studies that indicate that a person’s driving style is uniquely related to their
driving style. In other words, if you are an aggressive person in real-life, you are likely to drive
aggressively and if you are a sensation-seeker, you are likely to take risks. It is easy to see how
such personality typologies can manifest themselves in different accident risks.
Substantial insight can be gained from “Personality Tests” as certain personality traits have been
shown to be related to risky driving, crash involvement, style etc. Basic personality characteristics
which may lead to risky driving behavior are the following:
• Sensation seeking
• Impulsiveness
• Thrill and adventure seeking
• Emotional instability
• Anti-social
• Aggressive
• Being stressed

1.4 The Driver Distraction factor

Driver distraction is an important factor of increased risk of road accident worldwide.
Approximately 30% of drivers that were involved in a road accident reported some source of
distraction before the accident occurred. Distraction sources may be considered as a typical part
of everyday driving. The penetration of various new technologies inside the vehicle, and the
expected increase of use of such appliances in the next years, makes the further investigation of
their influence on the attention of drivers, on traffic flow and on road safety very essential.
Distraction accident risk – Key elements:
• Attentional demands: The amount of resources required to perform the distraction task.
• Exposure: How often and when drivers engage in the task. Driver strategies (if any) to
compensate for distraction.
• Risk compensation: can the additional mental or motor workload be counterbalanced by
adjusting driving behavior?
Earlier and recent studies agree that mobile phone use while driving may significantly affect
driver's behavior and safety. Research results suggest that mobile phone use may be the most
important in-vehicle distraction source for drivers. Drivers tend to reduce their speed during a
mobile phone conversation.

36
Although reduced speed is generally associated with lower accident risk, drivers using their
mobile phone while driving present up to 4 times higher accident risk, most probably as a result
of increased workload and delayed reaction time.
Although the physical distraction associated with handling the phone can present a significant
safety hazard, the cognitive distraction associated with being engaged in a conversation can also
have a considerable effect on driving.
Many studies have found that conversing on a hands-free phone while driving is no safer than
using a hand-held phone (Haigney et al., 2000; Matthews et al. 2003; Redelmeier & Tibshirani,
1997; Strayer, Drews, Albert & Johnston, 2003).
When drivers were engaged in a phone conversation using either a hand-held or hands-free phone,
they demonstrated similar driving deficits (Strayer et al. 2003).
Drivers tend to overestimate the ease of using hands-free phones while driving. (Mazzae et al.
2004.) The question of whether drivers actually modulate texting engagement is not well
addressed in the literature. Results indicated that drivers were particularly impaired when sending
text messages and less so when receiving (Hosking et al. 2009). When texting, participants express
greater following variability, greater lateral variability, reduced response time to the lead vehicle,
and increase in collision frequency. (Drews et al. 2009). A recent naturalistic driving experiment
suggests that the effects of texting may be significantly underestimated in previous (simulator)
experiments. (Cooper et al. 2011)
1.5 The Speed factor
Excessive or inappropriate speed contributes to about one third of all fatal accidents and is an
aggravating factor in all accidents. The level of exhaust emissions, fuel consumption and noise
increase with speed. Speed effects the quality of life of urban residents, especially the safe mobility
of vulnerable users
Higher speed means higher likelihood of crash. When driving with high speed, there is longer
distance travelled during driver perception / reaction time, longer distance travelled during
emergency braking, the driver is more likely to lose control, there is less time to take preventive
action, driver errors are magnified and other road users are more likely to misjudge speed.

37
Figure 2: Speed and braking distance

Normal driver response times have been found to be between 1.5 and 4 seconds (Evans L. Traffic
Safety and the Driver, USA, Van Nostrand Reinhold, 1991). Braking distance will vary according
to the type of road surface and its condition (longer on wet roads than dry roads). The above figure
illustrates the typical distance travelled by a car when the driver reacts to an emergency at different
speeds. The orange bar illustrates the distance travelled during a driver reaction time of 1 second.
The red bar illustrates the distance required for the vehicle to stop when the driver applies
emergency braking. The blue figure represents a person stepping into the road 37 metres ahead of
an approaching car. At 50km/h and 55km/h, the car is able to stop before hitting the pedestrian.
At 60km/h and above, the car hits the pedestrian – the higher the initial speed, the higher the impact
speed, the more serious the injury.
Speeding encompasses excessive speed (driving above the speed limit and inappropriate speed
(driving too fast for the conditions, but within the limits)

38
Factors influencing choice of speed are presented in the following scheme:

driver factors

education- vehicle
promotion factors

DRIVER
speed
zone/limit SPEED road factors
CHOICE

enforcement
traffic
and
conditions
sanctions
crash and
injury risk

39
1.6 The Age factor (Old-Young)
As a driver get older specific cognitive abilities are deteriorated: attention, perceptual speed, short
term memory, as well as sensory capabilities (especially vision). A the same time, musculoskeletal
strength and flexibility is deteriorated as there is progressive loss of muscle mass. The following
table presents some key characteristics of a driver and how there are changed over the years:

Table 2: Drivers skills change over the years

Reaction time Increase from 20s Emergency braking
onwards

Perception of angular Gradually declines Judging speed of traffic, gap acceptance

movement

Divided attention Reduced ability to Tendency to attend to one task at a time,

multi-task increased problems with distraction and
complex situations e.g. Busy roundabouts

Problem solving Ability improves to More time required to interpret complex

40-50 then declines information; traffic signals and signage

Learning & memory Good memory for Problems with new routes, new situations,
facts, reduced ability new systems
with new information
and procedures

The age factor is also critical for the younger road users. Young drivers usually overestimate their
competence, they are poor at identifying distant hazards. It takes longer to respond to hazardous
situations, thy recognise less risk in traffic situations. They are less likely to be able to stop within
limits of forward visibility
Finally they drive too fast for conditions (although this may be within the posted limit)

40
Age factor / Young – The Saudi Case
It is younger adult road users who are the dominant group when casualties are broken down by
age with 60% of casualties are adults aged 19-40.

Source: WHO – Documenting Road Safety Efforts on Saudi Arabia, 2019

1.7 The Fatigue factor

Fatigue is an important risk-factor for accidents. Estimates vary as to how many accidents are
caused by fatigue. Probably around 20% of fatal road accidents involve driver fatigue. According
to VicRoads Road Accident Facts in Victoria, Australia (1998 Edition), about 30% of severe single
vehicle crashes in rural areas involve the driver being fatigued. Driver fatigue is a significant factor
in ~20% of commercial road transport crashes. It is estimated that over 50% of long-haul drivers
have fallen asleep at the wheel. Increased crash-risk at night (peak levels at night can be 10X
daytime levels). Fatigue causes cognitive impairment, slower reaction times, reduced vigilance,
reduced information processing (remembering to cancel indicators etc.), and poor steering (after
2-3 hours continual some fatigued drivers drive faster others drive slower.

41
2. Human behavior modification - principles to affect human behavior
How can we modify and improve human behavior?
2.1 Public empowerment
People’s wide participation, awareness and support is a strong requirement for the success of any
road safety strategy
2.2 Creation of road safety culture
The level of road safety in a country reflects its culture and it is linked to the efforts on time and
resources put. Introduction of road safety values, attitudes, beliefs, expectations, decisions and
guidance of cultural-based behaviors in the society. Incorporation of safety in decision making by
governments - institutions. A proactive, long-term focus is needed to cultivate and sustain road
safety culture transformation, including traffic education and engagement of the mass media.
Traditional interventions must be combined with cultural-based interventions involving changes
in attitudes and perceptions of normative behaviors at the community and individual levels
2.3 Safety System Approach
A more holistic approach to road safety, adopted by many strategies, which builds on existing road
safety interventions, but reframes the way in which they are viewed and managed in the
community. Main responsibility lies to the people in charge for the planning and the operation of
the road traffic system, namely; road managers, vehicle manufacturers, politicians and the police.
Road deaths and serious injuries are not accepted to occur. The physical vulnerability of the human
body, human capacities and limitations as well as errors made by road users are recognized. It is
the responsibility of the individual person to abide by laws and regulations.
Safety System Approach key principles
• Human Behavior
no matter how well we are trained and educated about responsible road use, people make
mistakes and the road transport system needs to accommodate this
• Human Frailty
the finite capacity of the human body to withstand physical force before a serious injury or
fatality can be expected is a core system design consideration
• Forgiving Systems
roads that we travel on, vehicles we travel in, speeds we travel at and the attitudes of road
users to each other, need to be more forgiving of human error
In order a Safe System Approach to be implemented some practical steps are required to generate
greater community support for a safety culture:
• Identify the group of critical issues that need to be addressed
• Develop compelling information relating to the scale and significance of the issue
• Bring a wide range of private and public interest groups and citizens together to discuss,
verify, modify and develop contributions
• Develop options for interventions

42
In parallel, it is required to integrate road safety with other transport and wider societal goals.
Rather than focusing solely on road safety, the safety manager needs to look more broadly at
improving the quality of people’s interaction with the transport system and they need to meet the
need for people and goods to move (destination reached at scheduled time). Of course, economic
costs for firms can be reduced. Long term improvements in safety are essentially aligned with
better organized urban environments (land use vs transport). Finally, it is necessary to promote
synergies between road safety objectives and environmental protection (low speeds in urban areas)
A many interventions for this also we have to monitor the road user behavior performance
indicators:
• speeding, comparison to mean speed,speed variance, speed limit violations
• percentage of seat belts , “child restraints” and helmets’ use
• incidence of drinking and driving
• incidence of mobile phone use
• failure to stop - yield at junctions or at pedestrian crossings
• inadequate headways - close following
• use of reflective devices for cyclists and pedestrians
• use of pedestrian crossing facilities by pedestrians

Finally, in order to project improvement from the Safe System approach, it is usually feasible to
estimate the likely benefits arising from changes in key behaviors and in individual aspects of the
road environment. It is more difficult to make a firm quantitative estimate of reductions in serious
injuries and fatalities arising from a Safe System approach.
2.4 Safety Culture
Safety culture, as defined by Reason (1997), rests on five components, namely:
• Informed culture - The organisation generates significant data on incidents and accidents,
which are complemented by safety audits and surveys on safety environment
• Reporting culture - Employees are encouraged to report their errors or near misses and take
part in surveys on safety culture
• Just culture - The establishment of a trust relationship between employees and employer,
where reporting mistakes and incidents is encouraged and employees know they will be
treated fairly if they make unintentional mistakes
• Flexible culture - The organisation shows that it is able to adapt its practices when
warranted
• Learning culture - The organisation learns from incident reports, safety audits and
internally reported issues, resulting in improved safety.

Inspired by those five components, a Safety Management System (SMS) requires a just culture,
where staff can report mistakes, omissions or incorrect decisions without the fear of disciplinary
actions but where there will be no tolerance for gross negligence or unlawful acts. SMS efficiency
also depends on the ability to collect data and use it in predictive analysis to foresee issues and

43
manage them. However, the collection of this data may raise employee privacy concerns, which
must be weighted against the safety benefits. This rationale also concerns road safety.

2.5 Improving the safety of vulnerable road users

Firstly there is need for leadership from a government agency collecting and using good data on
crashes and injuries involving VRUs. Then strategy and action plan to improve vulnerable road
user safety shall include:
• Objectives and targets for improving the safety of VRUs
• Legislation and standards that prioritize the safety of vulnerable road users
• Enforcement of key laws affecting vulnerable road users (e.g. speed)
• Interventions focused on achieving concrete outcomes for vulnerable road user safety
• Monitoring of measurable performance indicators

3. Educational strategies
Education and awareness-raising campaigns are often the first and only tools tried when attempting
to influence behavior. In general, the goal of such campaigns is to communicate information with
the assumption that once the audience is aware of the information, they will then act in the desired
manner. In other words, educational campaigns appeal to the deliberative system and assume that
human behavior is usually a product of rational thought. Because of this, information alone almost
never works. However, information can be helpful as part of a more comprehensive program. Basic
road safety educational strategies are presented below:
• Implement public education/awareness campaigns which target the main causal factors for
collisions, deaths and serious injuries for all road users but in particular the high risk
groups
• Apply new media techniques/initiatives to road safety awareness/education campaigns
which are interactive and age-appropriate including the development of a social media
strategy
• Implement specific educational measures aimed at vulnerable road users
• Safe crossing by pedestrians
• Awareness of intoxicated pedestrians
• Use of personal protection equipment for cyclists and motorcyclists
• Awareness of blind spots on HGVs
• Road users and workers at road works
• Care for young and older people
• 30km/h zones in residential and sensitive built-up areas
• Promote the use of high visibility material for pedestrians, cyclists & motorcyclists
• Identify the most accident-prone places especially for children
• The approach could also be applied on other road user groups, e.g. cyclists, motorcyclists
or elderly
• Develop and implement education/awareness interventions aimed at the high risk 17 to 24
year age group

44
 • Monitor the implementation of guidelines to assist advertisers to depict positive images
of safe driving behavior and avoid showing unsafe road behavior
• Update the education/awareness program on the dangers of fatigue to include sleep
disorders such as Obstructive Sleep Apnoea Safety, Syndrome (OSAS)
• Ensure that education/awareness materials are readily accessible to the road-using public
via internet and in hard copy through public buildings
• Integrate international road safety awareness events such as UN Global Road Safety
Awareness Week, EU Road Safety Day and World Day of Remembrance for Road Traffic
• Implement a comprehensive integrated road safety education program for Pre-school,
Primary, Post-Primary, Third Level, Community and further education programs
• Develop a standardized guide to reducing road safety school gate risks and implement
nationally
• Increase educational offerings to offenders
• Increase educational courses in safer driving behavior - After a serious disqualification
offenders will take a course and assessment to regain a driving license
• Develop and implement a program for the compulsory initial and ‘periodic’ training of
driving instructors
• Reduce long term reliance on multiple learner permits by introducing measures to ensure
that learner permit holders sit a driving test before they can obtain a subsequent learner
permit
• Develop a network of accredited specialist driver trainer/assessors to support medical
fitness to drive clinical decision making
• Develop an Emergency Services Driving Standards & Syllabus for front line personnel
required to drive emergency vehicles
• Develop, introduce and oversee the implementation of a driving standard, training syllabus
and assessment process for emergency service drivers
• Educate drivers about tire safety, maintenance and checking through campaigns as well as
during driving license tests
• Educate drivers about vehicle maintenance and checking through campaigns as well as
during driving license tests
• Work with victim advocacy groups to inform the development and implementation of
support/advisory information/counselling for families affected by road fatalities and
serious injuries
• Work with and educate Fleet Management Companies and procurers of fleet vehicles to
ensure that vehicle safety (NCAP) ratings are given consideration when choosing and
purchasing vehicles
• Work with public and private organizations to ensure that road safety is a key determinant
in specification for services and tender documents when considering the procurement of
goods and services

4. Road Safety Campaigns

As described above, a basic pillar in order to improve road behavior according to the Safe System
Approach is the road safety campaigns through purposeful attempts to inform, persuade, and
motivate people in view of changing their beliefs and/or behavior.

45
The objective is to create awareness of road accident threats and vulnerability of (certain) road
users, to educate road users as to what constitutes safe or unsafe user behavior, to change attitudes
and beliefs to a more positive road safety approach and to inform road users of changes in traffic
regulations, operating conditions, etc. An effective methodology providing a sustained positive
change in road user behavior is much desired, and furthermore, seen as an investment with very
high road safety return.
Key elements for Road Safety Campaigns:
• Combined with other activities such as enforcement, education and/or legislation
• Theoretical model to be used
• The campaign to be based on prior research
• Single theme to be chosen rather than multiple themes
• A specific target audience is addressed
• The target audience is segmented (e.g. by attitudes, values etc.)

5. Road safety enforcement

Traffic law enforcement influences driving behavior through two processes: general deterrence
and specific deterrence. General deterrence can be described as the impact of the threat of legal
punishment on the public at large. Specific deterrence can be seen as the impact of actual legal
punishment on those who have been apprehended.
The primary objective of enforcement is to achieve compliance with legislation. Whether drivers
comply with traffic rules depends on the advantages and disadvantages of doing so and their
relative value as people don’t like being punished for not complying with laws. The amount of
enforcement activities determines the driver’s risk of apprehension of being detected and punished.
This objective risk of apprehension affects the subjective risk of apprehension, i.e. the risk people
believe there is. The subjective risk of apprehension can be increased by specific enforcement
strategies and publicity campaigns in the media increasing attention to enforcement activities. If
the probability of apprehension is considered to be high and apprehension is followed by a fine or
other sanction, most people will refrain from traffic law infringements. The most important
requirement for enforcement is that it deters drivers from offending; not only those drivers that
have been apprehended for a traffic violation but also those who have not. The key enforcemcent
objectives are to dramatically increase compliance with traffic laws, to ensure the certainty of due
process when a driver/rider is detected and to reduce the incidence, severity and costs of road
trauma.
Yannis et al. (2004) describe the hierarchy of road safety enforcement. The legal and
organizational framework enabling police enforcement provides the foundation for the actual
policing operations. Such a framework will result in well-planned, intensified police controls on
selected locations of the road network, resulting in an increase in the perceived risk of
apprehension. As a result violation rates will decrease. Changes in road user behavior will result
in less traffic crashes and less traffic victims, and in reduced monetary costs for society (social
benefits). The figure below describes this ideal scenario.

46
47
6. Roadway features and human behavior

The contributing factors inside the Human – Environment – Vehicle system and focusing on the
environment part are presented below:

Table 3: List of Accident Contributing Factors - Environment

before the accident during the accident after the accident

road Geometry Recovery area Accident warning

environment
• vertical • shoulders, Cleaning up the road
alignment, emergency lane
cross-section
horizontal • central refuge
alignment Roadside conditions
Surface characteristics Critical zone
• skid resistance, • transition zone
roughness
• work zones,
• debris, unusual
contamination surroundings
Surroundings defect

• urban, rural • maintenance

• advertising, • obstacle on
shops roadway

• traffic volumes

• main users

Equipment

• signs, markings,
etc.

The Safety Concept and the principles infrastructure adaption to human capacity rely on the self-
explaining roads: a traffic environment which elicits safe behavior simply by its design and
advocates the use of set categories of roads to ensure drivers are not confused by different types
of road with varying speed limit.

48
Another Safety Concept and the principles infrastructure adaption is the Psychological Sight
Distance: the depth of the driving space, of which the driver supposes to have it completely
registered. It consists of adequate road and road environment field of view as well as information
exposed forward, for which driver feels comfortable in avoiding abrupt maneuvers as well as speed
variations and to perceive the road environment (orientation sight distance).

6.1 External distraction - advertising signs, billboards etc.

Most studies agree that advertising signs do attract the attention of the majority of drivers, for a
non negligible proportion of their driving time. Research on labels and advertising signs is not
conclusive so far as regards their impact on accident risk. The location and position of labels and
advertising signs might have different impact to the distraction of driver attention. Labels and
advertising signs resembling (in size and form) to traffic signs or other traffic indicators create
confusion. Blinking and animation objects have proven to be difficult-to ignore, dynamically
changing advertising and information are ill-advise.

6.2 Road safety engineering strategies

Basic road safety engineering strategies are presented below:
• Carry out realignment schemes based on accident in-depth analysis on the national road
network
• Implement safety schemes at high risk locations on regional and local roads (e.g. safety
barrier installation, skid resistant surfaces, visibility improvements, curvature
improvements etc.)
• Develop design guidelines for forgiving roadsides
• Investigate the necessity of service areas on or immediately adjacent to the motorway
network
• Investigate the adequacy of designated and equipped enforcement checkpoint sites for
commercial vehicles
• Each Local Authority to publish/renew their prioritized plan on road building construction
and maintenance on an annual basis
• Increase the percentage of national road pavement that is measured in the annual skid
resistance (SCRIM) surveys
• Increase the percentage of the national road network in which the pavement condition, as
measured by annual survey indices, is characterized as fair or better
• Research emerging ‘in-road’ and ‘in-vehicle’ road safety technologies and make
recommendations Public and Private on their introduction
• Keeping drivers alert
• Implement a targeted program to reduce the likelihood of fatigue
• Retrofit the rural network and other facilities prone to cause fatigue with shoulder
rumble strips
• Reduce the number of commercial vehicle crashes resulting from loss of alertness
and driver fatigue
• Reduce the number of access points outside speed limit areas on national roads

49
 • Turn junctions into roundabouts
• Reduction of conflict points number
• Reduction of vehicle crossing speeds
• Sweden and Netherlands are examples of countries that already went far in completing
such transition from junctions to roundabouts, with more than 2000 roundabouts
respectively
• Improve safety and operation of highway intersections
• Improve the safety of intersections using automated methods to monitor and
enforce intersection traffic control
• Improve intersection safety by upgrading signalized intersection controls
that smooth traffic flow
• Utilize new technologies to improve intersection safety

• Design safer work zones

• Adopt improved procedures to ensure more effective practices, including traffic
control devices, for managing work zone operations
• Enhance and extend training for the planning, implementation, and maintenance
of work zones to maximize safety
• Enhance safe work zone driving through education and enforcement actions
• Incorporate road safety into sustainable urban mobility plans
• Governments should promote the use of motorcycle-friendly infrastructure guidelines
when they exist, and develop such literature where it is missing
• Implement measures to ensure timely and parallel implementation of the eCall service by
the three involved stakeholder groups
• Automotive industry
• Mobile network operators
• Public emergency services
• Review and determine the most appropriate method of making vehicle odometer reading
history available to the public to prevent odometer fraud (clocking) thereby protecting
consumers and providing more accurate data on vehicle servicing and replacement of
safety critical components

6.3 Infrastructure and Pedestrian Safety

There are a number of specific engineering measures that reduce pedestrian exposure to vehicular
traffic. Most of these measures involve separating pedestrians from vehicles or reducing traffic
volume.
• Sidewalks/ footpaths
• Marked crossings
• Overpasses and underpasses
• Mass transport routes

One of the most effective ways to improve pedestrian safety is to reduce the speed of vehicles, as
speed management is much more than setting and enforcing appropriate speed limits. It employs

50
a range of measures in engineering, enforcement and education with the aim of balancing safety
and efficient vehicle speeds on the road network.
Intervisibility between pedestrians and motorists to ensure pedestrians can see and be seen. Good
quality lighting is required to ensure that pedestrians are visible at night. Providing crossing
enhancements such as raised crossing islands and traffic signals. Increasing intensity of roadway
lighting increases visibility of pedestrians at night, especially at pedestrian crossings.
Another measure is to remove or reposition physical objects that affect visibility, such as trees and
billboards that make it difficult for drivers to see pedestrians, to install signals to alert motorists
that pedestrians might be crossing and finally to improve conspicuity of pedestrians.

7. The driving task model approach

The early skills and trait based models are placed back in 1940s. Driver skill is considered as a
function of perceptual, cognitive, and motor skills. These higher skills are the product of lower
level competencies (e.g. reaction speed and spatial orientation). With optimum training and
practice, a driver will reach a maximum level of skill determined by these characteristics

Motivational models (1970s)

There are three major motivation-based theories of driver behavior: Wilde’s (1982) risk
compensation model (later termed target risk; Wilde, 1994), Fuller’s (1984) risk avoidance model
(later termed the task-difficulty homeostasis model (Fuller, 2008; in this book, termed
riskallostasis theory), and Na ¨a ¨ta ¨nen and Summala’s (1974, 1976) risk threshold model (also
called the zero-risk model). The driver behaviour is considered as the sum of a set of psychological
states of the driver. Driving is a self-paced task with a strong orientation toward safety (accident
avoidance). The driver accepts a level of risk; the control variable for the quality of driving
performance. The compensatory models suggest that the driver selects a speed that matches their
perceived level of risk. Purely descriptive, no explanation of perceived risk.

Risk Threshold Models

However, one idea is quite central in motivational and risk theories; that is, the consideration of
the nature of risk, objective risks and subjective or observed risks and their interplay in accident
causation. One of the reasons for the high number of accidents according to risk theorists (Na ¨a
¨ta ¨nen & Summala, 1974) is that the subjectively estimated risk in a situation is lower than the
real risk. If subjectively estimated risk were higher than the real risk, then it would prevent such
behavior that would lead to risky situations. Risk Threshold Models involve a control process that
would enable a driver to maintain a stable balance between subjective, perceived safety, and
objective, physically or statistically determined safety.

51
Subjective Risk Control Model
Nataanen & Summala introduced this model suggesting that perceived risk in traffic (R) depends
on the product of the level of subjective probability of a hazardous event and the subjective
importance of the consequences of the event. For some drivers the threshold is too high for
cognitive, motivational, or physiological reasons. Improvements in traffic safety should be
expected only from better vehicles and better roads

Hierarchical models
The descriptive models can be divided into hierarchical models (e.g., Michon, 1985) and control
loop models (e.g., McRuer & Weir, 1969). The hierarchical models describe behavior in terms of
a hierarchy of three distinct types of behaviors, each building on the level below it. The lowest
level is an operational, control level. At this level, most behaviors are automatic and consist of
quick responses to the changing environment (e.g., braking when a lead car slows down). The
second level is a tactical, vehicle maneuvering level referring to how traffic situations are mastered.
The behaviors are less reflexive and consist of conscious decisions in the driving, such as a decision
to change lanes before exiting a highway. The third and highest level is a planning or strategic
level, and it consists of long-term decisions such as which route to choose or even whether to drive
at all. Thus, the three levels can be distinguished by the task requirements, the time frame needed
to carry them out, and the long-term decisions

The Task-Capability Interface model

Motivational models are influential but don’t emphasise the learned skills that are central to
driving. The Task-Capability Interface model puts risk aside and concentrates on the skills of the
driver and the demands of the situation. It can accommodate novel events (e.g. use of mobile
phone), without requiring driver to be continuously calculating risk.

The TCI model is an attempt to understand what motivates driver decision making, with a
particular emphasis on implications for performance safety. It starts from a recognition that driver
perceptual processes and control actions both have rate limitations. Thus, the driver needs to
continuously create and maintain conditions for driving within these limitations. That is, he or she
must ensure that the demands of the driving task are within his or her capability. Loss of control
occurs when, for a multitude of possible reasons, drivers allow task demand to exceed their
capability.

52
8. Conclusion

Human factors are the basic causes in 65-95% of road accidents. They involve a large number of
specific factors that may be considered as accident causes, including:
• Driver injudicious action (speeding, traffic violations etc.)
• Driver error or reaction (loss of control, failure to keep safe distances, sudden braking
etc.)
• Behavior or inexperience (aggressive driving, nervousness, uncertainty etc.)
• Driver distraction (mobile phone use, conversation with passenger etc.)
• Driver impairment (cerebral diseases, alcohol, fatigue etc.)

It is crucial to develop and implement strategies in order to modify and improve all road users'
behavior in order to reduce road accident fatalities and serious injuries. Educational, enforcement,
and engineering activities are necessary in order to meet this target by having our roads full of
safe, well trained, and educated drivers who drive safe vehicles, on a safe (and ready to forgive
possible drivers mistakes) road infrastructure.

53
References

AASHTO, Highway Safety Manual, 2010 (part A): http://www.highwaysafetymanual.org

FHWA, Road Safety Fundamentals, 2010 (chapter1-4): https://rspcb.safety.fhwa.dot.gov/RSF
Hauer, Ezra, The Art of Regression Modeling in Road Safety. Vol. 38. Springer, 2015.
Hedlund, J. (2007). Improving Traffic Safety Culture in the United States-The Journey Forward:
Summary and Synthesis.
Marshall, W. E. (2018). Understanding international road safety disparities: Why is Australia so
much safer than the United States? Accident Analysis & Prevention, 111, 251–265.
Papacostas, C. S., & Prevedouros, P. D. (1993). Transportation engineering and planning.
Tefft, B. C. (2008). Risks older drivers pose to themselves and to other road users. Journal of
Safety Research, 39(6), 577–582.
WHO. (2015). Global Status Report on Road Safety 2015. World Health Organization.
Al-Shammari, N. K., Darwish, S. M. H., & Abidi, M. H. (2019). Exploring the Risk Factors for
Road Traffic Collisions in Saudi Arabia, 1970-2018. Environmental Research and Public
Health.
Carter, D., Gelinne, D., Kirley, B., Sundstrom, C., Srinivasan, R., & Palcher-Silliman, J. (2017).
Road Safety Fundamentals: Concepts, Strategies, and Practices that Reduce Fatalities and
Injuries on the Road. United States. Federal Highway Administration. Office of Safety.
Evans, L. (1991). Traffic safety and the driver. Science Serving Society.
Part, D. (2010). Highway safety manual. American Association of State Highway and
Transportation Officials: Washington, DC, USA.
2014 ‫ عام‬،‫ الخطة الوطنية للسالمة المرورية‬،‫مدينة الملك عبدالعزيز للعلوم والتقنية‬
Global status report on road safety 2018: https://www.who.int/publications-detail/global-status-
report-on-road-safety-2018
Elvik, R. and T Vaa, The Handbook of Road Safety Measures, Elsevier, 2004. (Chapter 4)
Roess et. al., Traffic Engineering, Prentice Hall, 4th ed., 2010. (Chapter 11)
Garber, N. and L. Hoel, Traffic and Highway Engineering, Cengage Learning, 4th ed. 2008.
(Chapter 5)
World Road Association (PIARC), Road Safety Manual, 3rd ed., 2019:
https://roadsafety.piarc.org/en (Chapters 1 & 2)

54
Training programs for Road Safety Professional
Exam (Level 2):

RSP1-SA

Phase Two
10-11/03/2021

1
 Module 3
Data and its role in measuring road safety

Overview

Unit Modules Topics in this Module

1-Basics of Road Safety

Identify types, applications, and users of safety

2-Effects of Human Behavior on Road Safety data

The importance of data quality and its impact on

decision-making
3-Data and Its Role in Measuring Road Safety
The contributing factors to traffic crashes

4-Methodologies of Solving Road Safety Problems The major components of quantitative safety
analysis

5-Traffic Safety Strategic Plans and Programs

2
Outline
I. Data and Its Role in Measuring Road Safety ................................................................ 5
A. Introduction .................................................................................................................. 5
II. Traditional safety data ....................................................................................................... 7
A. Distinction between qualitative and quantitative safety data ............................. 7
B. Safety-Critical Data: crash data, road characteristics, and traffic data ............ 8
1. Road Network Characteristics .............................................................................. 8
2. Traffic Data .............................................................................................................. 9
3. Crash Data ................................................................................................................ 9
4. Crash reports .......................................................................................................... 10
III. Identifying key factors in safety data ........................................................................ 12
A. Elements of the crash report template and how it is used in road safety
analysis .................................................................................................................................... 12
1. Injury Severity ....................................................................................................... 13
2. KABCO Scale ........................................................................................................ 14
B. How key factors affect type, frequency, and severity of crashes .................... 14
1. Human Behaviour ................................................................................................. 15
2. Road Environment Factors .................................................................................. 16
3. Vulnerable Road Users ........................................................................................ 16
4. Traffic Data ............................................................................................................ 17
C. Discuss interrelationship and overlap of factors and typical data distribution
and ranges ............................................................................................................................... 17
IV. Safety Data and Key Factors in KSA Context ........................................................ 19
A. Crash data ................................................................................................................... 19
B. Road data .................................................................................................................... 21
C. Traffic data ................................................................................................................. 22
V. Non-traditional safety data and the importance of data quality .............................. 23
A. Required safety data and responsible partners to measure road safety .......... 23
1. Highway authorities .............................................................................................. 23
2. Emergency services .............................................................................................. 23
3. Government ............................................................................................................ 24
4. Commercial data ................................................................................................... 24
5. Academic research ................................................................................................ 24
3
 B. Supplementary safety data....................................................................................... 25
1. Behavioural data .................................................................................................... 25
2. Demographic data ................................................................................................. 26
C. How the quality of safety data impacts decision-making for more effective
interventions ........................................................................................................................... 26
D. Scientific methods of data preparation to improve its quality ......................... 28
E. challenges and limitations that road safety professionals face when collecting
and preparing related data.................................................................................................... 29
VI. Quantitative Safety Analysis ....................................................................................... 30
A. What can be done with traditional measures: Crash data, Volume Data, and
Road Data................................................................................................................................ 30
1. Importance of Objectives .................................................................................... 30
2. Objectives Help Sort and Organize Data ......................................................... 31
B. Skills of using related/available safety data to reach solid results and
effective treatments ............................................................................................................... 32
1. Pitfalls to Avoid..................................................................................................... 32
2. Road Safety is a Stochastic Condition .............................................................. 32
C. Definitions and primary components of quantitative safety analysis ............. 33
1. Metrics ..................................................................................................................... 33
2. Context .................................................................................................................... 34
D. Main concepts when analyzing safety data .......................................................... 34
1. Rates......................................................................................................................... 34
2. Finding the denominator: traffic ........................................................................ 34
3. Calculating the rate ............................................................................................... 35
4. Matching numerator and denominator.............................................................. 35
5. Challenges and mitigations ................................................................................. 36
6. Machine Learning ................................................................................................. 37
E. Safety Performance Functions and Crash Modification Factors ..................... 37
1. Safety Performance Functions (SPF) ................................................................ 38
2. Crash Modification Factors (CMF) ................................................................... 40
VII. Case Study: SPF for Urban four-lane divided roads in Riyadh, Saudi Arabia . 42

4
I. Data and Its Role in Measuring Road Safety
A. Introduction
Good quality information is at the heart of successful progress. Without information
to describe a problem, it is not possible to identify targeted solutions for the problem.
Information also enables us to evaluate the degree to which attempted solutions
succeed, and how they can be improved. In the absence of good information (i.e.,
good data), solving problems is reduced to guesswork.

In road safety, the primary problem which requires our attention is human casualty
reduction. Roads are complex open systems and using roads can result in substantial
harm to people who use them. This can take the form of death, injury, and material
loss. All round the world, reducing death and life- changing injury on the roads is a
pressing concern.

In road safety, information describing problems largely consists of data. Data are
numeric values assigned to variables which describe features of the real world.
Different sets of variables can be constructed describe almost anything: for instance,
people, objects, and events can all be described in data.

Data represents the real world but is not identical to it. Data sets provide a useful
approximation of reality which allows us to exploit information in an efficient,
reproducible, and productive manner. However, all data are only as accurate as the
collection processes which generate them, and only as accessible as the systems
which store them. Their ultimate value is determined primarily by the skill and
knowledge of the people who analyse them.

Because of its ability to expose real world events to analysis, road safety data has the
potential to make a real difference in the quest to reduce casualties. Collecting and
applying data appertaining to road safety has therefore become an international
priority. As a result of work by organizations like the International Transport Forum,
we know that about 1.3 million people are killed in road crashes globally each year.

However, this risk is not distributed evenly. For instance, the risk per person of
population is ten times greater in South Africathan in Norway. Casualties are also
distributed unequally between groups of people who use the roads in different ways.
We know this because rates of injury can be calculated for different modes of travel,
which make it apparent that road users who are not enclosed by a vehicle body
(mainly pedestrians, cyclists, and riders of powered two wheeled vehicles) experience
much higher rates of injury relative to distance travelled. For example, globally, these
types of vulnerable road users on urban roads are exposed to a death rate ten times
higher than other vehicle users.

As road and vehicles change in the future, both data itself and analytical techniques
applied to it will have to evolve to keep up. For example, automated vehicles are
currently coming into use in some countries, a trend which is set to continue. Such
vehicles will collect much valuable data about their interaction with the road
environment and its users, which is likely to be a valuable future source of
information. As data sources grow in scale and level of detail, new approaches will be
5
required to obtain the most useful intelligence from them. For example, machine
learning techniques can be used to facilitate the incorporation of huge amounts of
data, making it possible to arrive at analytical conclusions which take many more
factors properly into account.

B. Nominal versus Substantive Safety

As mentioned in Module 1, road safety has two dimensions; 1) Nominal Safety 2)

Substantive Safety. Nominal safety can be defined as an absolute statement about the
safety of a particular network, corridor, or intersection based only on compliance with
standards, warrants, and guidelines for design. Substantive safety is the historical and
long-term objective safety of a highway or an intersection based on crash data (i.e.,
actual crash frequency, crash severity, crash type, crash rate).

In various cases, nominal safety is an essential but not sufficient characteristic of a

safe road, meaning even if the road meets design criteria, based on standards and
guidance documents, it may still have higher than expected number of severe crashes.
Likewise, a road may be considered substantively safe; having experienced low
number of severe crashes, yet may not be nominally safe.

Substantive safety requires an evidence-based approach to estimate the expected

safety of a location through data and analysis rather than focusing solely on standards.
Therefore, transportation professionals should strive for both aspects of the road
safety dimension (i.e., nominal and substantive safety) in order to have a nominal safe
road as it was designed and then continue to monitor the road for crashes and other
measures of safety to make sure that the road is also substantively safe.

6
II. Traditional safety data
A. Distinction between qualitative and quantitative safety data
Data is often described as coming in two forms: quantitative and qualitative.
Quantitative information is counted and measured numerically and can be subjected
to statistical analysis. It aims to describe the real world in terms which are as objective
as possible. Qualitative data is less structured and is not necessarily collected or
measured numerically. It describes the real world by categorising activities and
responses subjectively into groups.

Most road safety data, including the most frequently used and traditional measures, is
often quantitative in nature. However, because road safety outcomes are strongly
influenced by human behaviours and attitudes, there is also an important contribution
which can be made by qualitative data.

All data is only as good as its sources. The way data is collected has a major effect on
both the quality and quantity of available information. As a result of this variability, it
is unavoidable that not all data which is desirable for a given task will be available,
and some of the data which is available will be unsuitable for use. It is good practice
for a road safety data analyst to seek out data from a broad range of sources and work
thoughtfully with what is available to gain the most benefit from it. Sometimes, it is
prudent to be pragmatic and use data selectively. Always, it is essential to report any
data unavailability or shortcomings so consumers of the resultant analysis can
understand the consequent potential weaknesses in results.

7
B. Safety-Critical Data: crash data, road characteristics, and traffic
data
Among data categories critical to use in road safety analyses, three are almost always
is part of the work: crash data, road characteristics, and traffic data.

The traditional approach to road safety data is quantitative and empirical. Data is
gathered by measuring observable facts about the road network and incidents which
occur on it, ensuring that data gathered is fit for purpose, then using it to quantify road
safety outcomes.

1. Road Network Characteristics

The first step in acquiring data for insight into road safety is to understand the nature
of the road network itself. Clearly it is important to understand and measure the
characteristics of the network, but this is not always a straightforward process. Roads
are in constant use and as a result are in a state of constant change. Records of what
changes were made to which road at what time is not always easily available, if at all.
Relating available road infrastructure data to other data sources is not always
straightforward, as accurate spatial and temporal information about the network may
be lacking and as a result matching with other information can be challenging. A
simple example of network information is the total length of different types of road
that exist within areas.

Different network characteristics are essential for different purposes. When

investigating issues related to how vehicle perform on the road, surface characteristics
are often the most essential. For instance, road alignment is important to understand
vehicle performance on bends, while the paved surrface characteristics affect vehicle
grip in adverse weather conditions. Driver behaviour issues like speeding and
8
overtaking are often more affected by layout charateristics such as the presence of a
central median (or reservation) and the number and type of lanes present. When
investigating vehicles leaving the carriageway, it is important to understand the
immediate environment of the road, such as the presence of crash barriers, signage
and traffic control.

2. Traffic Data

The next critical safety metric is knowing how much traffic is using the network, and
what types of vehicles are involved and their travel characteristics (speed, for
example). This information can be collected with a variety of equipment which is
sometimes built into the road permanently, and sometimes deployed temporarily.
Manual counts are also sometimes used. This information is of vital importance in
understanding how many vehicles use a road, therefore placing its road safety record
into the context of how many vehicles were therefore exposed to potential danger.
However, because count points are by their nature selective, providing definitive
information for only a limited scope in space and time, overall traffic flow usually
must be estimated based on incomplete information. Often, flow information is
available sparsely or not at all for minor roads, and counts gathered by different
methods can be hard to compare meaningfully with each other. Another complication
is that not all surveys indicate the type of vehicles being used, which is makes it more
difficult to understand actual utilisation of a route.

3. Crash Data

The most critical category of traditional traffic data is reports of crashes. “Crashes
“are incidents on public roads involving vehicles which resulted in death, injury,
damage, and/or risk. Crashes, which are also sometimes referred to as ‘accidents’ or
‘collisions’, frequently come to the attention of emergency services and/or authorities
responsible for maintaining roads. Most such organisations record crashes, although
often in different ways to suite their own purposes and priorities. Road safety
professionals are sometimes required to combine information about the outcomes of
crashes from different sources, some of which may be incompatible or inconsistent on
points of detail, to obtain a complete picture of a crash event

Crashes are often categorized by their relative severity, or by the degree of danger to
human wellbeing. Of greatest concern are crashes which have resulted in loss of life;
these events are usually investigated very thoroughly and recorded with
comprehensive crash data. However, a great many more crashes do not result in the
death of any person involved but do lead to injuries which are nevertheless very
serious and potentially life-changing events for the victims. Sometimes, the difference
between whether a crash may result in death or ‘only’ in injury is a matter of chance -
mere millimetres or moments - rather than any fundamental difference in the nature of
the incident itself. It is therefore important to treat information about injury crashes as
a very valuable basic data source concerning actual road safety outcomes.

Sometimes data is also available for crashes where no personal injury occurred but
some damage to vehicles or property resulted (“Property Damage Only” crashes). It is
even possible that some agencies collect information about events which could have

9
resulted in a crash occurring, but was narrowly avoided. These ‘damage only’ and
‘near miss’ incidents can provide useful background information about degree of road
danger in different situations, but because their data is often not collected
systematically it can be challenging to obtain, interpret and compare.

In most jurisdictions, the fundamental recording of crashes is the responsibility of the

police, who are called up on to respond to the largest proportion on these incidents.
Police forces use a variety of methods to record information about crashes, including
paper reporting, hand-held devices, and computer systems. Often the source for these
reports is an officer who attended the actual crash scene, therefore providing valuable
independent insight into the circumstances and participants. However, sometime
reports are compiled after the event or are based on information supplied to the police
by third parties who are not necessarily independent or reliable.

4. Crash reports

Crash reports typically come in multiple parts, as crashes have multiple participants.
As well as information concerning the overall event, such as date, time, location and
weather conditions, and data is also typically gathered separately about each vehicle
involved in the crash and each casualty which resulted from it. These different data
sets are typically held in separate tables which must later be related to each other for
analytical purposes. For example, a crash which involved three vehicles should have
three separate vehicle records each of which should be related to the overall crash
record. Similarly, if several occupants of a vehicle were injured, that single vehicle
should have several casualty records associated with it. Understanding how any why
these records are related to each other is critical to proper handling and use of crash
data. The figure below illustrates how the data from one crash event may be expressed
in different categories or types of crash records: the crash itself, vehicle reports, and
casualty reports:

10
Importance of Location Data - A most essential component of all critical road safety
data, whether related to the road network, the traffic it carries or the crashes which
occur on it, is the key metric of location. If the exact location of a junction cannot be
matched to traffic count points, or if the place where a crash occurred cannot be
accurately located relative to these things, then the relationship between them cannot
be adequately understood.

Because collecting and collating these various road safety data sources is challenging,
location information is not always recorded with compatible or consistent formats or
levels of accuracy. It is not uncommon for road safety analysts to find that a critically
important aspect of their work is to ensure that location information in accurate within
data sources and compatible between them. Time spent ensuring that location
information is accurate and fit for purpose is seldom time wasted. The best approach
to matching locations is to identify an element within each data source which can be
spatially located with precision, for example route numbers which are present in
national mapping layers or street names which can be identified in a gazetteer, then
combine these mappings using a GIS software tool.

11
III. Identifying key factors in safety data
A. Elements of the crash report template and how it is used in road
safety analysis
In most countries, police records of crashes are critical road safety data. Exact
procedures vary from place to place, but the most important data elements recorded in
the template are widely recorded and are among the first items you will require for
many analyses. This section identifies and discusses those elements.

The most crucial information concerning the attendant circumstances of the crash
include the where and when it occurred. The date and time should have been recorded
as accurately as possible at the time, but it is prudent to sense check this essential data
if possible, for example by cross referencing times to reported lighting conditions or
dates to reported weekdays. This will allow the information to be augmented at a later
stage: for instance, later data processing could categorise collision times into
convenient intervals, and days of the week and other items such as public holidays
and weather conditions can be calculated.

Understanding the exact location of a crash is also of critical importance. Because

integrating crash location with other data sets, such as road data and traffic count
locations as well as other supplementary data, is often an essential analytical step, it is
important to obtain spatial co-ordinates such as latitude and longitude if possible.
Again, if this data is accurately recorded at the time, it can be augmented later. For
example, exact route and junction co-locations can be identified, and the urban or
rural nature of the immediate environment can be derived analytically. If accurate
spatial co-ordinates are not obtained, the value of the data is reduced. Other
information germane to the scene of the crash can also be included, such as the
presence of roadworks.

Accurate and complete information about a crash cannot readily be recorded in a

single record. Individual records for each vehicle can add significant additional detail,
relating not only to that vehicle but also to the driver. In addition to basic details such
as vehicle type, information can include the manoeuvres such as overtaking and lane
locations such as vehicles on a hard shoulder. Driver details can include information
such as age and place of residence, which are both valuable data for understanding the
human factor in crashes.

Vehicle records also make it possible to understand the conflicts between vehicles
which occurred during the crash. For instance, if directions of travel and points of
impact are recorded systematically as data fields, then differences between front and
rear impact can be discerned analytically. If this information is not recorded in data
fields then it may be available from text descriptions of every incident. However,
reading such descriptions individually can be time consuming so an analytical
approach to identifying conflict types is often preferable. It should also be
remembered that vehicles can also be in conflict without colliding, for instance in a
close following situation.

12
Similarly, where crashes result in multiple casualties there is considerable value in
having a record held separately for each casualty. The most essential thing you need
to know about a casualty is in what way they became involved in the crash: whether
they were a driver, rider, passenger, or pedestrian. This allows casualties to be
analysed separately by mode of travel, which is particularly important for identifying
danger to vulnerable road users.

Additionally, demographic data on casualties is of considerable value for the same

reasons as for drivers. Other information such as pedestrian movements can also be
revealing, as they can indicate how a pedestrian came into conflict with traffic. For
instance, this may reveal the need for additional pedestrian infrastructure at a location.

However perhaps the most pressing datum of all, both for individual casualties and for
a crash overall, is to understand the severity of injures which resulted. The
international standard definition of fatality is where a death, not attributed to natural
causes, occurs within thirty days of the crash. Thankfully, such losses are relatively
rare compared to the occurrence of injuries short of death.

1. Injury Severity

It is very valuable to have the capability to assess the relative severity of non-fatal
crashes by reference to the most severe injury suffered by any victim. This table1
shows an international standard scale for assessing degree of injury in an objective
way. The severity of the crash is assessed to be the worst severity experienced by any
casualty.

Police officers do not generally have the skills required to make medical assessments
of injury, but it may be possible to obtain this information by other means. One
method is injury based assessment, where an officer is presented with a list of
common injuries and asked to select all those which applied. The appropriate severity
can then be looked up by the system without asking the officer to make and
judgement. An alternative possibility is to source medical records of crash victims,
although matching patient data to road crashes is complex and sometimes impossible

1
https://www.aaam.org/abbreviated-injury-scale-ais
13
because medical records are not kept in a manner compatible with police information
(for instance, not recording location of crash, or recording date and time of admission
to hospital rather than time of crash).

2. KABCO Scale

If injury based reporting is not possible, then the next best option is to ask officer to
make subjective judgements of relative severity. An example of this approach is the
so-called KABCO scale2 used in the US, where officers record a casualty on one of
five subjective categories: Killed (K); Incapacitating injury (A); Non-incapacitating
injury (B); Possible injury (C); or No injury (O).

The data fields mentioned above are those which are typically recorded in crash report
templates internationally, and which you are most likely to use frequently where they
are available. How these and other data are recorded by police attending crashes in
Saudi Arabia will be discussed later in this document.

B. How key factors affect type, frequency, and severity of crashes

While the crash report template can answer many important questions about location,
participants, severity and more, there is another question which sometimes can only
be answered tangentially if at all. That is the question of why the event occurred.

It is possible to analyse the crash record of a junction, route, or area without having
answers to the question of which factors contributed to it occurring, but road safety
value will be significantly limited. If you seek to enhance the value of your analysis
so that results can more readily be applied to real world outcomes, you will go beyond
the realms of quantitative information and will have to depend on judgement calls.
This could be the judgement of participants, officers attending the scene, or
subsequent investigators. While relying on these could potentially limit your sample,
because some reports may not include them, or even distort your results, because
apparently similar subjective judgements made by different officers may not in fact
refer to comparable events, excluding them may reduce the value of your conclusions.
It is necessary to strike a balance between these limiting factors; for example, you
may use subjective judgements in your analysis but point this out clearly when
presenting your results so consumers can make informed judgements about them.

2
For further information see https://safety.fhwa.dot.gov/hsip/resources/fhwasa09029/sec4.cfm
14
The answer to the question “Why” for any crash event is most often a combination of
human factors. Many studies3, as illustrated in the figure below, have shown that most
crashes are caused by one or more human beings making one or more errors. Road
environment factors most often play a secondary role, while vehicle factors are only
rarely involved. While it is important to render roads and vehicles as safe as possible,
this should be understood as an approach principally designed to mitigate the
consequences of human behaviour rather than intervening at the underlying source of
the issue. The most difficult question from a data perspective is how such human
factors, even if they are known or suspected at the time, can ever find their way into
data sets in a way which will make them accessible to the analyst.

1. Human Behaviour

If your analytical objectives suggest you should approach behavioural factors as they
contributed to crashes, there are a several considerations it is helpful to have some
way of assessing. One obvious avenue of investigation is illegal behaviour by crash

3
Example: Transport Research Laboratory report 297: Sabey and Taylor 1980
15
participants, such as speed above the legal limit, transgressing traffic sign or controls
or travelling in the wrong direction. An advantage here is that attending police
officers are likely to be aware of and alert to relevant regulations. However, they may
often not have sufficient evidence to proceed with formal action, and as a result
information of potential road safety value could be discarded as not worthy of report.

Human error is an aspect of behaviour which does not stem from deliberate choice;
errors are actions which stem from subconscious attitudes, lack of skill or knowledge,
and/or impairment of some kind. Error also a common factor which can contribute to
crashes, including behaviours such as poor observation or failure to control the
vehicle adequately. These are often related to factors personal to crash participants,
such as inexperience, difficulty in dealing with weather conditions, and various forms
of personal impairment such as ill health or intoxication. Another increasingly
common phenomenon is distraction caused by road users attempting to operate
phones or other devices at the same time as travelling.

While these behaviours are often harder for police officers to identify with certainty
than illegal acts, they may often be suspected by attending officers. If there is some
way an officer can record human factors, they think may have been relevant, this
avoids the data being lost altogether. However, it comes at the price of your analysis
being dependent on the officer’s qualitative judgement. This dilemma, between
losing information altogether or having to work with information of difficult
provenance, is a challenge for the analyst. Deciding how to use such data is a good
example of a decision best taken with reference to the identified objectives of the task
in hand.

2. Road Environment Factors

Other factors which relate back to critical road safety data can sometime be identified.
Road data, such as surface, profile and alignment are often relevant to how weather
conditions or vehicle manoeuvres influenced the outcome of crashes. The layout of
the road, such as the presence of a median or the numberand width of lanes can
fundamentally change how the dynamics of a crash are understood. Naturally,
knowledge of junction controls present at a location is important in the many crashes
which occur where streams of traffic meet while travelling in different directions.
Access to good quality road data can shed important light on these considerations.

3. Vulnerable Road Users

A factor with special relevance in urban settings is the possible presence of vulnerable
road users. Facilities designed for pedestrian or cyclist use affect road user behaviour
even if no such participants were injured. Road verges and footways are often
cluttered with other obstacles. Some of these are deliberate parts of the built
environment, such as crash barriers and traffic lights; others are not, such as parked
vehicles and privately erected signs. Both categories can present significant hazards or
mitigations in particular circumstances, which are of relevance if a vehicle leaves the
carriageway or driver vision is obscured. Obtaining accurate information about such
obstacles, while desirable, is not always possible.

16
4. Traffic Data

Finally, levels of traffic, particularly at times of day when volumes are very high or
very low, are important to understand. Traffic data, if available for a particular
location, can provide valuable insight into conditions on the road, both at the time a
crash occurred and also, for comparison, at other times. This is particularly true at
busy interchanges, where traffic entry from different arms of the interchange may
vary very considerably. Survey details can also include information about driver
behaviours relating to peed and headway, as well as discriminating between different
types of vehicle. For instance, locations where motorcycles commonly travel at high
speed present a very different type of hazard to very bust freeways where a large
volume of goods traffic typically travels very close together.

While all the factors discussed in this have an impact on crashes, their precise impact
will vary between situations and whether each factor needs to be incorporated in any
specific analysis depends on the objectives of that analysis. In general, the severity of
collisions is most heavily affected by the presence of vulnerable road users, who are
not protected from impact by vehicle bodies, and by illegal behaviours such as
speeding. Frequency of crashes is most heavily impacted by traffic volume and by the
prevalence of human error among road users. The types of crash which occur are
often heavily influenced by road environment factors such as the presence of
junctions or medians. However, these general categories are not absolute and where
time and available data permits a more inclusive approach to possible contributory
factors is to be preferred when designing your methodology. Finally, it should be
observed that some factors, such as weather conditions, have some degree of impact
across crashes of all types.

C. Discuss interrelationship and overlap of factors and typical data

distribution and ranges
The figure below illustrates the competing aspects of sample size and data quality that
often manifest in road safety analysis work. For example, if we view all available
crash data for a given location, we have a lot of data, but not so useful in analysis
(“very sparse [useful] data”). As we refine, sift and focus the available data, it
becomes more valuable (“value added data, peer reviewed”).

Sample size and selection is a common dilemma in road safety analysis. It is

sometimes hard to judge whether it is preferable to work with a large sample which
may lack detail or with a much smaller sample where many more metrics are
available with greater certainty but lacks breadth across the network and the full range
of events.

Larger sample sizes, such as those relating to damage-only crashes, offer more
comprehensive coverage and greater statistical certainty, but the data may be lacking
in focus even on important points such as time and location. Small samples of well
investigated crashes, on the other hand, offer much greater certainty of what happened
where, when, and even why. However, there is a danger than such a small sample
may have been selected for operational or other reasons for investigation, and

17
therefore could be unrepresentative or even materially distorted compared to more
normal incidents and conditions.

There is no perfect solution to this problem, Often, a compromise is the best

approach, for instance choosing to examine all crashes which warranted a police
report and resulted in some level of personal injury. These incidents are more fully
and accurately reported in the crash report template than non-injury crashes, which
are usually not reported at the scene and often have only sketchy information about
the circumstances. The final basis for making the decision depends, once more on the
ultimate objectives of the current research, as well as pragmatic matters of data
availability and quality. Whatever decision is made, the basis for the scope of data
employed in the task in hand should always be reported explicitly alongside the
results, so consumers of the data can understand the consequences when interpreting
them.

18
IV. Safety Data and Key Factors in KSA Context
Safety data is collection of facts and related information about car crashes. Generally,
they are divided into categories; critical safety data and supplemental safety data.
Critical safety data is major component of data that is needed to make an effective and
informed decision about traffic safety. Critical safety data are divided into three
categories; crash data, road data, and traffic data.

A. Crash data
Crash data is a group of variables or attributes that collected after a crash happened.
World Road Association provided key variables or attributes are needed to collect at
the event of crash. The following are a list of these variables.

• Crash location: geographic coordinates;

• Crash Time and Date: time of day, day of week, month of year;
• Information on those who were involved: road user type, age, gender, injury
severity;
• Details regarding the road: intersection, speed limit, alignment, marking;
• Details on the environment: light conditions, weather, road condition;
• Information regarding what happened in the crash: vehicle movement types,
objects struck (including off-road), and contributory factors such as speed,
alcohol use or driver distraction;
• Vehicle Type: type of vehicles involved, e.g., car, bus, and truck.

These variables provide detailed information about a single crash. However, these
variables describe the crash, the people involved in the crash, the vehicle in involved
in the crash, the road that crash occurred in, the environment that crash occurred in.
Usually, these are divided into three: Crash, Vehicle, and People. Crash data are
collected by filling forms, either paper-based or electronic based, that have three
sections designated for crash, vehicle, and people. Then, this information is keyed in
spreadsheet or database for each of the three. In addition, the three categories are
linked with unique identifiers.

In Saudi Arabia, there are several government sectors collecting the crash data. They
are:

• Traffic police department

• Najm Company
• Transport Ministry

19
Traffic police department at the ministry of interior collects crash data on the entire
area of Saudi Arabia. Recently, they have developed a system called “Basher” to
record the crash data. The following figures show the electronic forms of “Basher”.

Basher contains three main sections, information about the crash, vehicle, and people.
Most of the fields are drop list menus that have different predefined categorizes. For
example, a field in first person section called ‘health condition’ have five categories;
dead in the site, taken to the hospital, cure in the site, dead in the hospital, and no
injuries. This variable international called injury severity. However, based on this
variable one can derive another variable called crash severity that has three categories;
fatal crash, injury crash, property damage only crash. This are defined as follow;

• Fatal crash is defined if the crash results in at least one person died;
• Injury crash is defined if the crash results in at least one person injured;
• Property damage crash if the crash has no injuries or fatalities.

Najm Company is a private company that the first responder to non-injury crashes
that covers 13 cities of Saudi Arabia. The company has its own system or forms to
record only non-injury crashes.

Transport Ministry developed its own system to collect and store the crash data
occurred on roadways under their administration, most of their roads located outside
of the cities. However, people involved in crashes and died in the hospital, they

20
recorded as injured. Thus, the complete and accuracy of crash data can be found at the
traffic police department.

B. Road data
Road data is information about road that may affect the traffic safety outcomes. World
Road Association provides a basic list and they are presented as follow:

• Road class: freeway, arterial, and local road;

• Road width and type: divided (plus width and type of median) or undivided;
• Lane number and widths-in each direction;
• Intersection or crossing type: cross or “T” intersection, roundabout, railway or
other;
• Traffic control devices: signals and signs;
• Road alignment: horizontal and vertical curves;
• Street lighting: lighted or unlighted;
• Pavement type and condition: asphalt, concrete, brick, and unpaved –
roughness, potholes, and skid resistance;
• Shoulders width and type: paved or unpaved;
• Posted Speed Limit;
• Roadside hazard; including distance to hazard and type;
• Pedestrian facilities; presence of footpath, pedestrian crossing, and type.

Transport Ministry have records of road statistics and types only on roadways
administered by them. These roads usually located outside of cities and on the
boundaries of cities. These roadways are divided into three categories:

• Main roads: are the roads that start with one digit or two such as, 5 and 22 in
the middle of the Kingdome's map. They are also the roads that connect the
Kingdome's regions with the large cities.
• Secondary roads are the roads that are assigned by three digits such as, 222 in
the middle of the roads map. Also, they are the roads that connect between
medium-population cities with the main roads network.
• Sub-roads, are the roads that are assigned by four digits such as, 3333 in the
middle of the roads map. They are the roads that connect between villages,
hamlets, farms, and others other routes.

Ministry of Municipal and Rural Affairs and Housing have records of road statistics
and types on roadways inside the cities. In addition, there some government sectors
administrate roads on some part of the city such as The Higher Commission for the
Development of Riyadh. The roadways inside the cities can be classified into
highways (rings), arterials, collector streets, local streets. These mainly vary in two
functions: accessibility and mobility. As the accessibility increases the mobility
decreases. For example, ring roads have limited accesses to but higher speeds while
the opposite for local streets.

21
C. Traffic data
The same government sectors administered their roads they also collect traffic data.
The following provide some the traffic variables.

• Traffic volumes: Average Annual Daily Traffic.

• Traffic mix: percentage of different vehicle type; car, bus, and truck.
• Vehicle speeds: mean and 85th percentile speeds, compliance with speed limit

Data collection methods: through manual traffic counts or automated means,

(pneumatic tubes or permanent data collection device installed in the pavement.

22
V. Non-traditional safety data and the importance of data
quality
A. Required safety data and responsible partners to measure road
safety
Road safety data is typically collected by a wide range of different organisations.
Working effectively across and between organisations to gather the best and most
relevant data possible is an important part of the process of using road safety data
effectively. Always remember that organisations collect information for their own
purposes, following procedures which are designed to meet their own priorities. As a
result, data may not always be available to standards which would be ideal from a
road safety perspective. Be ready to work effectively and positively with data
providers, even though the data they hold may require you to make some additional
effort to make it fit for your purposes.

1. Highway authorities

These are public bodies that hold legal responsibility for the operation and
maintenance of road networks. They are a source of valuable information for all road
safety professionals and are also often actively engaged in road safety activities.

Among the data they hold may be:

• inventories of road infrastructure

• Surveys of their networks, including traffic counts, camera footage,
maintenance inspections and regular surveys.

The challenges presented by this data arise because these authorities naturally most
concerned with the state of the road as it is at present. They may be less focussed on
recording which changes were made in the past, and when they were put into effect.
This can be a problem for longitudinal analyses, covering crash records over a period
of several years. It is sometimes not possible to find out when engineering changes
were implemented, and as a result the condition of the road in the past may be hard to
reconstruct. This can sometimes present a challenge to the analyst.

2. Emergency services

These services can be a valuable source of road safety information. As well as police
recorded crash data, medical and rescue services may also record data concerning
crashes and casualties that can be of great value. Among the data they hold may be:

• Medical records which can reveal more detailed and reliable severity
information than is available from the police.
• Records of enforcement activity which can augment traffic and crash data.
• More detailed material about crashes from more extensive investigations than
those recorded by crash templates.

23
One challenge faced by analysts wishing to use these data is that it is not always easy
to access. Regulations concerning privacy and sensitivity must always be adhered to.
Also, matching between data sets collected by different agencies can be time
consuming or even impossible. Before embarking on substantive analysis with this
data, you may have to decide, in the light of your research objectives, if it is
worthwhile.

3. Government

Government data of many kinds is relevant to road safety.

• Government bodies hold details of driving license holders and registered

vehicles, which can provide valuable insight into fleet sizes and numbers of
drivers.
• Government level census and survey data provides useful information about
the communities served by roads, including their size, demographic profiles,
and nature.

A challenge with government data is that although it is valuable and usually freely
available, it may not necessarily be fit for road safety purposes as supplied. The
geographical and chronological scope is not always useful because it may not match
other data sources being used. Also, you should think about what apparently relevant
data are really telling us: for instance, you might like to know the number of active
drivers for specific areas or vehicle types, but all holders of a particular type of
licence cannot be presumed to be equally active drivers.

4. Commercial data

Some valuable data can be purchased or commissioned from commercial sources. For
example:

• Socio demographic profiling tools designed for marketing purposes can also
give useful insight into the community backgrounds of crash involved drivers.
• Commissioned surveys, of public opinion and of actual road user behaviour,
can provide valuable qualitative background to studies of behavioural road
safety issues.
• Telematic data is a growing new source of information, providing real time
records of driver activity and therefore is potentially a proxy for traffic surveys
where these are only sparsely available.

A disadvantage of commercial data is that it can be costly to obtain. It is important to

ensure that it is genuinely required to fulfil your objectives. You should also verify
that the data quality, format, and scope will be fit for your purpose before proceeding
to acquire it.

5. Academic research

Finally, much valuable insight can be obtained from academic research. Drawing on
the conclusions of wide scale studies can guide and inform your own investigations

24
and prevent you from needlessly repeating ground which has already been covered or
approaching a question with an unsuitable methodology. For example, studies of
driver behaviour using simulators can provide insight into how road users behave and
react without the difficulties of investigating real life events.

An advantage of academic studies is that the peer review process provides a level of
certainty about quality and reliability. However, before relying on it you should
examine the hypotheses upon which the studies were based, to ensure that they are
truly relevant to the needs of your research.

B. Supplementary safety data

Although road safety critical data is the best starting point for many important
activities, it is not the only way to achieve effective results. It is best practice to
supplement traditional data from secondary sources. Secondary sources provide useful
information about the communities served by road networks, and therefore provide a
valuable type of insight which critical data, with its tendency to centre on roads rather
than people, can easily overlook. Also, secondary data sources can sometimes fill in
gaps where traditional sources Are lacking.

1. Behavioural data

Data which sheds light on how road users behave on the network and can take your
analysis beyond what can be achieved with critical data only. Information about crash
involvement only represents a small proportion of overall road user behaviour, so a
wider view of behaviour provides clearer insight into the actions of road users and
their consequences.

. Examples of such data include

• information about road user actions from ‘near miss’ incidents; this can
sometimes be obtained by using machine learning to analyse CCTV footage
rather than costly and time consuming human observation.
• Data about enforcement activity is an important source of information about
how road users approach the potentially hazardous matter of speed choice. The
police often record when and where enforcement activity takes place, and how
many violations were detected. This gives a clear picture of how much unsafe
behaviour occurs at the locations, and how that behaviour changes over time
in reaction to circumstances, including the enforcement itself.
• Medical records, although difficult to relate accurately to police recorded
crashes on specific roads, can provide valuable context about frequency and
degree of road injury in an area. The quality of this information is enhanced by
the fact that it is sourced from medical professionals who are better placed
than police officers to make judgements about injury severity.

Although useful, behavioural data is sometimes challenging to integrate with critical

data sources. Also, there are often significant hurdles to overcome to obtain this data
in anything other than summary form, for instance due to the imperative of protecting
patient confidentiality.

25
2. Demographic data

Demographic data, sourced from both government and commercial sources, can
provide important background about the population’s exposure to road danger. It
provides insight into the differential nature of danger experienced by different
communities. For example, urban communities experience road danger differently to
inhabitants of rural areas, and it is well known that poorer communities generally
experience higher risk than more affluent ones.

Examples of this type of data include:

• The number of holders of certain licence types in different areas can be a

useful proxy for active driver population, and the length of time a driver has
held a licence likewise for driver experience.
• Vehicle history can provide insight into the distribution of owned vehicle
types and how they are distributed between fleet and personal ownership.
Increasingly in future it will become important to understand now many
vehicles carry different types of technology designed to reduce risk.
• Societal factors such as deprivation and rurality, as well as human factors such
as attitudes and peer pressure, are known to affect road safety outcomes and
cannot be ignored. The basic source for information of this kind can often be
sourced freely from government data such as that provided by census returns
and can usefully be augmented by socio demographic profiling information,
usually from commercial sources. For example, if it is possible to associate
participants in crashes with profiles, perhaps by matching postal codes, then
considerable insight into which communities are associated with which
behaviours and what dangers can be obtained.
• Direct automated observation of road users is a source of information with
considerable potential including telematic data and ‘black box’ in-vehicle
recorders which can include insightful data such as incidents of sudden
braking and excessive speed.
• Naturalistic observation using video sources such as CCTV and dashcams is
an alternative, although it is challenging to separate conclusions from the
prejudices of the observer.
• Modelling of road user behaviour is also valuable, including quantitative data
from sources such as studies using driving simulators, as well as qualitative
information from public surveys and opinion polls.

C. How the quality of safety data impacts decision-making for more

effective interventions
Data quality is a measure of the state of an information set, describing its fidelity to
the real world and its fitness for purpose. Data can never completely replicate the real
world objects, people, and events they represent, but they should approximate them
with sufficient precision to be useful.

To be fit for purpose, data should be consistent and accurate enough to match your
intentions. Therefore, data quality is not absolute, but is relative to context and

26
purpose. If you are clear about your objectives, this will help you to assess whether a
data set has sufficient quality to be fit for your purpose.

The best analysis requires both good analytical practice and good data quality. This
should always be the analyst’s goal, so time invested in obtaining and cleansing data
is not time wasted. On the other hand, a lack of sufficient data quality risks the worst
case scenario, where misleading results from poorly conducted analysis could lead to
bad decisions being made.

As this table suggests, high quality data by itself is not enough to guarantee useful
results, and poor decisions are likely to result if good analytical practice is not
followed. Sometimes it is possible to overcome inadequate data to produce useful
results, although the value will be limited by the inherent limitations of the data. It is
important that limitations which are consequent to data quality are clearly reported
alongside results, so consumers of the data will understand that these consequences
are present.

Sometimes you may find that data which would ideally serve your objectives are not
available, or if your data has inherent quality issues which will have an impact on
your ability to deliver your objective completely. For example, you may set out to
explore the road safety consequences of speed choice behaviour of road users on a
route. Ideally, to address this objective you would need:

• crash data which made it possible to identify crashes in which speed choice
was a contributing factor;
• comprehensive traffic data with vehicle speeds over time at key points on the
route;
• the location and duration of speed enforcement activity on the route with
numbers of detected violations; and
27
 • Knowledge of which speed limits were in force at what point and when.

If one or more of these data were not available to you, you may be able to use
secondary data to act as a proxy for the missing data. For instance, telematic data
exposing actual driver behaviour could serve as a proxy for speed survey data, or
driver opinions about speeding behaviour, limits and enforcement on the route may
give an alternative insight to enforcement data.

If some data were available but with known quality issues, you may be able to address
some of them as part of your analysis by employing some of the techniques discussed
in the next section. Alternatively, you may decide to limit your analysis, perhaps by
covering only places or periods for which data coverage was available or changing
your analytical approach to concentrate exclusively on parts of the issue which your
data covered adequately.

Either way, you should report the data quality issues you encountered, the steps you
were able to take to resolve them, the ways in which your analysis was limited by the
issues, and your recommendations for improving data quality in the future. In this
way, your results will not only give the best possible immediate support for good
decision making now but will help to signpost paths to future improvements.

D. Scientific methods of data preparation to improve its quality

Six important issues to consider when judging data quality are the data set’s accuracy,
completeness, timeliness, consistency, integration, and accessibility. These are not
necessarily of equal weight, as which data issues are the most important in any given
situation will depend on what is reasonable for the current task. For example,
completeness is highly desirable when compiling reports of fatal crashes where lives
have been lost; but is not particularly important when compiling sample data of
damage only crashes, when absolute completeness is not expected and probably not
even attainable.

Fortunately, there are steps which can be taken to improve data quality. Data
validation is the most desirable way of addressing the accuracy of data. This refers to
methods of checking data at the point of collection, applying constraints to the entry
process so certain specific examples of inaccurate information cannot be entered in
the first place. A simple example would be making it impossible to enter a date in the
future when completing a crash record or rejecting text entry where a number is
expected. Some data validation can be applied after data entry by automated systems,
and this is most often used to ensure consistency. For example, the reported road type
at a crash scene might be checked subsequently against road data to ensure that the
crash location did contain the reported features, such as the presence of a median.

Cleansing, also sometimes referred to as data correction, describes a process carried

out after data entry is complete, when a data set is being prepared for use in a
particular context. A simple example of cleansing might be examining a data set for
duplicated records and removing any identified cases. A more involved example
might be ensuring that complex data sets which require multiple record to record a
single item, such as multiple vehicles involved in a crash are correctly related with

28
each other. There should be no ‘orphaned’ vehicles which are not related to a crash,
neither any ‘widowed’ crashes which have no record of involved vehicles.

Data normalisation refers to how data sets relate to each other. It is most important for
ensuring consistency, although well implemented normalisation can also address
timeliness. At a simple level, information which refers to different aspects of the real
world should be stored in separate tables, which can then be logically related to each
other as required for the job in hand. For instance, a list of traffic counts need not
repeat the complete details of the road in every record; it is both more efficient and
easier to check if information about the route is stored separately and the list of counts
refers to it by a single code. This prevents the inadvertent creation of data anomalies
when editing or updating occurs; for example, a route disappearing from results
because no counts associated with it are present in the most recent version of the list.

Standardisation refers to understanding exactly what a datum means in relation to the

real world and ensuring that apparently similar measures from different sources do not
become confused. It is important for to achieve both accuracy and consistency. For
example, a measure reporting a date and time may appear to be measuring the same
thing, but if the data is derived from different sources this may not be the case. A
casualty may be recorded by the police at the time an officer attended the scene but
may only be recorded for medical record purposes at time of admission to hospital
which could be very different. To understand how data needs to be standardised, it is
sometimes necessary to return to the agency which supplied it and request
clarification of exactly what is meant by ambiguous fields.

E. challenges and limitations that road safety professionals face when

collecting and preparing related data
If achieving good data quality seems an imposing objective, it is worth remembering
that decisions about data should be considered relative to the objective of the research.
You only need to make data fit for your purpose, not for any conceivable purpose.
You may find that a certain data source is unavailable or unreliable, but you may be
able to fill it with a proxy. For example, if you lack data on which speed limits were
applied on which roads on certain dates, it may be that telematic data could inform
you the speeds at which a large sample of vehicles were travelling at on those roads,
which for some analytical purposes may be an acceptable proxy.

You may have to choose between larger data sets which provide a large sample but
lack detail or consistency, or far smaller ones with higher levels of detail and fidelity
but providing a smaller and therefore perhaps a less representative sample. Which one
to choose will depend on the priorities for the task in hand, and it may be that steering
a middle course, for example using a selective sample of the larger data set where
incomplete records are disregarded, could be the best way to proceed.

29
VI. Quantitative Safety Analysis
A. What can be done with traditional measures: Crash data, Volume
Data, and Road Data
1. Importance of Objectives

Before setting out on any data analysis task, including road safety analysis, it is
essential to be clear about the objectives of the project. A clear statement of intention
is helpful in many ways because if provides a point of reference for making decisions
at every stage of the process. Without such a vision, it is all too easy to lose sight of
the purpose of your research and be deprived of the focus which helps to produce a
coherent process. Objectives also provide a way of reflecting on a piece of work once
it is complete, providing perspective on which aspects were successful and where
opportunities for future improvement may lie.

Your objectives may take many forms: a brief from a client, a hypothesis to evaluate
or a set of questions to answer, or agency policy and procedures guidance can all
provide the basis for sound and meaningful objectives. If the circumstances of your
work do not provide an obvious objective, it is worthwhile to establish an explicit
objective for yourself.

Once an objective is established, you can use it a basis for decision making. For
instance, road safety projects often involve co-ordinating multiple data sources and
finding ways to render them compatible with each other. Having an objective clearly
in sight can help if difficult decisions must be made. Perhaps two sources cannot be
analysed together, and one must be chosen in preference to the other; or perhaps a
weighting factor must be applied to different results. A clear view of objectives is
invaluable in making these decisions for the right reasons.

30
2. Objectives Help Sort and Organize Data

When working with critical road safety data, there are common analytical types which
use each source, and use different sources in conjunction. This diagram below
(taken from Chapter 7 in Unit 3 of Road Safety Fundamentals) shows how the
traditional forms of road safety data can be analysed in conjunction to achieve
common analytical objectives. For instance, the process of identifying risk factors on
a certain route requires both an understanding of data relating to the road itself as well
as examination of the factors which are apparent from the crashes which occurred
there.

31
B. Skills of using related/available safety data to reach solid results
and effective treatments
Not all analytical outcomes are as clearly defined as the common purposes identified
in the diagram above. However, the fundamental principle of having clearly defined
research objectives remains central. By remaining focussed on objectives, it is
possible to follow through novel analyses with sound process, leading to relevant
results and useful conclusions.

The first stage of putting this into practice is to identify your preferred data sources
based on the objectives, and then to examine those sources for practical suitability.
This may involve discussions with data providers and putting data quality measures in
place. Once your data is assembled you should devise an analytical approach and
perform preliminary analysis. You should check both your approach and your process
by dip-checking results and correlating them with original data, to ensure fidelity.
When you believe your process is sound you should ideally seek out a peer reviewer,
for example a colleague or teacher, to examine your process from an outside
perspective as this can often lead to suggestions for useful improvements. Finally, you
should assemble the results in a format which makes them accessible to the intended
audience, along with an explanation of your methodology and an honest assessment
of any weaknesses you have not been able to remove, and for example reporting
issues in the original data or methods you may wish to have used which were not
realisable in practice.

1. Pitfalls to Avoid

Two important pitfalls are worth bearing in mind when deciding on analytical
process. The first is that correlation does not by itself imply causation. Merely
because two trends coincide does not prove they are necessarily related. If you suspect
that two correlated events or trends are related, you should examine the data to
discover if there are evidence-based reasons, such as statistical relationships, for
supposing this relationship exists. It is not good process to assume such a relationship
without evidence.

Another important pitfall to avoid is to forget the random nature of real world events.
Remember that data is not reality, it merely approximates to it. For example,
randomly occurring events will from time to time occur with greater frequency, but
over the long term they will tend to regress to their mean and stabilise. Short term
fluctuations in data may only be manifestations of the randomness of events and their
eventual regression and have little inherent meaning.

2. Road Safety is a Stochastic Condition

Road safety is necessarily a stochastic4 model: the open system nature of road
networks inevitably means a degree of randomness is present, so even where the same
or very similar conditions apply, diverse outcomes can result in different individual

4
A stochastic distribution refers to data which inherently contains a random probability which cannot be predicted
precisely.
32
cases. Data models can appear seductively deterministic5: a given set of inputs may be
presumed to result inevitably in some desired output. This is unlikely to be the case in
real world situations. Careful thought should be given at how a deterministic model
can be tested against reality to find out whether it truly provides a fair approximation
of actual outcomes. For instance,6 studies of speed enforcement have shown that
extensive deployment of speed cameras does lead to a reduction in casualties, but this
relationship is not deterministic in every individual deployment.

C. Definitions and primary components of quantitative safety analysis

The starting points for effective quantitative analysis are a clear objective and
appropriate data of sufficient quality. Armed with these prerequisites, you can set out
to achieve your goals.

1. Metrics

One of the first questions to consider is what to count when working with crash data;
that is, which measures connected with the crashes most accurately reflect your
purpose. As we have seen, crash data is not a single metric, but affords the analysis a
choice of approaches because of associated vehicles and casualties. The three most
common approaches are laid out below, with an indication of the advantages and
drawbacks of each. They are not mutually exclusive, and different metrics may be
used for different purposes in your analysis.

The most straightforward approach is to use numbers of crashes as a basic metric, and
for many applications that is a sensible approach. Counting crashes provides a clear
indication of known instances where the network has failed to deliver its most
fundamental purpose: getting users through their journeys without incurring harm. For
analyses where insight into network performance is a key objective, counting crashes
is usually a sound way to proceed. However, this is not appropriate in every case.
Counting crashes can sometimes obscure the number of people involved in a crash,
and the nature of their involvement.

Where the objective of the analysis is to gain insight into driver behaviour, it is often
more appropriate to count vehicles, and the drivers in charge of them, as a basic
metric. This approach provides a clearer picture of vehicle involvement by type and
manoeuvre, and the mix of vehicles in multi vehicle crashes. Counting drivers allows
study of the mixture of demographics and behaviours present among crash
participants. However, the relationship between drivers and the overall crash severity
is not direct, and if severity is involved in the analysis then this approach can be
misleading.

Where information is to be provided in the public domain or used to measure progress

against targets, it is often most useful to count casualties. This provides a direct
relationship between your metric and injury severity and has value in providing
demographic insight. It is also the most appropriate metric for studying pedestrian

5
A deterministic data model is one where outputs can be predicted precisely given complete knowledge of inputs.
6
https://www.racfoundation.org/wp-content/uploads/2017/11/efficacy_of_speed_cameras_allsop_181110.pdf
33
involvement in crashes. However, the number of casualties reported from crashes can
create distorted picture of relative danger, as sometimes large numbers of casualties
result from relatively minor crashes, with smaller numbers resulting from much more
serious incidents.

2. Context

Once you have identified the basic metrics most appropriate to your purpose, you
should consider the most appropriate context in which your analysis should set them.
Many analyses require road safety information to be set into context so consumers of
the analysis can understand whether outcomes are expected or unusual. This requires
a sufficiently broad base of information against which to judge metrics, so they can be
evaluated in a meaningful way.

Sometimes this can be done with a longitudinal approach, examining outcomes over a
long period of time. Alternatively, a comparative approach may be preferred, using
other similar areas, locations, areas, or networks as controls against which to judge
performance. It is also possible for both these approaches to be used together, for
example by comparing change over time in the area being studied against change over
the same period in a control area.

Whichever approach to analytical comparison is used, it is often useful to normalise

measures against an underlying trend. These are discussed in more detail in the next
section. The best approach to normalising metrics depends partly on which metrics
you chose to use.

A common way to normalise road safety crash data is by using traffic flow,
expressing performance by as outcomes relative to the total amount of traffic. This is
referred to as a rate. This has the advantage of factoring out high volume as simply a
function of high traffic but has the disadvantage that the relationship between traffic
volume and road danger is not a straightforward one.

A similar method used for personal outcomes, where the metric chosen is a count of
people rather than events, by expressing performance relative to a relevant measure of
population. This is referred to as an expression of risk. It has the advantage of
expressing danger in relatable terms for individuals, but the disadvantage that
exposure to danger is not distributed evenly throughout a population.

D. Main concepts when analyzing safety data

1. Rates

2. Finding the denominator: traffic

To express road safety metrics such as a count of crashes as a rate, it is first necessary
to calculate the amount of traffic. Traffic measures are produced from point counts as
an average daily flow, in other words the mean number of vehicles which crossed the
point over a 24 hour period during the time the count was active. Ideally, this number
should include every day of a complete annual cycle, to capture seasonal fluctuations
fully. This is known as Annual Average Daily Flow, or AADF.
34
If AADF is not available, then it should be estimated using the available average flow
as a basis. Care should be taken to decide which directions of travel are being
included. Where a median is not present, it is generally preferable to use flow data
from both directions and combine them.

AADF is used as a basis to estimate total traffic by expressing it as the total distance
travelled by vehicles at the given point over the course of a year. This involves
multiplying AADF by the length of road to which the base count is relevant, usually
meaning either the distance between count points or the distance on either side of a
count point without a major junction. This total vehicle distance travelled figure forms
the denominator for calculating a rate.

3. Calculating the rate

The rate is then expressed as the number of events being measured (often the total
crashes on the same or a similar length of road to that used to calculate the flow)
divided by the total vehicle distance travelled. The resulting figure is the crash rate,
usually simplified to crashes per vehicle kilometres travelled at an appropriate order
of magnitude. In the example shown below, an AADF of about 25,000 km per annum
over an 8 km route with seven fatal and serious injury collisions over three years gave
a rate of 31.5 collisions per billion vehicle km.

4. Matching numerator and denominator

Risk is calculated in a similar manner to rates but using an underlying population

figure instead of traffic flow. Where the denominator for the calculation is a count of
people, it is generally preferable also to use a head count metric as the numerator. For
example, the number of involved drivers of a certain vehicle type or age group, or the
total number of persons killed, is typical numerators for such a calculation.

It is important to choose numerators and denominators which are as relevant to each

other as possible: for instance, if the metric used describes road users in a certain age
group, then the underlying population group used should refer to the same age group
as closely as possible.

35
This approach is often used for national comparison. For
example, Romania and the United Kingdom over the last
five years have similar absolute numbers of road
fatalities; around 1,800 to 1,900 per year on average.
However, the population of Romania is around 20
million, while the population of the UK is over 65
million. Consequently, the population’s risk of road
death differs considerably; in 2017 Romania had around
99 deaths per million inhabitants, while the UK had
around 287.

5. Challenges and mitigations

There are potential pitfalls in working with both rates and risk.

Rates require the underlying flow data to be as relevant to the denominator as

possible, but by their very nature most traffic flows are estimated and may not refer to
the same spatial or chronological scope as the accompanying metric. Also, traffic data
is often not available for long enough periods or in sufficient detail. For example, it
may only cover major roads where count points are plentiful or permanent, or it may
not be available for different vehicle types.

To mitigate these difficulties, it is sometimes possible to use other denominator

metrics such as road length, although such approaches require care to ensure that such
a relatively crude measure is not misrepresented by comparing dissimilar routes.

Another potential pitfall is to assume that the relationship between traffic and road
danger is linear. In fact, the relationship is more complicated, especially varying for
different road and junction types. This can be mitigated by using techniques which
inherently take account of the underlying road type, for example by using Safety
Performance Factors.

Risk can also be a misleading measure. The most immediate problem is that rarely do
all members of a population have equivalent exposure to the risk being measured. For
instance, some members of a population may cross busy roads frequently, while
others do so only rarely. This is particularly problematic in the case of risk by vehicle
type, as distances travelled in vehicles are almost never available.

Sometimes it is possible to mitigate these problems by using proxies such as fleet

sizes or number of licence holders, either in palace of or as a weighting factor to
populations. However, these data these carry their own biases and using them will
involve making assumptions about population behaviour which may have limited
validity. When faced with such decisions, analysts are generally well advised to refer

7
https://ec.europa.eu/transport/facts-fundings/scoreboard/compare/people/road-fatalities_en
36
to the original objectives of the analysis and making decisions about which
assumptions may be regarded as valid for the present purpose. Such assumptions
should always be described explicitly alongside results, so consumers of outcomes
can make reasoned judgements about the value and limitations of the results.

6. Machine Learning

An alternative analytical approach which is becoming increasingly practical for

analysts is to deploy machine learning. This involves training a neural network to
carry out multivariate analysis.

In summary, a neural network consists of an array of calculation nodes, called

neurons, which associate data by finding patterns in large input data sets. This is done
by repeatedly passing a set of inputs with known outputs (for instance, the
characteristics of a road network with a known number of casualties) through the
network unto the output of the network consistently resembles the known real world
output. This ‘trains’ the model to provide a sound estimate of the likely output for any
other similarly structured input array.

The advantage of a machine learning approach is that it can take a very large number
of input variables into account, potentially including numeric expressions of human
factors such as behaviours which led to collisions. Because the model is trained over
much iteration with these large input data sets, it can often recognise subtle
differences which would not be apparent to a human analyst or to conventional linear
statistical analyses.

However, machine learning also has disadvantages. How the model is trained is
highly dependent on type of neural network used and the scope of input variables
chosen for training. It tends to be more suitable for wide scale applications, for
example a whole city rather than a specific route. In practice it can be complex to
apply, and because the exact methodology used to generate the output is not apparent
it can feel like a ‘black box’ system where it is not possible for consumers of the data
to scrutinise its process. It is likely that this style of analysis will become more refined
and widely used in the future.

E. Safety Performance Functions and Crash Modification Factors

Quantitative safety analysis is a method that uses safety data to build predictive
models. These models can then be used for evaluation and improvement of traffic
safety on a network, facility, or specific site. The main purpose of these models is to

quantify the effect of associated variables on the outcomes. For example, as the
AADT increases the crashes generally increases but the amount of this directional
effect is unknown without developing these predictive models.

37
 Safety Data

regression/machine
learning methods

Predicted No. of Crashes

This analysis gives ways to expect the outcomes, number of crashes, through changes
applied or occurred on input variables, traffic variables and/or road characteristics
variables. However, these models are developed based on the safety data and by
regression or machine learning methods.

1. Safety Performance Functions (SPF)

Safety performance functions are predictive regression models that take a few input
variables, primarily AADT and segment length, to predict total crashes or certain type
of crash severity frequency per year for a base condition. SPFs can be developed for
different type of roadway for either segments or intersections. Highway Safety
Manual (HSM) provided detailed information on safety performance functions for
three roadway types; rural two-lane two-way roads, rural multilane highways, and
Urban and Suburban Arterials. Generally, these roads are divided into homogenous
segments or intersections with similar attributes, road characteristic, to drive or
develop SPFs. An example of developed SPF for rural multilane highways by HSM is
shown below.

38
 Example 1: SPF for four-lane divided rural highway segments

The base condition for safety performance function for divided roadway
segments on multilane highways are:

➢ Lane width : 12 feet

➢ Right shoulder width : 8 feet
➢ Median width : 30 feet
➢ Lighting : None
➢ Automated speed enforcement : None

The developed SPF for four lanes divided rural highway segments for
base condition are:

N = 𝑒 (𝑎+𝑏×ln(𝐴𝐴𝐷𝑇)+ln(𝐿)) (1)

Where

N: Expected number of crashes per year;

AADT: annual average daily traffic (vehicle/day) on roadway segment;

L: length of roadway segment (miles);

a,b = regression coefficients.

The coefficients for total and fatal-injury crashes are

Severity Level a b

Total -9.025 1.049

Fatal and injury -8.837 0.958

Highway Safety Manual (2010)

Thus, the effect of traffic volume (AADT) on number of crashes is estimated in the
SPF, while Crash Modification Factors (CMFs) quantify the effects of changes on
geometric design and traffic control features from the base condition (explained

39
below). However, a good practice to calibrate the coefficients based on the safety data
collected on the local condition (see the case study)

2. Crash Modification Factors (CMF)

Crash modification factors are measures to quantify the effect of a particular treatment
applied to the base condition of SPFs. Thus, CMF is a factor that estimating the
expected change on predicted number of crashes of the base condition after a
particular treatment is implemented. CMFs are interpreted as follow:

• CMF < 1: it means that the treatment reduces the number of crashes
• CMF > 1: it means that the treatment increases the number of crashes

For example, suppose the estimated CMF for installing a traffic signal at non-
signalized intersection is 0.70 and the predicted number of crashes computed by SPFs
is 10 crashes per year for non-signalized intersection.

Then, the expected number of crashes after installing the traffic signal is calculated as
follow:

Predicted No. of crashes = 0.70 * 10 = 7 crashes / year

Thus, installing the traffic signal is expected to reduce the number of crashes by 30%.
However, this is just an example to illustrate CMFs and not reflect the real change.
For more information about CMFs, you can consult HSM. Below, a continuing
example of CMFs of four lanes divided rural highway segments SPF provided by
HSM.

40
Example 2: CMFs for four-lane divided rural highway segments

Any changes applied to the base condition, or it is called treatment, they

are quantified by CMFs. The following table shows the CMF for modifying
the median width in feet:

Median width (ft) CMF

10 1.04

20 1.02

30 1.00

40 0.99

50 0.97

60 0.96

70 0.96

80 0.95

90 0.94

100 0.94

This CMF for median width is adopted by HSM from a study conducted by
Harkey et al.

This CMF can multiply directly by the predicted No. of crashes computed
by SPF presented in Example 1. The study assumes the median width only
affects cross-median crashes and not affecting other types of crashes.
Highway Safety Manual (2010)

41
VII. Case Study: SPF for Urban four-lane divided roads in
Riyadh, Saudi Arabia
As mentioned earlier, good practice to calibrate the SPFs to local condition. It
improves the accuracy of the SPFs to predict number of crashes as well as counting
for unobservable variables or differences among populations or countries. The
following is a paper study conducted to calibrate and developed the SPF for urban
four-lane divided roads in Riyadh, Saudi Arabia. The study also discusses three
CMFs; angle parking, lighting, and Median width. However, the developed CMF for
angle parking only presented here. The other information about other CMFs in the
study can be found in full text of paper.

42
Title: Transferability and Calibration of Highway Safety Manual
Performance Functions and Development of new Models for Urban
Four-Lane Divided Roads in Riyadh, Saudi Arabia

Objective: Developing Safety Performance Functions (SPF) and Crash

Modification Factor (CMF) for Urban Four-Lane Divided Roads in Riyadh.

The study collected the safety data; crash data, traffic data, and road
data, from three sources:

➢ Traffic Police Department

➢ The higher commission for the development of Riyadh
➢ GIS environment and google maps

They collected all roads that fallen in the category of urban four-lane
divided roads in Riyadh city. Then, they divided these into homogenous
segments. Lastly, they mapped crash data and traffic data into these
segments to have only one file to do the analysis on it. The following
shows the attributes or variables on 172 segments for the entire city of
Riyadh.

➢ Crash Frequency (Fatal + Injury Crashes)

➢ Segment Length
➢ AADT
➢ No. of driveways
➢ Speed Limit
➢ Median width
➢ Shoulder width

The base condition for this category as follow:

On-street parking: none, Median width= 15 ft, Median width: none,

Lighting: none

Automated speed: none

Study By: Khalid Al Kaaf, Mohamed Abdel-Aty

43
Developing CMF for angle parking using negative binomial regression:

Then, the authors fitted negative binomial regression model to the data
with indicator for angle parking; 0 for without angle parking and 1 for
with angle parking. The model is shown below:

𝑁 = 𝑒 (−4.84+0.507×ln(𝐴𝐴𝐷𝑇)+0.735×𝐿+0.422×𝑃𝐴𝑅𝐾) (2)

Where;

N: No. of predicted crashes per year;

AADT: Annual Average Daily Traffic, veh/year;

L: segment length;

PARK: an indicator for the presence of angle parking.

Then, the CMF can compute from the model as follow:

CMF for angle park = 𝑒 0.422 = 1.52.

This indicates the angle parking increased the predicted number of

crashes by 52%.

Developed Model of SPF for Urban four-lane divided roads in Riyadh:

Model based on HSM:

N = 𝑒 (𝛽0+𝛽1×𝐴𝐴𝐷𝑇+𝛽2×𝐿) (3)

Model based on HSM:

N = 𝑒 (−3.9+0.435×ln (𝐴𝐴𝐷𝑇)+0.73×𝐿) (4)

Study By: Khalid Al Kaaf, Mohamed Abdel-Aty

44
Developed New Model:

N = 𝑒 (−4.76+0.583×ln (𝐴𝐴𝐷𝑇)+0.77𝐿−0.025×𝑆𝐿+0.0144×𝐷𝑆) (5)

Where;

SL: Speed limit, DS: driveway density.

Study By: Khalid Al Kaaf, Mohamed Abdel-Aty

45
Training programs for Road Safety Professional
Exam (Level 3):
RSP1-SA

Phase Three
22-24/03/2021

1
 Module 4
Solving Road Safety Problems

Overview

Unit Modules Topics in this Module

Different approaches of road safety management

1-Basics of Road Safety process

Network screening methodologies and identifying

their strengths and weaknesses
2-Effects of Human Behavior on Road Safety

Tools and techniques needed to diagnose crash

trends

3-Data and Its Role in Measuring Road Safety

The importance of using the surrogate safety
measures.

The distinction of the needed data used to measure

4- Solving Road Safety Problems safety at the macro- and micro levels

Overview of the road safety audit (RSA) in

diagnose process

5-Traffic Safety Strategic Plans and Programs

Suggest potential treatments based on the
outcomes of safety data analysis

The concepts of the safety performance function

(SPF) and crash modification factor (CMF)

Evaluate the feasibility of road safety treatment

using economic appraisal and the methodology of
comparing alternative countermeasures

2
 Outline

1 Road Safety Management Overview

(Corresponds to Presentation 8.1)

1.1 Decade of Action

1.2 Safe System Approach
1.3 Forgiving Roads
1.4 Road Safety Management (RSM)
1.4.1. 1st RSM Dimension - Institutional Management (MGT) Functions
1.4.2. 2nd RSM Dimension – Interventions
1.4.3. 3rd RSM Dimension – Results
1.5 Highway Safety Manual (HSM) - Roadway Safety Management Process
1.5.1. Network Screening
1.5.2. Diagnosis
1.5.3. Countermeasure Selection
1.5.4. Economic Appraisal
1.5.5. Project Prioritization
1.5.6. Safety Effectiveness Evaluation

2 Understanding Diagnostic Tools and Collision Patterns (Corresponds to

Presentation 8.2)

3 Focus on Road Safety Audits

(Corresponds to Presentation 8.4)
3.1 Road Safety Audits as part of Management of Road Infrastructure Safety
3.2 Road Safety Audits
STEP 1: Identify the Project or Road/Intersection in Service to be Audited
STEP 2: Select a RSA Team
STEP 3: Conduct a Pre-Audit Meeting to Review Project Information
STEP 4: Analyze Drawings/Perform Field Observations Under Various Conditions
STEP 5: Conduct Audit Analysis and Prepare a Report of the Findings
STEP 6: Present Audit Findings to Project Owner/Design Team
STEP 7: Project Owner/Design Team Prepares a Formal Response
STEP 8: Incorporate Findings into the Project

4 System-wide Management supporting Countermeasure Selection

(Corresponds to Presentation 9.1)
4.1 General
4.2 An enhanced systemic approach to road safety

3
5 User-Specific Interventions & Countermeasure Resources
(Corresponds to Presentation 9.3)
5.1 Haddon's ten strategies for road traffic injury prevention
5.2 What specific interventions can be implemented?
5.3 Managing exposure to risk through transport and land-use policies
5.4 The role of education and publicity
5.5 Remedial action at high-risk crash sites
5.6 Setting and enforcing seat-belt
5.7 Setting and enforcing child restraint use
5.8 Strategies to Reduce Crashes Involving Young Drivers
5.9 Strategies to Reduce Crashes and Injuries Involving Older Drivers
5.10 GRSF - Guide for Road Safety Interventions Report, March 2021

6 Multidisciplinary Considerations in Safety Solutions

(Corresponds to Presentation 9.4)
6.1 United Nations Road Safety Trust Fund Global Framework Plan of Action for Road
Safety, 2018

7 Effectiveness Evaluations: Safety Benefits

(Corresponds to Presentation 10.1)
7.1 Crash Reduction Factors (CRF), Crash Modification Factors (CMF) & Benefit/Cost
Ratios (BCR)
7.2 The CMF CLEARINGHOUSE
7.3 The PRACT Repository

8 Effective Evaluations: Economic Tools

(Corresponds to Presentation 10.2)
8.1 General
8.2 The SafetyCube DSS Calculator

9 Comparing Alternatives and Identifying Positive and Negative Impacts

(Corresponds to Presentation 10.4)
9.1 Economic Appraisal Process in HSM
9.2 Prioritize Projects Process in HSM
9.3 Current Road Safety DSS Worldwide
9.4 The SafetyCube DSS

4
1. Road Safety Management Overview

1.1 Decade of Action

On May 11, 2011, the World Health Organization, through the United Nations
Collaboration (UNRSC) announced the creation of the Decade of Action with the goal
to reduce projected fatalities between 2011 and 2020 by 50%. Five Pillars were created,
Pillar 1: Road Safety Management, Pillar 2: Safer Roads and Mobility, Pillar 3: Safer
Vehicles, Pillar 4: Safer Road Users, and Pillar 5: Post-Crash Response.

Pillar 1 Road Safety Management emphasizes the need to have a Lead Safety Agency
that is strong enough to mandate the use of global best practices for road safety. For
this reason, ideally the Lead Agency should chaired by someone in a position of power
such as the Transport Minister.

Pillar 2 deals with the infrastructure and focuses on the implementation of the Road
safety Audit and the Safe System Approach, which emphasizes safer road users and
vehicles, proper speed limits and “Forgiving Roads” and “Positive Guidance” to ensure
roads are safe for all road users. Roads must be designed and maintained using state
of the art technologies and global best practices. Unfortunately, it is estimated that
approximately 55% of the world’s roads are one or two stars on a five star basis. Too
often these roads are not designed with the safety of pedestrians, bicyclists or
motorcyclists in mind. Road designers must design roads for the vehicle population that
will be using the roads. Some countries vehicle mix have more than 70% motorcycles,
yet the roads are designed for cars and trucks.

Pillar 3 deals with Safer Vehicles and encourages the mandated use of safe cars, trucks,
and motorcycles that meet minimum crash performance levels based on the New Car
Assessment Program (NCAP) as well as regular, periodic vehicle inspections. With the
evolution of electric vehicles and autonomous vehicles, the recommendations from Pillar
3 are becoming even more critical considerations for road safety. Too many countries
allow vehicles to be used that do not include engineering developments that make the
vehicles safer to avoid a crash or to improve the protection of the occupants during an
impact.
Pillar 4 “Safer Road Users” recommends best practices for vulnerable road user safety,
as well as proper behaviors for motorists. These include, but are not limited to seat belt
usage, mobile phone usage, child restraints, drinking and driving, helmet use by
motorcyclists, and speeding.

Pillar 4 makes recommendations for educational practices for all road users from
children to graduated drivers’ licenses for new drivers to elderly drivers. It emphasizes
the need for significant punishments for driver violations that will be deterrents for the
drivers. Pillar 4 also encourages the use of electronic enforcement to improve
enforcement levels and remove the possibilities for police corruption.

5
The last Pillar, Pillar 5 Post Crash Care, emphasizes the need for procedures and
resources that will help to ensure proper care for road users involved in a crash. The
“Golden Hour” is critical for the survival of crash victims. Pillar 5 is also concerned with
crash re-construction as well as how quickly a road authority can respond to an incident
on the roads to reopen the roads to traffic.

Unfortunately, in spite of the existence of the First Decade of Action, fatalities rose from
1,240,000 in 2011 to 1,350,000 in 2020. Global road fatalities now are the number one
cause of death for ages 5-29. While the First Decade of Action did not achieve the
desired statistical goals, it did help to raise global awareness regarding the need for
road safety.

In February 2020, the UNRSC convened the Third Global Ministerial Conference in
Stockholm, Sweden and created the Second Decade of Action. The stated goal of the
Second Decade of Action is to reduce road traffic deaths and serious injuries (Road
Traffic Injuries) by at least 50% between 2021 and 2030. This proposal was adopted by
the United Nations General Assembly on August 18, 2020. While a 50% reduction sounds
overwhelming, it may be easier for Road Authorities to accept a 7% to 8% annual
reduction, which would be the equivalent of the 50% reduction over ten years.

The Academic Expert Group for the Third Global Ministerial Conference made nine
recommendations for the Second Decade of Action. Since each 1% increase in speed
results in about a ~ 4% increase in deaths each of the nine recommendations includes
language that relates to managing speed.
RECOMMENDATION #1 Sustainable Practices and Reporting
RECOMMENDATION #2 Procurement
RECOMMENDATION #3 Modal Shift RECOMMENDATION
#4 Child and Youth Health
RECOMMENDATION #5 Infrastructure
RECOMMENDATION #6 Safe Vehicles across the Globe
RECOMMENDATION #7 Zero Speeding
RECOMMENDATION #8 30 km/h Speed Limits in Urban Areas
RECOMMENDATION #9 Technology

The Draft for the Global Plan of Action for the Second Decade of Action that was
prepared by the WHO and UN regional Economic Commissions on February 2, 2021
proposes changing from the Five Pillars in the First Decade of Action to Five Focus Areas:
Infrastructure, Vehicle Safety, Behaviors, Post-Crash and Modal Shift. Unlike the First
Decade of Action that only reported on road fatalities, the Second Decade of Action will
focus on road fatalities plus serious injuries, aka Road Traffic Injuries.

The Second Decade of Action developed global performance targets on key risk factors
and service delivery mechanisms that were aligned to the UN Sustainable Development
Goals for 2030.to reduce road traffic fatalities and injuries.

6
1.2 Safe System Approach

The “Safe System Approach” started in Australia. It is similar to “Vision Zero” from
Sweden or “Towards Zero Deaths” in the United States. It is different from conventional
safety practices because it is being human-centered. It advocates for extensive vehicle
or roadway design and operational changes rather than relying primarily on driver
and/or vulnerable road users’ behavioral changes. It is concerned with the safety needs
of all road users, including pedestrians, regardless of their age or physical conditions,
bicyclists, motorcyclists, and non-motorized vehicles.

The Safe System emphasizes that road users will make mistakes and that accidents will
happen. Roads must be designed, installed and maintained to ensure road users will
survive these inevitable crashes. No one deserves to die a Needless Death.

The Safe System Approach emphasizes that the human body can only withstand a
certain level of impact force. Since the formula for force is Force = Mass times Velocity
Squared, it is critical that the proper speed (velocity) limits are used, especially in urban
areas where the road is shared by motorists, pedestrians, bicyclists and motorcycles. For
example, pedestrians have been shown to have a 90% chance of survival when struck by
a car travelling at 30 km/h or below, but less than 50% chance of surviving an impact at
45 km/h.

1.3 Forgiving Roads

Road Users make mistakes for a variety of reasons. Approximately 30% of all fatal
accidents are single vehicle, non-pedestrian crashes. Too often drivers and pedestrians
are distracted by cell phones, texting, eating or children. In more than 50% of the single
vehicle, non-pedestrian crashes, the driver and/or pedestrian was drinking. In about
30% of these single vehicle, non-pedestrian crashes, the driver was asleep. Single vehicle
crashes are up to 25 times more likely to occur during nighttime hours.

The Safe System Approach endorses the use of “Forgiving Roads” to reduce the
consequences of run off the road crashes. According to the United States Federal
Highway Administration (FHWA), approximately two-thirds of road fatalities are the
result of run-off-the-road crashes on rural roads.

FHWA recommends that efforts initially are made to keep the vehicle on the road. This
could entail the use of proper vertical and horizontal signage, longitudinal audible tactile
strips (rumble strips), high friction surface technology, especially on curves, safety edge,
which is the application of a 30 degree slope on the pavement on the edge of the road
to allow the errant vehicle to safely re-enter the road, or in the future, autonomous
vehicles.

7
FHWA encourages the removal of rigid objects located close to the road. This may be
a simple cure during the design stage of a project. However, this will become more
difficult once the road has been built.

If the hazard cannot be eliminated, FHWA also recommends the use of countermeasures
that will allow for the safe recovery of the errant vehicle that leaves the road. This can
be accomplished by moving the hazard further away from the road to create recovery
zone or a clear zone along the side of the road. Long mast, cantilevered luminaire
supports are a good example of this solution. The roadside clear zone is the distance
from the edge of the travel lane that should be free of any non-traversable hazard such
as steep slopes or rigid fixed objects. According to the American Association of State
Highway Transportation Officials (AASHTO) Roadside Design Guide, the recommended
clear zone distances are designed to allow approximately 80 to 85 percent of all run-
off-the-road vehicles to recover or come to a safe stop. The width of a clear zone along
the horizontal alignment is dependent on roadside geometry, design speed, radius of
horizontal curve, and the ADT.

If the hazard cannot be eliminated nor moved to create a clear zone, the third option is
to reduce the consequences of a crash with the rigid object. This can be accomplished
in a variety of ways, such as covering a culvert with a grating or by making a roadside
pole more “forgiving.” Poles can be made forgiving through a variety of options, such
as the use of slip bases, or Omni-directional couplings or through the installation of
collapsible poles.

The fourth option to eliminate or reduce the consequences of run off the road crashes
is to shield the hazard with a barrier or a crash cushion. Motorists are protected.
Hazards are shielded, not protected.

1.4 Road Safety Management (RSM)

A comprehensive and integrated RSM is a crucial element in achieving a safe

environment. RSM contributes to set, manage and maintain the objectives regarding
Road Safety. During a longer period of time, functions nearby the road, road usage, etc.
will change, RSM aims to react flexible on these circumstances and to keep low levels of
traffic accidents.

As defined by Muhlrad (2009) RSM is “the Task of preparing and implementing road
safety policies including defining goals and targets, programming, implementing
interventions, evaluation and research”. RSM is the process which effectively implements
road safety policies, including organisation, coordination and management of road
safety interventions meant to reduce fatalities and injuries in traffic1.

1
Bliss & Breen (2009) “Country Guidelines for the Conduct of Road Safety Management Capacity Reviews
and the Specific action of Lead Agency Reforms, Investment Strategies and Safe System Projects” - World
Bank
8
Why do we need RSM?
• Because of the high cost of motorized mobility to society and public health
• Because road traffic injury is largely preventable
• Because achieving better performance in road safety by more effective safety
management
• The shift to safe system – the new performance frontier
• Road safety in a complex multi-sectoral context
• Leadership, ownership, and accountability

The RSM Model derives from New Zealand’s comprehensive 2010 target setting
framework which linked desired results with interventions and related institutional
implementation arrangements (Land Transport Safety Authority, 2000). The New
Zealand framework was adopted by the European Transport Safety Council (Wegman,
2001). It highlighted its results management framework, and it was further elaborated
by the Sunflower Project (Koornstra et al., 2002). The first World Bank guideline
concerning the implementation of the World Report recommendations (Bliss, 2004)
used the framework to introduce prototype safety management capacity review tools.

Safety can be viewed as a management system with three dimensions. World Bank uses
detailed checklist to assess management capacity of a country to deliver road safety.

1.4.1. 1st RSM Dimension - Institutional Management (MGT) Functions

First dimension refers to 7 Institutional Management (MGT) Functions as presented in

the figure above. They are the basis to build road safety management systems and they
should get the highest priority in the road safety planning and political initiatives.
Institutional and management functions apply to all government, civil society and
business structures. In general, they include system-wide policies and programs of
activities to address safety targets. These 7 functions relate to all government, civil
9
society and business organizations that produce interventions and contribute to results.
They are delivered primarily by all the government agencies producing interventions
and also in government partnerships with civil society and business entities. Effective
institutional management is a pre-requisite of successful results-focused intervention.

In managing for improved road safety results, the foremost and pivotal institutional
management function is results focus. All the other institutional management functions
are subordinate to this function and contribute to its achievement. A country’s results
focus can be interpreted as a pragmatic specification of its ambition to improve road
safety and the means agreed to achieve this ambition. In the absence of a clear and
accountable focus on results all other institutional functions and related interventions
lack cohesion and direction and the efficiency and effectiveness of safety initiatives can
be undermined.

Results focus in its ultimate expression concerns a strategic orientation that links all
actual and potential interventions with results, analyzes what can be achieved over time,
and sets out a performance management framework for the delivery of interventions
and their intermediate and final outcomes. It defines the level of safety that a country
wishes to achieve expressed in terms of a vision, goals, objectives and related targets.

Legislation concerns the legal instruments necessary for governance purposes to specify
the legitimate bounds of institutions, in terms of their responsibilities, accountabilities,
interventions and related institutional management functions to achieve the desired
focus on results.

This function ensures that legislative instruments for road safety are well-matched to
the road safety task. Road safety legislation typically addresses land use, road, vehicle,
and user safety standards and rules and compliance with them, as well as post-crash
medical care. A mixture of specialist legislative and technical expertise is needed within
government to develop and consult on legislation promoting enforceable standards and
rules with due consideration to cost, effectiveness, practicality and public acceptability.

Funding and resource allocation seeks to ensure that road safety funding mechanisms
are sufficient and sustainable. As part of this a rational framework for resource allocation
supports the building of strong business cases for road safety investments based on
cost-effectiveness and cost-benefit analyses. To achieve more ambitious performance
targets countries may need to establish new funding sources and mechanisms.

• Promotion concerns the sustained communication of road safety as a core

business for government and society and emphasizes the shared societal
responsibility to support the delivery of the interventions required to achieve the
desired focus on results. This function goes beyond the understanding of
promotion as road safety advertising supporting particular interventions and
addresses the overall level of ambition set by government and society for road
safety performance. Countrywide and sustained communication of road safety as
10
 a core business for Government and society. Emphasising the shared societal
responsibility to support the delivery of the interventions required to achieve the
desired results. In good practice, the ‘promotion’ function is addressed by the
following:
• Promoting a far-reaching road safety vision
• Multi-sectoral promotion of effective intervention and shared responsibility
• Leading by example with in-house road safety policies
• Developing and supporting safety rating programmes
• Encouraging promotion at local level

Monitoring and evaluation concerns the systematic and ongoing measurement of road
safety outputs and outcomes (intermediate and final) and the evaluation of
interventions to achieve the desired focus on results.

Periodic monitoring and evaluation of road safety targets and programs is essential to
assess performance and to allow adjustments to be made. The establishment and
sustainable funding of transport registries for drivers and vehicles, crash injury databases
and periodic survey work to establish performance and exposure data is typically the
responsibility of several different government agencies—transport, police, and health.
In some countries government insurance departments or organizations and university
departments also share responsibility. The organization of independent inspection, audit
and review is also part of this function.

Next institutional management function is research and development and knowledge

transfer which has guided the design and implementation of national strategies that
have sustained reductions in road deaths and injuries, in the face of growing mobility
and exposure to risk. It aims to produce a cadre of international, national and local
professionals who can contribute research-based approaches and knowledge to road
safety policy, programs and public debate. Research and development and knowledge
transfer concerns the systematic and ongoing creation, codification, transfer and
application of knowledge that contributes to the improved efficiency and effectiveness
of the road safety management system to achieve the desired focus on results.

It aims to produce a framework of international, national and local professionals who

can contribute research-based approaches and knowledge to road safety policy,
programs and public debate. Knowledge transfer must be grounded in practice by a
learning by doing process, backed with sufficient targeted investment to overcome the
barriers presented by the evident capacity weaknesses at the global, regional and
country levels. Strong and sustained international cooperation will be required to
mobilize knowledge transfer resources and support services to low and middle-income
countries commensurate with the sheer scale of the global losses arising from escalating
road deaths and serious injuries.

The lead agency plays a dominant role in most of the institutional management
functions for road safety; in others it plays a guiding, encouraging or catalytic role. The
11
lead agency takes responsibility within Government for the development of the national
road safety strategy and its results focus. It is engaged in the delivery of and supported
by strong horizontal intergovernmental coordination arrangements; good vertical
coordination of national, regional and local activity; coordination of the necessary
delivery partnerships between government stakeholders, the professional, non-
governmental and business sectors and Parliamentary groups and committees.

It is engaged in the delivery of and supported by strong horizontal intergovernmental

coordination arrangements; good vertical coordination of national, regional and local
activity; coordination of the necessary delivery partnerships between government
stakeholders, the professional, non-governmental and business sectors and
Parliamentary groups and committees; a comprehensive legislative framework;
sustainable sources of annual funding and a rational framework for resource allocation;
high-level promotion of road safety strategy across Government and society; regular
monitoring and evaluation and strong research and technical support.

1.4.2. 2nd RSM Dimension - Interventions

Intervene refers to intentionally become involved in a difficult situation in order to

improve it or prevent it from getting worse. Interventions are the second level of the
road safety management pyramid and they are designed to achieve results.
Interventions are shaped to achieve the desired focus on results. They address the safe
planning, design, operation and use of the road network, the conditions under which
vehicles and road users can safely use it, and the safe recovery and rehabilitation of
crash victims; they set specific standards and rules to achieve this safety and aim to
secure compliance with them.

1.4.3. 3rd RSM Dimension – Results

Road safety results are expressed in long term goals and interim quantitative targets.
Desired safety performance endorsed by governments at all levels, stakeholders and the
community. They must be achievable with cost-effective interventions. The
measurement of the desired results and their expression as targets in terms of 3 main
results:
• Final outcomes
• Intermediate outcomes
• Outputs

The level of safety targeted is ultimately determined by the quality of the delivered
interventions, which in turn are determined by the quality of institutional management.

Final outcomes can be expressed as a long term vision of the future safety of the road
traffic system (e.g. Vision Zero and Sustainable Safety) and as more short to medium
term targets expressed in terms of social costs, fatalities and serious injuries presented
in absolute terms and also in terms of rates per capita, vehicle and volume of travel.
12
Intermediate outcomes are of value for their contribution to improved final outcomes
and they include average traffic speeds, the proportion of drunk drivers in fatal and
serious injury crashes, seatbelt-wearing rates, helmet wearing rates, the physical
condition or safety ratings of the road network, and the standard or safety ratings of the
vehicle fleet.

Outputs represent physical deliverables or alternatively they can correspond to

milestones showing a specific task has been completed.

Good practice countries set quantitative outcome and intermediate outcome targets to
achieve their desired results focus. They can also set related quantitative output targets
in line with the targeted outcomes as in the New Zealand example:
• The overall target is to reduce the socio-economic costs of road crashes
• To be achieved by meeting the second level of targets, requiring specific
reductions in the numbers of fatalities and serious injuries
• A third level of targets consists of intermediate outcomes (including those related
to speed, drink driving and rates of seat-belt wearing) that are consistent with
the targeted reductions in final outcomes;
• A fourth level of targeting is concerned with institutional delivery outputs such
as the enforcement outputs that are required to achieve the third-level targets

Implementing the recommendations requires capacity building at the global, regional

and country levels, to create the resources and tools necessary to target initiatives on a
scale capable of reducing significantly and sustainably road deaths and injuries in low
and middle-income countries.

At the country level implementation requires an integrated framework that treats the
recommendations as a totality and ensures that institutional strengthening initiatives
are properly sequenced and adjusted to the absorptive and learning capacity of the
country concerned.

Emerging global and regional initiatives aiming to assist the acceleration of knowledge
transfer to low & middle income countries and the scaling up of their road safety
investments must be harmonized. Opportunities must also be taken to combine and
leverage the weight and effectiveness of resources being mobilized to improve the
results being achieved.

1.5 Highway Safety Manual (HSM) - Roadway Safety Management Process

According to Highway Safety Manual (HSM) Roadway Safety Management Process

includes 6 steps:

13
1.5.1. Network Screening
Network screening is a process for reviewing a transportation network to identify and
rank sites from most likely to least likely to realize a reduction in crash frequency with
implementation of a countermeasure.
Those sites identified as most likely to realize a reduction in crash frequency are studied
in more detail to identify crash patterns, contributing factors, and appropriate
countermeasures. Network screening can also be used to formulate and implement a
policy, such as prioritizing the replacement of non-standard guardrail statewide at sites
with a high number of run-off-the-road crashes.

There are five major steps in network screening:

1. Establish Focus: Identify the purpose or intended outcome of the network
screening analysis. This decision will influence data needs, the selection of
performance measures and the screening methods which can be applied.
2. Identify Network and Establish Reference Populations: Specify the type of sites or
facilities being screened (i.e., segments, intersections, at-grade rail crossings) and
identify groupings of similar sites or facilities.
3. Select Performance Measures: There are a variety of performance measures
available to evaluate the potential to reduce crash frequency at a site. In this step
the performance measure is selected as a function of the screening focus and the
data and analytical tools available.
4. Select Screening Method: There are three principle screening methods described
in this chapter (i.e., ranking, sliding window, and peak searching). The advantages
and disadvantages of each are described in order to help identify the most
appropriate method for a given situation.
5. Screen and Evaluate Results: The final step in the process is to conduct the
screening analysis and evaluate results.
14
Step 1 - Establish Focus
The first step in network screening is to establish the focus of the analysis. Network
screening can be conducted and focused on one or both of the following:
• Identifying and ranking sites where improvements have potential to reduce the
number of crashes; and/or,
• Evaluating a network to identify sites with a particular crash type or severity in
order to formulate and implement a policy (e.g., identify sites with a high number
of run-off-the-road crashes to prioritize the replacement of non standard
guardrail statewide).

Step 2 - Identify Network

The second step is to identify the network elements to be screened and group them into
reference populations. Elements that might be considered for the screening process
include: intersections, segments, facilities, ramps, ramp terminals, or at‑grade crossings
which are then grouped by reference population or sites with similar characteristics. By
establishing a reference population, the performance at a particular site is compared to
the expected safety of the reference population, yielding a relative measure of
comparison for determining sites with potential for improvement. Reference
populations can be established based on several characteristics.

Step 3 - Select Performance Measures

Selecting a problem identification methodology to use for the analysis of the network
elements is the third step in the network screening process. The evaluation can be
based on one or multiple problem identification methods. The use of multiple problem
identification methods may provide more certainty in the evaluation, if the same sites
are ranked among the top sites with multiple methods.

Several problem identification methodologies can be used to identify sites for safety
improvements, and the specific problem identification method used varies from agency
to agency. Agencies should use problem identification methods in the network
screening process that suit their specific purpose and available data.

The following are the 13 problem identification methods (Referred to as performance

measures in the HSM.) identified in the Highway Safety Manual (HSM); however, states
are using additional methods not included in the HSM:

1. Average Crash Frequency – Sites are ranked based on the total number of crashes
or by a particular crash severity or type during a given time period. The site with
the highest number of crashes is ranked first.
2. Crash Rate – The crash rate normalizes the crash frequency based on exposure.
3. Equivalent Property Damage Only (EPDO) Average Crash Frequency – Each crash
is weighted based on the crash severity and the equivalent property damage only
crash cost.

15
 4. Relative Severity Index – Average monetary crash costs are assigned to each crash
at a site, and the total average crash cost for a site is compared to the average
crash cost for the reference population.
5. Critical Crash Rate – A critical crash rate or threshold value is calculated for each
site and compared to the observed crash rate. Sites with an observed crash rate
greater than their critical crash rate are flagged for further investigation.
6. Excess Predicted Average Crash Frequency Using Method of Moments – With this
method, the observed crash frequency for a site is adjusted based on the variance
in the crash data and the average crash counts for a site’s reference population,
which is then compared to the average frequency of crashes for the reference
population.
7. Level of Service of Safety (LOSS) – This method compares the observed crash
frequency and/or severity to the mean value predicted for the reference
population using a SPF. The difference between the two values yields a
performance measure that ranges between LOSS I and LOSS IV, with LOSS I
indicating a low potential for crash reduction and LOSS IV indicating a high
potential for crash reduction.
8. Excess Predicted Average Crash Frequency Using SPFs – This method represents
the difference between the observed crash frequency for the site and the
predicted crash frequency based on the SPF with information specific to the site.
9. Probability of Specific Crash Types Exceeding Threshold Proportion – This
method is based on the probability that the long-term proportion of a specific
crash type exceeds a threshold proportion for the site’s reference population.
10. Excess Proportion of Specific Crash Types – This is the difference between the
observed proportion of a specific crash type for a site and the threshold
proportion for the reference population.
11. Expected Average Crash Frequency with EB Adjustment – The expected crash
frequency is calculated using a calibrated SPF, which is then weighted with the
observed crash frequency using the EB method. The EB method accounts for
regression to the mean bias and is discussed in detail in Unit 6.
12. EPDO Average Crash Frequency with EB Adjustment – This method combines the
expected crash frequency method with EB adjustments and the EPDO crash
frequency method. The expected crash frequency is calculated using a calibrated
SPF and weighted with the observed crash frequency using EB, which is then
weighted based on crash severity and the equivalent property damage only cost.
13. Excess Expected Average Crash Frequency with EB Adjustment – The expected
crash frequency determined from a SPF is weighted with the observed crash
frequency using the EB method. The resulting weighted crash frequency is then
compared to the expected crash frequency using the SPF to determine the
difference between the two values.

Step 4 - Select Screening Method

Simple Ranking

16
As the name suggests, simple ranking is the simplest of the three screening methods
and may be applicable for roadway segments, nodes (intersections, at‑grade rail
crossings), or facilities.

Sliding Window
With the sliding window method, the value of the problem identification methodology
selected is calculated for a specified segment length (e.g., 0.3 miles), and the segment is
moved by a specified incremental distance (e.g., 0.1 miles) and calculated for the next
segment across the entire segment.

Peak Searching
Similar to the sliding window method, the peak-searching method subdivides the
individual roadway segments into windows of similar length; however, the peak-
searching method is slightly more meticulous. The roadway is first subdivided into 0.1-
mile windows; with the exception of the last window which may overlap with the
previous window.

Step 5 - Screen and Evaluate Results

Finally, the selected problem identification method(s) and screening method(s) are
applied to the study network. The result will be a list of sites identified with potential
for safety improvement with the sites most likely to benefit at the top of the list. These
sites should be studied further to determine the most effective countermeasures.
Applying multiple problem identification methods to the same data set can improve the
certainty of the site identification process. Sites listed at the top of the list based on
multiple problem identification methods should be investigated further.

1.5.2. Diagnosis
Diagnosis is the process of further investigating the sites identified in network screening
to identify specific safety concerns and contributing factors at potential project sites.
Basic diagnosis involves a traditional engineering study, including an office-based
review of crash data, such as crash report documents, summary statistics, and collision
diagrams. Intermediate diagnosis capability builds on basic analysis with a
multidisciplinary team field study and road safety audit to observe site conditions and
road user behaviors to identify underlying contributing factors that may not be obvious
from the crash data alone. Advanced-level diagnosis incorporates statistical methods
to focus the diagnosis.

1.5.3. Countermeasure Selection

Following diagnosis, agencies select potential treatments (engineering-related or other)
for each site of interest. Appropriate countermeasures target the underlying
contributing factors identified in the diagnosis. Data used in countermeasure selection
include crash data, budget and other program data, stakeholder input, and project-
related factors (e.g., costs and environmental and right-of-way needs). Intermediate
practice involves a list of common or preferred countermeasures by facility type along
17
with preselected CMFs for those treatments. The data needs for countermeasure
selection focus on countermeasure feasibility (e.g., what improvements would work well
at the site) and countermeasure effectiveness (i.e., inputs for specific CMFs).

1.5.4. Economic Appraisal

Economic appraisal compares the relative benefits and costs of proposed alternatives or
treatments. Analysts usually express benefits as the dollar value of the estimated lives
saved as well as other injuries and property damage avoided (e.g., based on a CMF).
Costs may include construction costs, maintenance costs, right-of-way acquisitions, and
other costs incurred due to implementing the treatment. Common economic measures
include benefit-cost (B/C) ratio, net present value, and cost-effectiveness. Economic
analysis also allows project managers and designers to develop project budgets to
maintain economic justification for safety improvement with known countermeasures
and desired B/C ratio.

1.5.5. Project Prioritization

Project prioritization results in an optimized portfolio of projects for a given budget.
Agencies have different processes for prioritizing projects. Even within safety programs,
safety benefits are usually not the only decision factor. Rankings based in part on project
costs, safety benefits, project risk, project readiness, agency goals and performance
targets, public perception and acceptance, and political influence, among others, can
maximize the B/C ratio of the overall program for a given available budget.

1.5.6. Safety Effectiveness Evaluation

Agencies conduct safety effectiveness evaluation to quantify the actual safety benefits
achieved from project implementation (e.g. a reduction in fatal and serious injury left-
turn crashes). The evaluation provides critical feedback for future planning. Agencies can
and should conduct evaluations at the project, countermeasure, and program levels. The
objective of project-level evaluation is to quantify the safety effectiveness of individual
projects.

18
2. Understanding Diagnostic Tools and Collision Patterns

Crashes are the result of different factors interacting:

Background factors
• Training
• Safety Culture
• Vehicle fleet characteristics
Crash Specific Factors
• Type of road users
• Location where crash occurs
• Trajectories of the road users

Reason proposed what is referred to as the “Swiss Cheese Model” of system failure.
Every step in a process has the potential for failure, to varying degrees. The ideal system
is analogous to a stack of slices of Swiss cheese. Consider the holes to be opportunities
for a process to fail, and each of the slices as “defensive layers” in the process. An error
may allow a problem to pass through a hole in one layer, but in the next layer the holes
are in different places, and the problem should be caught. Each layer is a defense against
potential error impacting the outcome.

For a catastrophic error to occur, the holes need to align for each step in the process
allowing all defenses to be defeated and resulting in an error. If the layers are set up
with all the holes lined up, this is an inherently flawed system that will allow a problem
at the beginning to progress all the way through to adversely affect the outcome. Each
slice of cheese is an opportunity to stop an error. The more defenses you put up, the
better. Also the fewer the holes and the smaller the holes, the more likely you are to
catch/stop errors that may occur.

There is need for data describing the location of the crashes and the role of the different
elements in the crash. Starting with the location of the crashes It is important to consider
what the boundaries of a crash location are. There needs to be a defined cut-off point,
such as between crashes that occur at an intersection and crashes that are considered
19
‘mid-block’. It may be necessary to look beyond these defined boundaries when
analysing crash data. For example, crashes within 10 metres on the approach roads to
an intersection may be considered as located at the intersection; however, it may be of
value to look beyond this boundary for other crashes that may be related to an
intersection (e.g. 100 metres). The crash location is also generally identified as the point
at which an impact occurred. However, this may only be the end point of a sequence of
events. Factors relating to the cause of the crash may have started earlier on the
roadway.

Crash locations can sometimes be poorly or inaccurately defined. In built-up areas, the
common practice is to measure the distance from the nearest intersection, junction or
landmark using some distance measuring device. Other common systems are the Linear
Referencing System and Link-Node System. These too rely on road names or reliable
kilometer post markers along roads. In many countries hectometer points (or milepost)
on road are a part of infrastructure for a very long time. Now more and more common
is to use Global Positioning Systems (GPS). Gathered in this way latitude and longitude
coordinates may show us the exact location more precisely.

There is generally not enough funding to treat all identified crash locations.
The selection process varies depending on the aim of the project and the types of
actions that may be considered, and include:
• site action – treating a specific site or short length of road that has a
concentration of crashes;
• route action – investigating crashes along a section of road, looking for common
crash characteristics along the route, but also individual problematic sites;
• area action – investigating crashes throughout an area, where the main issues to
be addressed may be on a broader scale, such as traffic management and
network problems (e.g. pedestrian safety may be a recurring theme) ;
• mass action – this looks for common crash characteristics over a larger area, such
as delineation problems or vehicles running off the road.

Diagnosing the contributing factors to crashes is the foundation for selecting an

effective solution to a safety problem. To properly understand the problem, one must
consider that:
• A crash is the result of a sequence of events (not a single cause).
• Each event or circumstance is linked to a component of the Safe System – the
persons involved, the vehicle, or the road environment.
• Each event is strongly influenced by the preceding event or circumstance.
Diagnosis of the contributing factors at a crash location is a four-step process:
• Gather the relevant information about the site, such as crash data, volumes, and
the history of the network or location in terms of land use or physical layout
changes.
• Analyse the crash data for the location by looking for common crash types or
factors, particularly those that are happening in groups.

20
 • Inspect the site from the perspective of the road user, and closely examine the
site’s physical features and layout.
• Based on the notes and findings from the above steps, draw conclusions about
the likely contributing factors to the crash groupings (of similar type and/or
location).

Crash data is very important, but for all incidents there are very sparse data and for
investigated and researched crashes there is full investigation detail and peer reviewed
detailed data.

High Risk Sites Identification is one of the easiest approaches to diagnosing problems
even with limited data. It is always based on geographic location of the crashes. It
requires the analyst to define their high risk sites criteria: Number of crashes and types
of crashes (pedestrian, fatal, side impact, etc).
We can analyze the data with highly advance statistical tools but let’s stick with the main
concepts and leave advanced statistics for another course. The analysis of crash data is
best approached by determining crash rates. Crash rates are derived from the local,
regional, or national data primarily police records. The most common approach is a
crash rate CR
# Crashes
𝐶𝑅 =
𝑅𝑒𝑓𝑒𝑟𝑒𝑛𝑐𝑒
where:
# Crashes can be All crashes, Fatal crashes, intersection crashes
Reference = #Vehicles/day, population, #vehicles entering intersection per day

Using our knowledge of Crash Rate and crash information we can proceed with Network
screening . Many tools exist to automate the process but all are based on common
objective: “What parts of my road network have an over representation of crashes”.

The final concept for Network Screening and Countermeasure selection refers to a
practical tool that opens up our possibility to analyze safety issues and identify
countermeasures and their benefits is the Safety Performance Function

𝑆𝑃𝐹 = (𝑐𝑟𝑎𝑠ℎ𝑒𝑠, 𝐴𝐴𝐷𝑇, 𝑙𝑒𝑛𝑔𝑡ℎ, 𝑒𝑡𝑐)

The SPF is a mathematical expression for a network that allows comparison between
local, regional, and national data. SPF provide an important baseline for comparison,
most Network Screening tools use SPFs as the reference for identifying over-
represented road elements. SPFs are a requirement for the application of Crash
Modification Factors that help guide countermeasure selection and evaluation.

21
3. Focus on Road Safety Audits

3.1 Road Safety Audits as part of Management of Road Infrastructure Safety

Road infrastructure safety may be critical for road safety enhancement, especially in
emerging economies. Traditional «reactive» approach to road safety (e.g. high risk site
management) is becoming ineffective in more advanced countries. Moving towards a
Safe System approach where the Road Administration has responsibility for the safety
of the infrastructure. The Road Infrastructure Safety Management procedures include:
• Road Safety Impact Assessment (RIA)
• Road safety measures Efficiency Assessment Tools (EAT)
• Road Safety Audit (RSA)
• Network Operation (NO)
• Road Safety Performance Indicators (SPIs)
• Network Safety Ranking (NSR)
• Road Assessment Program (RAP)
• Road Safety Inspection (RSI)
• High Risk Sites (HRS)
• In-depth Investigation

The management of Road Infrastructure Safety as described in IRTAD/ITF Management

of Road Infrastructure Safety2 includes 6 major categories which are conducted with the
tools and actions presented in the figure below:
1. Planning & - Road safety Impact Assessment
- Efficiency Assessment Tools
Design
- Road Safety Audit

2. Construction &
- Road Safety Audit
Pre-opening

3. Normal - Road Network Operation

- Safety Performance Indicators
Operation - Network Safety Ranking
Road Infrastructure
Safety
Management 4. Maintenance & - Road Network Operation
- Road Safety Inspection
Renewal - Road Assessment Program

- High Risk Sites

5. Error correction, - Road Safety Inspection
Hazard elimination - Road Assessment Program
- In-depth investigation

6. Major upgrading - Road safety Impact Assessment

& Renewal - Efficiency Assessment Tools

The key messages derived are:

• Road Authorities are key players for improving road safety
o In the USA, road conditions contribute in 53% of all road deaths and 38% of all
injuries.

2
https://www.itf-oecd.org/sites/default/files/docs/15irtadsafetymanagement.pdf
22
 o There are substantial opportunities, programs, and tools to improve safety of
road infrastructure.
o Road safety measures should be defined and implemented according to
preference and circumstances of each country.
• Road Infrastructure Safety Management procedures are effective and efficient
o RIAs and EATs provide better information to policy makers in order to make
better decisions.
o RSAs and RSIs have shown positive cost-benefit-ratios, up to 99:1.
o Regular use of RAP has shown improvements in Spain, UK and Sweden.
o HRS (and potentially NSR) approach results in an 18% reduction in casualties.
• Several tools supporting road infrastructure safety management are already
available.
o International guidelines and manuals are nowadays available.
o National guidelines and software are available in many countries (e.g. Australia,
Austria, Canada, Finland, Germany, Greece, Ireland, Japan, the Netherlands,
Norway, Spain, Sweden, UK, USA, etc.)
• Each country has specific needs and has to cope with specific barriers to the
implementation of RISM as different conditions exist.
o There is no procedure better than other, and it is not the case that a more
extensive use of these tools automatically ensures a superior road safety
performance.
o Procedures should be adapted to the specific conditions of each country.
o Low and medium income countries can focus on low budget procedures.
• Road design standards cannot guarantee road safety in all conditions.
o Designers may be encouraged to reduce construction cost and are not aware
of future traffic conditions.
o The relationship between design standards and road safety is unclear and the
level of safety designed into roads is unpremeditated (Hauer, 1999).
o Design standards are important to keep up with nominal safety. Substantive
safety must be considered in design process to care for safety in principle.
• Road infrastructure should be improved with the development of self-explaining
roads to guide drivers to adopt appropriate behaviours.
o Evidence of increased safety after the implementation of the self-explaining
roads.
o In general, a more pro-active approach to road infrastructure design and
management is desired (e.g. Vision Zero, Sustainable Safety, Safe System, Safety
Conscious Planning).
• Road safety performance monitoring helps to achieve safety target of road
authorities.
o A target should be defined and progress toward the safety target should be
monitored.
o Fundamental road safety performance indicators can be the number of road
accidents or fatalities per unit distance or unit number of vehicles or vehicle
travelled.

23
 o Monitoring can be effective if the exact location of accidents or x, y coordinates
are available.

3.2 Road Safety Audits

A variety of countermeasures are available to make roads safe. One of the most effective
countermeasures is Road Safety Audit (RSA). Pillar 2 of the UNRSC Decade of Action
strongly recommends the use of RSA on roads around the world. RSA is a formal safety
performance evaluation of an existing or future road or intersection by an independent,
experienced & multidisciplinary team.

The concept of RSA can be traced back to the quality programs developed by British
Rail. Road Safety Audits were introduced in the United Kingdom with the Road Traffic
Act 1988: › “… in constructing new roads, (local authorities) must take such measures as
appear to the authority to be appropriate to reduce the possibilities of such accidents
when the road come into use.” In 1991 Safety Audits were mandatory on trunk roads
and motorway schemes in the United Kingdom.

Some European countries followed the United Kingdom’s lead and mandated that all
road design schemes be subjected to a RSA. The concept of RSA became more
widespread in 2008 when the EU Directive 2008/96/EC (which was revised recently by
Directive2019/1936) on Road Infrastructure Safety Management was issued. This
directive required that all Member States undertake Road Safety Audits, Road Safety
Impact Assessments, Road Safety Inspections, and Network Safety Ranking/ ranking of
high concentration accident sections on the Trans European Road Network (TERN.)

It further required that Safety Auditors have suitable training and experience based on
a Certificate of Competence.

RSA were required on new designs of roads while Member States were told to carry out
Road Safety Inspections (RSI) on existing roads in order to identify the road safety
related features and prevent crashes. These inspections were to be performed
periodically and by a competent entity. Member States were also encouraged to apply
this Directive to other national roads, which were not included in the Trans-European
Road Network.

Some other non-European countries, including Australia, New Zealand, Chile, United
Arab Emirates, South Africa and the United States implemented programs to use RSA
and RSI on their roads.

In 2014 the concept of RSA received a huge boost when the Multilateral Development
Banks issued their “Road Safety Guidelines-Road Safety Initiative” that stated, “Road
Safety Audits (RSA) from feasibility level to opening of the completed project, carried
out by independent, certified road safety auditors and on post-opening of the project.”
It further encouraged the Inclusion of road safety specialists on the project
24
planning/design team and the Application of recognized design standards which are in
line with the Safe System Approach. The Guidelines also stated “Evaluations of the safety
ratings of existing road networks and strengthening of capacity for periodic Road Safety
Inspections (RSI) of the road network (hot spots, spatial analysis in urban areas and
hazardous road sections.) “

The Multilateral Development Banks were frustrated that so many roads that they had
financed were being built as unsafe. This Initiative was designed to require road
authorities who wanted loans or grants from the banks to use best practices as well as
state of the art concepts and technologies to meet global road safety requirements. In
2018 in Leipzig, Germany the Multilateral Development Banks further reinforced the
requirement need for RSA by stating “Depending on project scope and stage, we expect
road authorities to apply RSA on five stages:
• Initial design/feasibility study/Road Safety Impact Assessment
• Draft design
• Detailed design
• Before completion
• Early operation

Because some road authorities apparently did not totally understand the concept of
RSA, and would conduct the audit, but not implement the recommended changes, the
Banks noted: “Road Safety Audits must be subject to decision-making by the responsible
authority, and this should be documented.
1) What problems were identified?
2) What solutions are suggested by the RSA team?
3) Can the responsible authority accept them or not?
4) If not – why not? What alternative measures will be applied instead?”

To avoid any possible confusion, the Multilateral Development Banks further stated “We
expect personnel with road safety experience to be responsible for road safety in the
projects we finance and to be a part of the implementing agency’s Project Management
team, Design team, Supervision team and Contractor’s team.”

In its October 2019 Good Practice Note on Road Safety, the World Bank clearly stated:
“10. The Borrower will identify, evaluate and monitor the potential traffic and road safety
risks to workers, affected communities and road users throughout the project life-cycle
and, where appropriate, will develop measures and plans to address them. The Borrower
will incorporate technically and financially feasible road safety measures into the project
design to prevent and mitigate potential road safety risks to road users and affected
communities”.

11. Where appropriate, the Borrower will undertake a road safety assessment for each
phase of the project, and will monitor incidents and accidents, and prepare regular
reports of such monitoring. The Borrower will use the reports to identify negative safety
issues, and establish and implement measures to resolve them.”
25
Road authorities around the world were being forced to understand and use RSA if they
wanted their projects financed through a Multilateral Development Bank. This greatly
increased the awareness and the use of road safety audits. The Term “Road Safety Audit”
means different things to different people. In Europe, Road Safety Audits only exist on
new road projects. The Road Safety Audit Stages in Europe include an Initial planning
stage (Road Safety Impact Assessments) followed by
STAGE 1) Completion of Preliminary Design (Road Safety Audits)
STAGE 2) Completion of Detailed Design (Road Safety Audits)
STAGE 3) Completion of construction prior to road open to public (Road Safety Audits)

STAGE 4) Monitoring - Periodic checks of the existing network after 12 months and then
36 months

When an existing road or intersection is evaluated for safety in Europe, it is referred to

as a Road Safety Inspection (RSI). The RSI would be conducted by a qualified Road Safety
Auditor.

In the United States and many other countries, the term “Road Safety Audit” is used for
safety evaluations of design projects as well as existing roads or intersections.

Regardless of the terminology used, it is critical that the Road Safety Auditors have an
influence during the Planning, Preliminary Design and Final Design Stages (Pre-
Construction Road Safety Audits Phase), Work Zone, Construction, Pre-Opening Stages
(Construction Phase), and on existing roads (Post Construction Phase).

Development Project RSAs may be conducted on industrial, commercial or residential

land use developmental projects that may have an impact on the characteristics of
existing adjacent roads. These developmental projects may have the potential to change
traffic volumes, traffic patterns, and vehicle mix or road user perception of the road.

A road safety auditor will use a “Prompt List” to consider potential issues or problems.
These prompt lists will change depending on the Project Phase. Note that these are
called “Prompt Lists” and not “Check Lists.” The Prompt Lists are intended to make the
road safety auditor consider a variety of issues depending on the site conditions. It is
not all-inclusive. This is one of the reasons experience is so important for a Road Safety
Audit team. Once it has been determined that a project should be considered for a RSA
or a RSI, a typical RSA or RSI will consist of the following eight (8) steps:

STEP 1: Identify the Project or Road/Intersection in Service to be Audited (Project

Owner)

It is important to start the RSA process as early as possible. It is far easier to implement
road safety improvements during the design stage than after the road has been totally
constructed. In Step 1, a design project may be identified for a RSA because it is required
26
for financing from the bank. An existing site may be chosen due to the high number of
crashes, or recent high profile crashes or due to changed traffic characteristics cause by
an expansion development such as a new industrial park, or school or commercial mall.

In countries where detailed crash and traffic data are available, Risk Mapping may be
used to identify a Road Safety Audit project. Risk Mapping can help to identify high-
risk routes rather than “hot spots” or “cluster sites” for fatal and serious injury crashes.

Another approach to determine what types of projects should be considered would be

through the application of nominal and substantive safety concepts. Nominal Safety
refers to compliance with standards and Substantive Safety refers to actual crash
performance.

STEP 2: Select a RSA Team (Project Owner)

When selecting a RSA team in Step 2, it is extremely important to have a team that is
independent of the Project Owner/Designer. Independence ensures a fresh set of eyes
and an unbiased assessment. Creating an experienced and multi-disciplinary team
ensures that the problem will be viewed from multiple perspectives and draws upon
expertise from several disciplines. The core team, which typically will include a highly
qualified Team Leader as well as Team Member and Observers from a variety of
disciplines such as Safety, Design, Operations, Enforcement as well as Special Needs
such as Pedestrian, Bicycle, Human Factors, Motorcyclists, Elderly, and Disabled.

The Best Practice is to have the smallest team that provides all of the necessary
knowledge and experience for the project. This could be as few as two Members and
should be no more than six Members. The optimum number for a RSA Team is four to
six Members. In Europe. A RSA team typically consists of two people. Both are qualified
Road Safety Auditors and one is named the RSA Team Leader. These Road Safety
Auditors will solicit comments and support from other road safety specialists as needed
during certain stages. In Belgium, three Road Safety Auditors are used and one is a
Motorcycle Safety Specialist.

STEP 3: Conduct a Pre-Audit Meeting to Review Project Information (RSA Team)

The Pre-Audit Meeting may be the first time some of the team members meet each
other. The Road Safety Audit Team Leader will coordinate this meeting.
The Road Safety Audit Team Leader is typically the single point of contact with the
project owner regarding the RSA. He or she is responsible for selection of the other team
members. A good Team Leader should be able to co-ordinate tasks and liaise with
others within deadlines. He or she also should be able to write clear, concise reports,
and visualize schemes from plans taking into account the perspective from ALL road
users. An effective Team Leader has good attention to detail, understands complex
schemes and knows where to look up standards and control data.

27
The Team Leader will set the agenda for the Pre-Audit Meeting. This agenda will include,
but is not limited to introductions to the Team Members, an explanation of project
objectives and project design. A description of the RSA process with a proposed
schedule will be presented at the Pre-Audit Meeting.
All Team Members should introduce themselves and explain their connection with the
project. For example, in a pre-construction audit, members of the team should identify
their role in the project, such as project manager or drainage engineer.

The Team Leader will provide the team with all necessary data (plans, photos, accident
history, environmental documents, traffic volumes, design criteria, aerial photos, design
drawings, etc.) to allow them to make good decisions. If available, the Team Leader
should present a two to three year crash history of the project to give the Team a clear
understanding of the situation.

STEP 4: Analyze Drawings/Perform Field Observations Under Various Conditions

(RSA Team)

If the audit is for a new project in the Pre-Construction Phase, the Team will review all
available plans or drawings. The Team must recognize that they are only concerned
with the safety issues of the design. This is NOT an opportunity to redesign the project.
The Team also must understand the RSA is not a simple standards check for adherence
to design guidelines. The RSA goal is to ensure state of the art technologies and
concepts, as well as best global road safety practices are included in the project. Field
reviews for Pre-Construction RSAs are conducted to observe the ambient conditions in
which the new facility will operate.
For the field review, the Team Leader will arrange transportation and designate a
secretary and a photographer. Ideally the Team should take with them to the site at
least drawings and aerial photographs as well as a camera and a video camera, a
measuring wheel, a stopwatch, high-visibility vests and emergency lights for the vehicle.

Once on site, the Team should drive the site as well as walk the site. All Team Members
should drive in the same vehicle to allow them to all hear everyone’s comments
regarding the site. Field reviews for post-construction RSAs are conducted to observe
conditions “on the ground” that create safety hazards. Team Members should observe
road user characteristics: For example, what are typical speeds? What is the typical
traffic mix, including heavy vehicles, farm equipment, cyclists? Does traffic tend to
queue at certain times of the day or in certain lanes?

Team Members should observe surrounding land uses: What are the existing
developments contributing traffic to the audit site? Are there any driveways that might
affect the planned roadway? What are the typical traffic patterns associated with the
adjacent land uses (for example, weekend traffic near a sports field)?

28
Observe link points to the adjacent transportation network: For example, are there at-
grade railway crossings in the vicinity of the audit site that could delay traffic? Are
interchange ramps close to the site?

There is no substitute for getting out of the car and walking the audit site, especially if
traffic at the audit site includes (or will include) pedestrians. The audit team can more
closely observe roadside and pavement conditions, as well as pedestrians’ perspective.
Time spent walking around or observing the audit site also gives greater insight into
driver and pedestrian behavior, although the presence of the audit team may influence
this behavior. Team Members should talk to road users, especially pedestrians who are
disabled to better understand the current conditions.

While on site, use the prompt list to review the location. Check for sight distance
obstructions. Sight distance obstructions typically encountered include buildings, trees,
embankments, and other landscaping features. It is not unusual to find street signs that
are obstructed by overgrown vegetation. Team Members should also look for roadside
hazards including fixed objects, deep and/or steep drainage ditches, or unprotected
barrier ends within the roadside clear zone. Another issue to look for would be driveway
issues: Do cars turning into and out of driveways interfere with through traffic?
Interference will be particularly unwelcome close to an intersection, where driver
workload is already high, or around a curve where approaching drivers cannot see it.
Team Members should observe pavement conditions, including potholes and friction
conditions. Team Members should keep referring to the prompt lists.

The Field Review should be conducted on peak and off-peak traffic periods. It also
should be conducted in the day and the night. A review of the crash history data will
help to identify the number of incidents occurring at night. Single vehicle crashes are
up to 25 times more likely to occur during nighttime hours. Ideally the Field Review
should be conducted in wet and dry conditions. Too often horizontal lines are not clearly
visible in the rain. If the pavement is too slick, the rain can increase the number of run
off the road incidents. Rain can also increase the likelihood of

STEP 5: Conduct Audit Analysis and Prepare a Report of the Findings (RSA Team)

The Team will review the information gathered during STEP 4 and will prioritize the
issues discovered during the Field Review. Many highway engineers are used to thinking
about safety in terms of adherence to design criteria and standards used by the road
authority. This is referred to as ‘nominal safety’. The performance of a road section or
intersection as determined by crash frequency and severity, is referred to as ‘substantive’
or quantitative safety. The term “nominal safety” was coined by Dr. Ezra Hauer to
describe a road situation based solely on its adherence to design standards and
practices. A road may be considered “nominally safe” if it meets the minimum standard
of care and is current with respect to published standards and guidelines. One of the

29
first things a Road Safety Auditor must realize is that the Audit is not just a review of the
adherence to the local standards.

The term substantive safety, also known as quantitative safety), is the actual or expected
performance of a road sector or an intersection in terms of crash frequency and severity.
For example, the standards may require a 4 meter minimum clear zone at a site. If the
site meets this minimum, then the site is nominally safe. However, crash data reveals
that this nominal width was not adequate at this site due to the number of crashes by
vehicles not being stopped within the clear zone. This would mean the site is
substantively unsafe and should be considered for a countermeasure.

The Team must prioritize the issues discovered during the Field Review. A variety of
methods can be used to prioritize the issues including the Haddon Method, which was
developed by William Haddon in 1970. This matrix looks at factors related to personal
attributes, vector or agent attributes and environmental attributes; before, during and
after an injury or death. By utilizing this framework, one can then think about evaluating
the relative importance of different factors and design interventions.

Once the priority level has been determined, the RSA Team will suggest
countermeasures to improve the level of safety on the selected site. The Team will make
Short -Range, Mid -Range and Long- Range Recommendations. As an example, at a
curve on a rural road, the RSA Team may make Short- Term Recommendations that
could be implemented quickly at a low cost. These could include lower speeds, better
Enforcement, better Education for road users, and the use of Engineering
countermeasures such as advanced signage, the use of dynamic signs in the curve,
vegetation control to improve sight distance, better vertical signage, pavement
markings, maintenance, high friction surface technology, etc.

Mid-range measures which will be more expensive and will take more time to implement
to make the rural curve safer may include curve widening, repaving, and will also still
include Enforcement & Education.

Long-range measures may include flattening a curve or modifying a roadway’s vertical

alignment. Once again Enforcement & Education should continue to be recommended
by the RSA Team.

STEP 6: Present Audit Findings to Project Owner/Design Team (RSA Team)

The RSA Team must present its findings and recommendations to the Project Owner. At
this meeting, the RSA Team will present the safety concerns and recommendations in a
written report. This meeting is an opportunity to clarify findings and suggestions and
for the RSA Team to assist the project owner to make best choices.

The RSA Team Report will identify the scope of the project. It will introduce the RSA
Team Members and explain when the Field Reviews were conducted, including time of
30
day and weather conditions, if appropriate. The Report will summarize the project,
identify and prioritize safety concerns. The RSA Team must be careful not to use terms
like “unsafe,” “sub-standard,” or “unacceptable.” A lawyer or newspaper reported could
use these terms against the Project Owner if there is an incident in the future. It is better
to use terms like, “Issues,” “Examples,” and “Recommendations.”

STEP 7: Project Owner/Design Team Prepares a Formal Response (Project Owner)

The “MISSING LINK” in many RSAs is the Formal Response by the Project Owner/Design
Team. By definition, a RSA is a “FORMAL safety performance evaluation of an existing
or future road or intersection by an independent, experienced & multidisciplinary team.”
This Response is a part of the Formal Contract signed by the Project Owner. The
Response should be prepared by the local road agency (with possible input from
designer). For each audit issue, the Report should identify what action will (or will not)
be taken with a brief explanation. This Response from the Project Owner becomes a
part of the official project record.

STEP 8: Incorporate Findings into the Project (Project Owner)

In a Pre-construction RSAs, the changes may be made to the design drawings. In Post-
Construction RSAs, the improvements may be incorporated in operating budgets or
maintenance programs. Implementation may depend on policy, staffing, and/or
funding. Typically, a Project Owner will incorporate findings based on ranking and
feasibility. Some improvements can be implemented relatively quickly (Short-Range),
and some may require more time and budget (Long Range).

31
4. System-wide Management supporting Countermeasure Selection

4.1 General
Network screening tools can provide more details as we expand out analysis approach.
We can consider 3 different ways to assess safety performance of a road network:
1. Observed Crash Performance
2. Predicted Crash Performance
3. Expected Crash Performance

The benefits from using observed crashes are that they use available data usually already
available, are relatively simple tools (visualization tools, Excel) and reflect the historical
performance of the region/area/site. Crash rates are susceptible to “regression to the
mean”. Crashes are a stochastic process and the variations may mean we can sample at
a temporary high or low value. The basic drawbacks of using observed crashes include
the fact that they does not account for fluctuations (regression to the mean), they cannot
account for changes in traffic volume, they may not account for weather and finally they
are dependent on the supply of crash data.

Although observed crash performance is only retrospective with limited forecasting

ability, predictive crash performance is possible with a model of crash performance. A
model of predicted crashes with a Safety Performance Function provides ability to
predict crashes based on relationships between crashes and road features. The general
expression of SPF for a road segment is the following:

The general expression of SPF for intersections is the following:

32
SPFs allow us to identify local areas with overrepresentation of crashes. It is similar to
hot spot Identification yet enhanced with local reference information. Predictive power
of SPF means we may be able to predict future performance due to changes in the
systems (more traffic, introduction of a new feature). Key benefits for predicted crash
performance is that it counts for fluctuations in traffic. On the other side it requires
extensive data to model all SPFs of interest and it loses specifics of types of crash

With expected number of crashes we can use a statistical correction for predicted
crashes using the observed crash history (Empirical Bayes Method)

𝐶𝑟𝑎𝑠ℎ𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 = 𝐶𝑟𝑎𝑠ℎ𝑝𝑟𝑒𝑑𝑖𝑐𝑡𝑒𝑑 + (1 − 𝑤)𝐶𝑟𝑎𝑠ℎ𝑜𝑏𝑠𝑒𝑟𝑣𝑒𝑑

1
𝑤=
1 + 𝑘 𝐶𝑟𝑎𝑠ℎ𝑝𝑟𝑒𝑑𝑖𝑐𝑡𝑒𝑑

Where k is the quality of the SPF and should be reported by the SPF

The key benefits of using Expected Crash performance is that it combines observed
information with predictive tools and can address better the amount of traffic, changes
in road features and general time trends. ON the other side it requires extensive data to
model all SPFs of interest, it loses specifics of types of crash and finally it does not
account for the changes from the original road characteristics if we want to predict a
countermeasure performance.

4.2 An enhanced systemic approach to road safety

According to Grembek et al, 2019 the systemic approach is a desirable methodology to

transition from existing practices in road safety to approaches such as safe systems that
fundamentally change the way road systems are conceived and designed to ensure that
none of their users are severely injured.
The systemic approach stands at a midpoint between traditional hotspot and corridor
approaches, and emerging fully proactive approaches such as safe systems, which is why
it constitutes an exceptional opportunity for road safety agencies.

The systemic approach is a flexible, data-driven methodology that aims to identify

recurring safety concerns within a road network, by identifying the crash profiles that
are associated with certain roadway features. The analysis takes the form of a
transparent systemic crash matrix that shows what types of crashes occur on what types
of facilities, with rows representing crash characteristics and columns corresponding to
facility types.

The framework set by the systemic approach allows agencies with varying degrees of
data availability to implement it - regardless of the level of performance their data
management systems - and with different safety priorities.

33
It is not specific to a particular database and lays out the main steps to be undertaken
by any agency that wishes to conduct systemic improvements on its road network.
Adequate data collection on past collisions and detailed infrastructure features, followed
by well-documented data cleaning, ensures that the systemic matrix is generated with
the right information.

Developing a systemic crash matrix is only the first step on the way to systemic road
safety. Once systemic hotspots are identified, engineering countermeasures can be
selected to address their crash profiles across the corresponding facility types.

34
5. User-Specific Interventions & Countermeasure Resources

5.1 Haddon's ten strategies for road traffic injury prevention

1. Prevent the initial aggregation of the particular energy form.

This is usually done by discouraging the use of vehicles and designs that are
particularly hazardous and by encouraging alternative travel modes.
2. Reduce the amount of energy aggregated.
Examples are the setting of speed limits on roads, making engines which are not
very powerful, and installing speed limiters on existing vehicles.
3. Prevent the inappropriate release of energy.
This can be achieved by designing vehicles and the environment such that road
users do not make mistakes easily, for example, through the use of better brakes,
safer intersections and roundabouts, and skid resistant roads.
4. Alter the rate or spatial distribution of release of the energy from its source.
Making pointed and sharp surfaces rounded and flatter distributes the forces
over a larger area during an impact and thus reduces stresses on the body.
Vehicles with appropriate crashworthiness criteria will transfer less energy to
occupants.
5. Separate susceptible structures from the energy being released by means of
space or time.
Separate lanes for bicycles and pedestrians reduce the probability of the riders
or walkers being hit by motor vehicles. Daytime curfews for trucks in cities reduce
the number of crashes involving pedestrians.
6. Interpose a material barrier to separate the released energy from susceptible
structures.
Examples are physical road dividers on highways, and bollards and fences
between pedestrian paths and roads.
7. Modify contact surfaces or basic structures that can be impacted.
Padded interiors and absence of sharp objects prevent injury. Examples include
softer car and bus fronts, breakaway poles on highways, and use of helmets by
two-wheeler riders.
8. Strengthen human beings who are susceptible to damage by the energy transfer.
An example is treatment for osteoporosis of older road users.
9. Quickly detect and evaluate damage, and prevent its continuation or extension.
Damage can be limited by efficient systems for extraction of victims from vehicles,
emergency care, and management of crash sites.
10. Carry out all necessary measures between the emergency period immediately
following damage and ultimate stabilization of the process.
Such measures include intermediate and long-term repair and rehabilitation.

5.2 What specific interventions can be implemented?

There is no standard package of interventions suitable for all contexts and countries.
Interventions proven in one setting may not easily be transferable elsewhere, and will
require careful adaptation and evaluation. Where effective interventions are altogether
35
lacking, scientific research is needed to develop and test new measures. Whether in
high-income, or low-income and middle-income countries, there are several good
practices that can be followed:
1. reducing exposure to risk through transport and land-use policies;
2. shaping the road network for road injury prevention;
3. improving visibility of road users;
4. promoting crash-protective vehicle design;
5. setting and securing compliance with road safety rules;
delivering post-crash care.

5.3 Managing exposure to risk through transport and land-use policies

Safety-conscious planning and design of the road network and of land use is necessary
to minimize the risk of road traffic injuries. Exposure to risk of road traffic injury can be
decreased by strategies that include:
• reducing the volume of motor vehicle traffic by means of better land use;
• providing efficient networks where the shortest or quickest routes coincide with
the safest routes;
• encouraging people to switch from higher-risk to lower-risk modes of transport;
• placing restrictions on motor vehicle users, on vehicles, or on the road
infrastructure;
• promoting safety-centred planning, design and operation of the road network.

5.4 The role of education and publicity

Public health sector campaigns in the field of road traffic injury prevention have
encompassed a wide range of measures, but education has always featured as one of
the key activities of prevention. Ongoing research and experience have led to re-
examination of the role that education plays in road traffic injury prevention. The
following are the key conclusions from research on this issue:
• Informing and educating road users can improve knowledge about the rules of
the road and about such matters as purchasing safer vehicles and equipment.
• Basic skills on how to control vehicles can be taught.
• Education can help to bring about a culture of concern and develop sympathetic
attitudes towards effective interventions.
• Most programmes providing highway safety education do not work in isolation
– they need to be linked or used in combination with other measures.
• There is a need for a balanced approach to the role of education and publicity,
taking into account evidence from research on behaviour change, the
interventions that may be promising, those to avoid, and those to encourage, as
well as the need to implement proven measures.
• General non-specific road safety campaigns should be avoided. Campaigns
should rather be used to put important questions on the agenda, and should
preferably support other measures such as new legislation or police enforcement.

5.5 Remedial action at high-risk crash sites

36
The systematic implementation of low-cost road and traffic engineering measures is a
highly cost-effective method of creating safer patterns of road use and correcting faults
in the planning and design of the roads that have led to traffic crashes. The use of road
safety audits and safety impact assessments can prevent such faults from being
introduced into new or modified roads.

Low-cost road and traffic engineering measures consist of physical measures taken
specifically to enhance the safety of the road system. Ideally, they are cheap, can be
implemented quickly, and are highly cost-effective.

5.6 Setting and enforcing seat-belt

The level of seat-belt use is influenced by: whether there is legislation mandating their
use; the degree to which enforcement of the law, complemented by publicity campaigns,
is carried out; incentives offered to encourage use. Mandatory seat-belt use has been
one of road injury prevention’s greatest success stories and has saved many lives.

5.7 Setting and enforcing child restraint use

The high level of effectiveness of child restraints in reducing fatal and serious injuries
was discussed in the previous chapter. Good protection requires that the type of
restraint used is appropriate for the age and weight of the child. Several restraint types
exist and are covered by international standards.

Effective interventions for increasing child restraint use include: laws mandating child
restraints; public information and enhanced enforcement campaigns; incentive
programmes and education programmes to support enforcement; child restraint loan
schemes.

5.8 Strategies to Reduce Crashes Involving Young Drivers

Graduated driver licensing (GDL) addresses both the inexperience and immaturity of
young drivers. GDL provides a structure in which beginning drivers gain substantial
driving experience in less- risky situations. GDL raises the minimum age of full licensure
and helps parents manage their teenage drivers. GDL’s effectiveness in reducing young
driver crashes has been demonstrated many times (Masten, Foss, & Marshall, 2013;
Russell, Vandermeer, & Hartling, 2011).

Driver education was developed to teach both driving skills and safe driving practices.
Based on evaluations to date, driver education for beginning drivers does a good job at
teaching driving skills, but has not definitively been shown to reduce the number of
crashes or crash rate. Rather, some research has suggested that it lowers the age at
which teenagers become licensed, and therefore increases exposure, so its overall effect
is to increase the number of crashes (Roberts et al., 2006; Thomas, Blomberg, & Fisher,
2012a; Vernick et al., 1999).

5.9 Strategies to Reduce Crashes and Injuries Involving Older Drivers

37
 • Educating and training older drivers to assess their driving capabilities and
limitations;
• Improving skills when possible;
• Voluntarily limiting driving to circumstances in which they can drive safely;
• Helping drivers adapt to medical or functional conditions that may affect driving
through treatment (such as eyeglasses or cataract surgery to improve vision) or
through vehicle adaptations (such as extra mirrors, extended gear shift levers, or
hand controls);
• Using license renewal procedures or referrals from law enforcement, physicians,
family, or friends to identify older drivers who cannot drive safely, in certain
situations or at all, and restrict or revoke their driver’s licenses

5.10 GRSF - Guide for Road Safety Interventions Report, March 2021
The World Bank Global Road Safety Facility (GRSF) has published a Report titled: Guide
for Road Safety Interventions: Evidence of What Works and What Does Not Work. The
guide will help road safety practitioners understand that what sometimes appear to be
“common-sense” approaches often not deliver the best road safety outcomes. Although
some interventions provide benefits, others have very limited or even negative impacts,
despite being commonly -and mistakenly - recommended and adopted.
https://www.roadsafetyfacility.org/publications/guide-road-safety-interventions-
evidence-what-works-and-what-does-not-work

38
6. Multidisciplinary Considerations in Safety Solutions

6.1 United Nations Road Safety Trust Fund Global Framework Plan of Action for
Road Safety, 2018

The future should thus see a more enhanced and multidisciplinary effort towards
creation of comprehensive and effective national road safety systems. This should be
founded on the international regulatory framework following the holistic approach of
the safe system. In addition, the United Nations Road Safety Trust Fund (UNRSTF)
should support this coordinated effort, building on the safety system considerations, as
it was agreed during the consultation on the objective of the Trust Fund. The New Global
Framework Plan of Action for Road Safety is presented below:

All 4 pillars interact with the 5 areas, providing individual actions of each interaction. For
more info:
https://unece.org/DAM/Road_Safety_Trust_Fund/Documents/UNRSTF_Global_Framew
ork_Plan_of_Action_21_Nov_2018.pdf

39
7. Effectiveness Evaluations: Safety Benefits

7.1 Crash Reduction Factors (CRF), Crash Modification Factors (CMF) &
Benefit/Cost Ratios (BCR)

Road Safety Countermeasures recommended by the RSA Team will cost money, and the
Project Owner may be hesitant to make the necessary expenditures without an
understanding of the potential payback. RSA Teams must be familiar with Crash
Reduction Factors (CRF) and Crash Modification Factors (CMF.) CMFs and CRFs help to
create the benefit/cost ratios needed to justify selected road safety countermeasures.

A Crash Reduction Factor (CRF) is the percentage crash reduction that might be
expected after implementing a given countermeasure at a specific location or along a
specific road segment. It is a generic estimate of the effectiveness of a countermeasure
based on thousands of other actual studies at similar sites.

A Crash Modification Factor (CMF) is a multiplicative factor used to compute the

expected number of crashes after implementing a given countermeasure.
The CRF is the inverse of the CMF. For example, if the CRF is 15%, then the CMF would
be 85%.

If more than one countermeasure is implemented at a site, the RSA Team will multiply
together the CMFs for each countermeasure. For example, if the CMF for
countermeasure 1 is 0.70 and the CMF for countermeasure 2 is 0.80, the RSA Team may
recommend that both countermeasures be implemented. If both countermeasures are
implemented, the anticipated CMF would be 0.70*0.80=0.56, and the corresponding REF
would be 0.44.

Two sources for CMFs and CRFs are the CMF Clearinghouse
(www.cmfclearinghouse.org) and the US Highway Safety Manual. Make sure you
understand the differences between CRFs and CMFs because different publications may
use one or the other, and this could cause confusion. CMFs/CRFs give road safety
specialist the data or the “muscle” they need to convince someone to implement a safety
project.

40
CRFs and CMFs alone may not be enough to persuade the Project Owner to implement
road safety countermeasures and improvements. CRFs and CMFs and Value of a Life
can help to create conclusive arguments in the form of a Benefit/Cost Ratio (BCR). A
BCR will use the Benefits in the form of the reduction or severity of the crashes resulting
in a Reduction in Number of casualties (death, severe and minor injuries); Reduction in
damage to property; and Reduction in costs of medical, insurance and police services
and divide the Present Value of these Benefits by the Present Value of the Costs of the
initial treatment + future maintenance and future upgrades. If this BCR is greater than
one, then it should be considered.

In a paper by Fai Deng of World Bank, which considered the Australian experience, RSA
recommended countermeasures had Benefit cost ratios (BCR) of between 3:1 and 242:1.
Seventy five percent (75%) of the recommendations had a BCR greater than ten (10),
and ninety percent (90%) of recommendations had a BCR greater than one (1).

7.2 The CMF CLEARINGHOUSE

The CMF Clearinghouse serves three important roles for the transportation safety field:

1. Provides CMF Data

The CMF Clearinghouse is a comprehensive and searchable database of published CMFs.
It contains all CMFs published in 2010 or after as well as many CMFs published before
that date, such as those compiled in the first edition of the AASHTO HIGHWAY SAFETY
MANUAL and the FHWA DESKTOP REFERENCE FOR CRASH REDUCTION FACTORS. It
provides information on all available CMFs, such as the CMF value and all published
details about the CMF; citations and related information about the study that produced
each CMF; and a star rating that provides an indication of the quality of each CMF. It is
regularly updated to include newly published CMFs.

2. Educates CMF Users

The CMF Clearinghouse provides guidance material to instruct users about the
appropriate use of CMFs. Through a series of FREQUENTLY ASKED QUESTIONS, the
Clearinghouse provides answers to many important CMF-related questions posed by
city and state transportation engineers, planners, and researchers. The CMF
Clearinghouse also sponsors an annual WEBINAR to provide guidance on appropriate
use of CMFs and best practices from state agencies. In addition to the guidance provided
from the Clearinghouse, the website also provides links to many EXTERNAL RESOURCES
in categories such as “How to Develop and Use CMFs”, “Cost-Benefit Analyses”, and
“Training”.

3. Facilitates CMF Research

Researchers often use CMF Clearinghouse data to determine if there are any CMFs
existing on a potential research topic and where research gaps are present. To direct
future research, the CMF MOST WANTED LIST is provided to show researchers and
41
funding agencies the countermeasures and topics that are high priority to
Clearinghouse users but not present in the Clearinghouse data. The CMF Clearinghouse
also provides safety researchers with a mechanism to SUBMIT CMFs for inclusion in the
Clearinghouse.

It is important to understand what the Clearinghouse does and does not include. The
CMF Clearinghouse presents CMFs from studies that meet the following criteria:

• The study must be based on crash data, not surrogate measures of safety such
as speed reductions, near misses, or yielding behavior.
• The study must have the objective of quantifying the safety effect of a roadway
feature or characteristic (i.e., as opposed to an academic exercise of comparing
model forms).
• The study must be focused on determining the safety effect of an infrastructure
characteristic, feature, or modification that would fall under engineering
responsibilities (e.g., not planning-level or area-wide characteristics such as land
use or demographics; not safety efforts unrelated to engineering such as public
safety awareness campaigns or law enforcement efforts).
• The study must explicitly present quantified CMF values or CMFunctions (i.e., the
Clearinghouse team does not derive CMFs if they are not explicitly reported by
the author).The Clearinghouse presents the CMFs as they are presented by the
author in the original source document. There is no modification made to the
CMF value or adjustment to any reported standard error.

For more info: http://www.cmfclearinghouse.org/collateral/CMF_UserGuide.pdf

7.3 The PRACT Repository

Stand-alone Regression Equation APMs are not available in any of the above web
databases. SPFs are available only in SPF Clearinghouse (to subscribers only), without
however providing adequate background information. Existing Databases include
mostly data from USA and Australia. Results from European studies are very uncommon.
In PRACT Repository all types of data required in accident prediction are available: CMFs,
SPFs, and Regression Equation APMs. The quality of included CMFs has been verified
through an evaluation process. User is provided with additional information to verify
the quality and the transferability of CMFs and APMs.

The repository has two parts: the CMF part and the APM part. Both parts are based on
the respective inventories developed within PRACT review process. All reviewed APMs
were included in the repository. Only high-quality CMFs were included in the repository,
on the basis of specific criteria. Quality criteria refer to: statistical design, testing for
statistical significance, and sample size. CMFs originating from the Highway Safety
Manual were considered “a priori” of adequate quality and were included in the
repository. All other CMFs were assessed prior to inclusion in the repository, on the basis
of fulfilling all of the quality criteria.
42
The quality criteria were applied to the CMFs (1,526 Factors and Functions) gathered
during the review process. 889 CMFs were found to satisfy the quality criteria and were
included in the repository.

For more info: www.pract-repository.eu

43
8. Effective Evaluations: Economic Tools

8.1 General
All countermeasures have costs associated with them. These costs can be separated into
3 main categories: Capital/Installation/Implementation, Operation and
Repair/Disruption costs.

Cost-benefit analysis (CBA) is a formal analysis of the impacts of a measure or

programme, based on welfare economics, designed to assess whether the advantages
(benefits) of the measure or programme are greater than its disadvantages. Unlike other
tools of efficiency assessment, CBA involves monetary estimates of both costs and
effects/benefits of a measure. Thus, in the road safety context, it may be useful for
prioritizing various road safety measures or programmes, particularly different accident
severities or additional impacts (on mobility, environment, etc.) are taken into account,
or trade-offs of road safety against other policy objectives need to be considered.

A Cost-Benefit Analysis (CBA) can be applied in order to compare the economic benefits.
It refers to costs to install and operate a countermeasure and calculates the financial
savings for a countermeasure:

𝐶𝐵𝐴=𝑆𝑎𝑣𝑖𝑛𝑔𝑠−𝐶𝑜𝑠𝑡𝑠

The equation gives an absolute number that may be hard to compare to other
countermeasures

A Cost Benefit Ratio (CBR) compares the benefits to the costs:

𝐶𝐵𝑅=𝑆𝑎𝑣𝑖𝑛𝑔𝑠/𝐶𝑜𝑠𝑡𝑠

When conducting CBAs of road safety measures, the following cost categories are
usually examined:
• Medical costs: costs resulting from the treatment of casualties, e.g. costs of
hospital stays, rehabilitation, medicines and adaptation.
• Production loss: loss of production and income resulting from the temporary or
permanent disability of injured, and the complete loss of production of fatalities.
• Human costs: immaterial costs of suffering, pain, sorrow, loss of life or quality of
life.
• Administrative costs: the costs of police services, fire services, law courts and
administrative costs of insurers.
• Property damage: damage to vehicles, freights, roads and personal property.
• Other costs: congestion, vehicle unavailability and funeral costs

8.2 The SafetyCube DSS Calculator

44
It combines information about the effectiveness of a measure (i.e. the percentage of
crashes or casualties prevented) with the costs of this measure. Integrates updated
information of crash costs in the
European countries. It allows to express all costs and benefits of a measure in
monetary values and conduct cost benefit analysis. The main functions include: perform
cost-benefit analysis with own input data, select one of the SafetyCube examples of cost
benefit analyses, measures with high effectiveness

For more info: https://www.roadsafety-dss.eu/

45
9. Comparing Alternatives and Identifying Positive and Negative
Impacts

9.1 Economic Appraisal Process in HSM

The 4th step of Roadway Safety Management Process in the Highway Safety Manual (as
discussed in Presentation 8.1) is the economic appraisal. Economic appraisals are
performed to compare the benefits of potential crash countermeasure to its project
costs. Site economic appraisals are conducted after the highway network is screened,
the selected sites are diagnosed, and potential countermeasures for reducing crash
frequency or crash severity are selected.

In an economic appraisal, project costs are addressed in monetary terms. Two types of
economic appraisal – benefit-cost analysis and cost-effectiveness analysis – address
project benefits in different ways. Both types begin quantifying the benefits of a
proposed project, expressed as the estimated change in crash frequency or severity of
crashes, as a result of implementing a countermeasure. In benefit-cost analysis, the
expected change in average crash frequency or severity is converted to monetary values,
summed, and compared to the cost of implementing the countermeasure. In cost-
effectiveness analysis, the change in crash frequency is compared directly to the cost of
implementing the countermeasure.

9.2 Prioritize Projects Process in HSM

The 5th step of Roadway Safety Management Process in the Highway Safety Manual (as
discussed in Presentation 8.1) is project prioritization. Prior to conducting prioritization,
one or more candidate countermeasures have been identified for possible
implementation at each of several sites, and an economic appraisal has been conducted
46
for each countermeasure. Each countermeasure that is determined to be economically
justified is included in the project prioritization process described in this chapter.

In the HSM, the term “prioritization” refers to a review of possible projects or project
alternatives for construction and developing an ordered list of recommended projects
based on the results of ranking and optimization processes. “Ranking” refers to an
ordered list of projects or project alternatives based on specific factors or project
benefits and costs. “Optimization” is used to describe the process by which a set of
projects or project alternatives are selected by maximizing benefits according to budget
and other constraints.

Ranking by Economic Effectiveness Measures

The simplest method for establishing project priorities involves ranking projects or
project alternatives by the following measures, including:
• Project costs,
• Monetary value of project benefits,
• Number of total crashes reduced,
• Number of fatal and incapacitating injury crashes reduced,
• Number of fatal and injury crashes reduced,
• Cost-effectiveness index, and,
• Net present value (NPV).

Incremental Benefit-Cost Analysis

Incremental benefit-cost analysis is an extension of the benefit-cost ratio (BCR) method.
The following steps describe the method in its simplest form:
1. Perform a BCR evaluation for each individual improvement project.
2. Arrange projects with a BCR greater than 1.0 in increasing order based on their
estimated cost. The project with the smallest cost is listed first.

47
3. Beginning at the top of the list, calculate the difference between the first and second
project’s benefits. Similarly calculate the difference between the costs of the first and
second projects. The differences between the benefits of the two projects and the
costs of the two are used to compute the BCR for the incremental investment.
4. If the BCR for the incremental investment is greater than 1.0, the project with the
higher cost is compared to the next project in the list. If the BCR for the incremental
investment is less than 1.0, the project with the lower cost is compared to the next
project in the list.
5. Repeat this process. The project selected in the last pairing is considered the best
economic investment.

9.3 Current Road Safety DSS Worldwide

• Crash Modification Factors Clearinghouse (www.cmfclearinghouse.org) by
NHTSA (USA) - 6.251 CMF on infrastructure only
• SafetyCube - Road Safety Decision Support System by the EC (www.roadsafety-
dss.eu/) - 1.250 studies 7.500 estimates of risks and measures (behaviour,
infrastructure, vehicle, post-crash)
• PRACT Repository (www.pract-repository.eu) by CEDR (Europe) - 889 CMF and
273 APM on infrastructure only - high quality
• iRAP toolkit (http://toolkit.irap.org/) by iRAP - 58 treatments (42 on
infrastructure)
• Safety Performance Factors Clearinghouse (http://spftool.com/) by Tatum Group
LLC, - few SPF - subscribers only

9.4 The SafetyCube DSS

The SafetyCube DSS objective is to provide the European and Global road safety
community a user friendly, web-based, interactive Decision Support Tool to properly
substantiate their road safety decisions for the actions, measures, programmes, policies
and strategies to be implemented at local, regional, national, European and international
level. The main contents of the SafetyCube DSS concern:
• road accident risk factors and problems
• road safety measures
• best estimate of effectiveness
• cost-benefit evaluation
• all related analytic background
Special focus on linking road safety problems with related measures.

SafetyCube methodology relies on:

1. Creating taxonomies of risk factors and measures
2. Exhaustive literature review and rigorous study selection criteria
3. Use of a template for coding studies, to be introduced in the
DSS back-end database
4. Carrying out meta-analyses to estimate the effects of
risk factors / measures.
48
 5. Drafting Synopses summarising results of risk factors / measures.
6. Carrying out cost-benefit analyses for the most effective measures
• Systems approach: links between infrastructure, user and vehicle risks
• Rigorous assessment of the quality of the data / study methods

For more info: https://www.roadsafety-dss.eu/

49
Training programs for Road Safety Professional
Exam (Level 3):

RSP1-SA

Phase Three

22-25/03/2021

1
 Module 5
Road Safety Strategic Plans and
Programs

Overview

Unit Modules Topics in this

Module

1-Basics of Road Safety

2-Effects of Human Behavior on Road Safety

How strategic safety plans are prepared & used.

3-Data and Its Role in Measuring Road Safety

The important elements of successful road safety
policies and programs

Role and value of leaders, safety champions, and

4- Solving Road Safety Problems coalitions in influencing road safety programs

Elements of successful communication and

outreach strategies
5-Traffic Safety Strategic Plans and Programs
How multidisciplinary teams and partnerships can
achieve road safety goals

Safety program evaluation & explain how results

influence future program delivery

2
Outline
I. Implementing Road Safety Strategic Plans and Programs ............................... 4
A. Introduction ................................................................................................................ 4
B. Historical Road & Traffic Safety in KSA................................................................. 4
II. Strategic Safety Plans ..................................................................................................... 5
III. Success with Safety Policies and Programs .......................................................... 16
IV. Roles of Leaders, Champions and Coalitions.................................................... 23
V. Communications and Outreach ................................................................................ 30
VI. Multiple Disciplines in Road Safety ...................................................................... 33
VII. Safety Program Delivery ........................................................................................... 38

3
I. Implementing Road Safety Strategic Plans and
Programs
A. Introduction
This module is primarily about MANAGEMENT. Management of road safety
involves a variety of processes including strategic planning, coordination of
different multi-sectoral stakeholders, agency coordination, understanding of
legislation, promotion of safety to the public, monitoring, evaluation and many
other functions. One of the most important lessons to remember is that above
all, making the political and economic case for road safety investment will be a
constant process that must be continually looked after at EVERY level: political,
social, environmental, and down to the level of being an individual champion
for road safety at your job.

B. Historical Road & Traffic Safety in KSA

In this section, several important aspects of the strategic plan for traffic safety
in the Kingdom of Saudi Arabia (KSA) will be reviewed and discussed. This
module provides a comprehensive view of traffic safety in the Kingdom.

National Road and traffic safety strategy plan had passed through several
stages as follows:

▪ Traffic Safety committee, King Abdelaziz for science and technology

(KAST) sponsored by KAST, 1984.
▪ Strategic Traffic safety plan, Ariyadh Development Authority (ADA)
sponsored by ADA, 2003.
▪ Traffic safety committee, Eastern province Sponsored by Aramco
company, 2009.
▪ Traffic Safety Association at Imam Mohammed bin Abdurrahman
University, 2010.
▪ Transportation strategy sponsored by ministry of transportation (MOT),
2011.
▪ National strategic plan for traffic safety, sponsored by KAST, 2012.
▪ National Transformation program 1, (NTP1) sponsored by Government,
2016.
▪ National Transformation program 2, (NTP2) sponsored by Government,
2017.
▪ The Ministerial committee of traffic safety – General secretarial, 2018.

4
II. Strategic Safety Plans
Strategic safety plans should be developed and thought of as a resource for
directing planning efforts and allocating available resources. The strategic
safety plan should be thought of as a roadmap to assist in planning for all the
separate priorities of the road system. These priorities include:

• Economic Vitality
• Safety
• Security
• Mobility and Accessibility
• Environmental Protection
• Integration and Connectivity
• Management and Operations
• System Preservation

The transportation planning process is much like other strategic planning

processes. The first step is to develop an understanding of the environment
and the problem to be solved. Along with the above listed priorities for every
strategic road safety plan, the plans should be developed to help understand
and address following overall transportation planning components:

• Environmental Scan
• Understanding the Problem
• Contributing Factors
• Roadway conditions
• Road user demographics
• Environmental conditions
• Behavioral characteristics

It should also be understood that the initial strategic safety plan and its
implementation is a process which will need to be continually assessed and may
potentially need to evolve over a period of time. The Strategic safety plan is an
initial formwork which sets the overall vision for the program, creating goals for
the program to meet to achieve that vision, creates objectives in which to

5
measure whether goals are being achieved, and to provide performance and
evaluation criteria for assessment.

Figure 1: Shows general relationship among vision, goals, objectives, evaluation criteria,
performance measures, an targets

Evaluation of these metrics will be all the level of the Ministry to know what is
or is not working in the strategic safety plan and provide direction for
improvement as iterations of the plan are implemented. The Strategic Safety
Plan should be continually evaluated and altered as necessary to keep with
current trends and technologies both in construction and design, as well as in
the vehicles on the roads.

Development of Performance Measures:

As part of creating the plan evaluation metrics, it is important to identify,

understand and gain insights from:

• Stakeholders/Constituents
• Decision Makers
• Legislative/Program Requirements
• Data Analysis
• Current Goals and Objectives

6
 Figure 2: Performance measures

The performance measures and evaluation of the plan should be constant,

sourcing feedback from all partners. Internationally, these partners are
commonly members of the following groups:

• Department of transport (DOTs)

• Highway Safety Offices
• Departments of Motor Vehicles
• Operation Lifesaver and Highway Rail Grade
Crossing Managers
• State and Local Law Enforcement
• Motor Carrier Safety Assistance Program Administrators
• Other Stakeholders
These partners can assist with the selection and evaluation of the candidate
performance measures, providing feedback on a continual basis to determine
which plan aspects are providing better results than others, or where
unintended outcomes maybe occurring:

7
 Figure 3: Performance measures feeding into different stages

At the global level, the United Nations began passing Resolutions for road
safety back in 2005 and at the same time the World Bank and WHO published
the World Report on Road Traffic Injury Prevention (2005) which laid out a
strategic framework to reduce global traffic fatalities, 90% of which happen in
the developing world.
A road safety management system can be thought of like a pyramid, with
RESULTS at the top, INTERVENTIONS in the middle and INSTITUTIONAL
MANAGEMENT FUNCTIONS at the base. Most countries focus on interventions,
without enough attention to the various elements of the other two.

8
 Figure 4: Road Safety Management System Strategy chart

Strategic Safety Plans are critical documents for achieving outcomes. At the
broad level, they should:

• Address management, engineering, operations, education,

enforcement, and emergency services as key factors of evaluating
highway safety projects
• Focus on safety needs of public roads (especially high-fatality safety
areas)
• Describe strategies to reduce or eliminate safety hazards
• Have political approval

A specific example in the USA is the NHTSA/ State Traffic Records Coordinating
Committee (TRCC) Strategic Planning Guide. It: “provides State TRCC’s with a
practical, replicable process for developing and implementing effective
strategic plans that will improve safety data quality and reduce roadway
fatalities”

Another in the USA is the Governor’s Highways Safety Association (GHSA)

Guidance on Highways Safety Plans. Identify the highway safety problems It
helps describe a State’s overall highway safety performance measures, define
the State’s performance targets, and develop and select evidence-based
counter measure strategies and projects.

Strategic plans influence funding and resources by identifying critical areas of

investment, agreed-on areas of prioritization by leaders, specialists, and the
public, anticipating needed federal funding vs. state funding, strongly
9
influencing project proposals (compliance documentation), and using safety
data network screening as a precursor to funding/resources.

The UK is a best practice coutry when it comes to creating strategic safety

plans, as well as casualty reduction targets that measure AND project fatality
reductions based on reccomended interventions (per the KSI figure below).
The analsyis of the top 10 KSI collision contributory factors from 2015 below
does not cover the full multi-disciplinary nature of road safety but the clear
emphasis on statistics that have been attributed to specific factors allows for
current and future policy designs in interventions.

Top 10 KSI Collision Contributory Factors

2015 (Stats 19)
• Failed to
• Failed to judge other
•Loss of •Poor turn or
look person’s
control manoeuvre
properly path or
speed
523 368 353 223

•Careless,
• Travelling
Reckless or in • Sudden
• Fa gue too fast for
a hurry Braking
condi ons
222 145 136 127

• Distrac on
• Swerved
in vehicle

120 118
80.9% of
KSI NRUSS - Over 90% of
casual es road users felt safe on
due to their journey on the
human SRN
factors 9

Figure 5: Top 10 KSI Collision Contributory Factors

A safety management system should MOVE AWAY FROM a philosophy of

REACTING TO SAFETY PROBLEMS AND FIXING THEM and MOVE TOWARDS a
philosophy of PRE-EMPTIVE SAFETY, INCIDENT PREVENTION, AND
PARTENRSHIPS. It is critical for students of road safety to create a mindset early

10
in their career that CRASHES ARE PREVENTABLE and PREDICATBLE and lives are
saved when planning accordingly.

This is why the SAFE SYSTEM is so widely popular—it places the reduction of
casualties at the heart of its philosophy and this is why it has become the
framework for global action under the United Nations Road Safety
Collaboration. At the country level, its application on a systematic level can
allow for the SAFE SYSTEM to become a long-term philospophy at BOTH a
political and application level while also setting SHORT TERM targets.

• All countries can adopt this “elimination” vision for the long term

• Not necessary to specify when it will be reached – but strong statement

about unacceptability of serious casualties

• Informs communities - this is a reasonable long term expectation

• Sets the context for short term decisions

• Responds to a strong and growing market for safety as community

awareness grows

National strategic plan for traffic safety

In 2012, the KSA had established the national strategic plan for traffic safety as
a comprehensive plan for traffic safety. The vision of the plan was adopted
towards a safe society by reducing traffic accidents and their consequences. The
vision aims at resilient plans to achieve a safety society including “improvement
of the traffic safety”, significant reduction in the rates of traffic accidents,
minimizing causalities including deaths and injuries, and mitigate the direct and
indirect social and economic impacts.

Accordingly, 10 road safety "Key areas" have been approved, 6 of which are
main, and the others are possible as a basis for the plan, as follows:

• Strategic Pillars
- Urban planning and Transportation
- The road
- Traffic management and new technologies
- The vehicle
- Education, awareness and education
- Ambulatory and therapeutic services

11
 • Enablers
- Regulation & organization
- Human and resource capabilities
- Research & information
- Funding of traffic safety

The "kay areas" of the traffic safety system:

Traffic regulations and legistlations, resourses and human

potenitials, traffic tasks structure

Statistical indicators and traffic data

The road and recent technologies

Traffic behaviors, education, training and rehabilitation

Urban planning and school and public transportation

The vehicle

Medical and emergency services

Traffic studies and research

The National Traffic Safety Plan consists of 70 projects distributed to legislation,

developmental and constructional.

12
 Figure 6: National Strategic Plan for Traffic Safety

National transformation program (NTP)

In 2016, the ambitious goals to the vision of the first Vision
Realization Program (VRP) was established and named the national
transformation program (NTP). Enhancing Traffic Safety (ETS) is
playing a significant role in the National Transformation Program
out of the 12 programs of the Vision 2030. The NTP 2.0 VRP
comprised of four Strategic Objectives/ portfolios under Ministry of
Health that ETS constitutes. To make a real and sustainable
transformation in the traffic safety system based on enhancing
cooperation between all related parties/agencies and lead to a
reduction in the mortality rate resulted from traffic accidents to 23
per 100,000 population and to reduce the number of injuries to 66
injury per 100,000. The strategic goals for NTP program are:

• Ease the Access to healthcare services

• Improve Value of healthcare services
• Strengthen Prevention against health threats
• Enhance Traffic Safety (ETS)
- 18 initiatives
▪ with 4 KPIs for level 3
▪ 8 strategic objectives for level 4
▪ 8 KPIs for level 4

Strategic Pillars

- Data collection and standardization

13
 - Awareness, education, and social contribution
- Superior Road Infrastructure
- Safe Vehicles
- Firm Law Enforcement
- Efficient Traffic Management
- Effective Accidents Management
- Comprehensive Trauma System

Business Model
➢ Government buy-in: Ensuring that the Strategy or business case developed
is endorsed by the government and it will be supported by the government.
If the government is not “interested”, you will have low buy-in.
➢ Identify Root Causes: Root Causes analysis helps in revealing pain areas
that are not shown in the surface and give insights to the direct cause of the
problem rather than the obvious. Root Cases analysis usually sheds the
lights on solutions that will mitigate the origins of the problem.
➢ Building Executable Strategy: Building a strategy that is not impossible to
implement and execute is a key factor. The executable strategy must be
aligned with the organization’s goals and the Stakeholders, operating model
of the government, how decisions are made, and ease of implementation.
➢ Establishing Governance: The governance model governs the structure of
government and reflect the interrelated relationships in the organization.
The governance model helps in taking key decisions from the authorities
and cascades them in addition to requesting required support from
execution level to authorities. The Traffic Safety Committee is an example of
an established governance.

14
Figure 7: Key Performance Indicators

15
III. Success with Safety Policies and Programs
Road safety management systems have specific characteristics that are useful
when applied in any given country situation.

• Place and emphasis on the production of safety

• Neutral to country structures and cultures
• Accommodate evolutionary development
• Work within any given land-use /transport system and its current and
projected risk exposure
• It takes the road network as its frame of reference and locates the
deaths and injuries that are avoidable.

Safety projects that give priority to road safety management/coordination, data

and safe infrastructure often are the foundations of successful reductions in
death and injury. Secondary priorities include a mix of enforcement
enhancement, campaigns, EMS response establishment/improvement,
improving vehicle inspection and quality of stock in a country etc. However,
each prioritization is highly specific to the local conditions which MUST first be
assessed with a strategic review that has recommendations that are agreed
upon by relevant agencies and politicians, and ideally the public has provided
a review and endorsement.

Safe System project components can be broken down as follows:

1. Capacity strengthening priorities:

- Lead agency development

- Crash database development

- Other institutional reforms

2. High-risk corridors and areas to be targeted with good practice

interventions:

- Infrastructure safety improvements

- General deterrence-based traffic safety enforcement programs, supported by

intensive publicity & awareness campaigns (e.g. speed, alcohol, safety belts &
helmets, fatigue, commercial vehicles)

- Improved post-crash response and emergency medical and

rehabilitation services

16
3. Policy reforms (e.g. driver licensing, vehicle safety standards)

4. Project management arrangements:

- Lead agency role

- Coordination with sectors

5. Monitoring and evaluation system:

- Performance targets for high-risk corridors and areas

- Procedure monitoring

- Reporting arrangements to evaluate results back to implementing agencies

and agreed targets

The above is a general outline and specialists can be recruited to make sure
each aspect is sufficiently covered (e.g. monitoring and evaluation specialists,
police specialists, campaign specialists, etc.). The lead agency should play a
central role in oversight.

Common attributes of successful programs can include:

• Documentation of the process

• Coordination with partners

• Relationship between different plans

• Data-driven decisions

• Use advanced safety analysis methods and tools

• Address local road needs

• Consider all 4E’s (or Safe System)

• Establish multi-year plan and budget

• Streamline Project Delivery

• Evaluate the success of the program

Research is a core part of the Safety Management Pyramid (part of the

management function). Specifically:

17
• TRB, FHWA, NHTSA, AAA, Development Banks, Universities, etc.: all are part
of the road safety global research chain

• Research further develops human factors, vehicle technology, road

infrastructure, countermeasures, database/accident stats, risk factors,
communication strategy, speed cameras etc. ALL of the 5 Pillars of the
Decade Plan have various long histories of research.

• Research influence outcomes by shaping the intervention decision process

upstream BUT ALSO ensuring future interventions are “challenged” and
“evolved” by new and continuous research processes.

In this section will clarify the implement the policies and programs. Policies are
an essential part of implementing, operating, and managing the programs and
initiatives involved in a road safety strategy. The NTP2 program has been taken
as a case study.

Data collection and standardization

Data collection is one of the main pillars of the comprehensive traffic safety
system, ensuring that decisions are based on accurate data. Build Institutional
Capacity to Collect, Document, and Analyze Traffic Safety System Data.

18
 Figure 8: Data collection Basher System

Awareness, education and social contribution

Awareness and education are a cornerstone of the comprehensive system of

traffic safety. They contribute to reducing serious traffic violations by raising the
level of education and awareness of traffic safety. For example (innovative
national media campaigns focusing on positive behaviors)

Figure 9: Media, GDOT Tweeter collection Basher system

Superior Road Infrastructure

The optimum road infrastructure is a vital components of the comprehensive
system of traffic safety. The roads on which the vehicles are to be fully equipped

19
must be designed/operated against deaths and injuries resulted from traffic
crashes.

Figure 10: Road Infrastructure

Safe Vehicles
Vehicle safety is also another part of the comprehensive traffic safety system
as vehicle safety contributes significantly to accident prevention. Ensure
vehicle compatibility with traffic safety norms and standards.

20
 Figure 11: Vehicle periodic inspection

Firm Law Enforcement

The implementation of the law is one of the most important pillars in the
comprehensive system of traffic safety, which contributes greatly to the
discipline of drivers and the community's compliance with traffic laws and
regulations.

Figure 12: Updated traffic law

21
Efficient Traffic Management
Effective traffic management is one of the cornerstones of the comprehensive
traffic safety system, as it allows for a significant reduction in traffic congestion
and effective and rapid response to traffic problems. Improve supervision and
control techniques to ensure effective traffic management within cities and on
highways.

Figure 13: Operation measurement

Effective Accidents Management

Accident management is one of the main pillars of the comprehensive
system of traffic safety. It ensures that accidents are dealt with quickly and
efficiently, thus contributing to the reduction of deaths and injuries resulting
from traffic accidents.

Figure 14: Basher dashboard

22
IV. Roles of Leaders, Champions and Coalitions
The world is facing a major global health challenge with road crashes. However
despite a UN Decade of Action 2011-2020 (and now another 2021-2030), and
many UN Resolutions, road safety still is not a priority for most countries’
leaders.

To make a concrete, positive change in improving road safety, strong leaders

are vitally important. While safety champions and coalitions are useful and
necessary, a leader can:

• Bring cohesiveness to a diverse field

• Ensure technical knowledge is used
• Make safety a priority in a competitive policy environment
Leaders are those people who can inspire their followers and provide direction
to a group or cause. The field of safety, especially as it pertains to roads, is a
tremendously diverse field with a hugely diverse field of contributors ranging
from educators to politicians, to public servants and designers. While these
groups can work in harmony, each has their own agenda based on their own
backgrounds. A leader is needed to corral these fields and experts to ensure the
focus stays on the strategic goals common to the safety of the public. In the
policy environment, politicians are tasked with tackling so many different tasks,
without a strong leader, any one task, including road safety, could easily be
omitted or passed over inadvertently or purposefully. A strong leader is needed
to ensure the safety policy remains a priority in political discussion and
decisions.

As discussed, there are many types of leaders in the realm of road safety. Some
are those placed in positions of leadership or power directly, others are less
obvious. Below is a brief list of some possible examples of leaders:

• Examples:
• DOT Secretary – directly affects road safety programs and funding
• Law enforcement – directly sees and understands actual human
behavior on the roads, understands the human behavior what
policies/road safety practices may not work as intended
• Trauma nurse – directly sees the results and aftermath of road
safety failures. Has the ability to reach individuals directly affected
road incidents

23
When defining leadership, or leaders themselves, it is important to understand
that leading is NOT managing.

• Leadership focuses on: setting a direction or goals and motivation of a

group or mission.
• Management focus on: the execution of tasks.

Both are needed to accomplish the end goals. Both can be exhibited by a single
individual.

While there are no defined set of skills or traits which leaders must all exhibit to
be successful, the following are a list of common skills and traits exhibited at
some level by most successful leaders:

Traits:

• Self-confidence and decisiveness – Not making a decision, is still

making a decision
• Resilience and energy – Failure, at some level, is almost guaranteed.
Maintain the resilience and energy to get back up and try again.
• Need for achievement/service mentality – Leaders want to serve, and
want to achieve, for themselves and others
• Flexibility – Leaders must adapt to the situation, the people, and the
goals to be successful
• Emotional maturity – Leaders must have self-awareness and self-
control. They need to understand their own weaknesses in order to
mitigate their affect
Skills:

• Communication – Leaders must communicate and motivate their

constituents
• Social – Leaders must be able to connect with others. Be able to
understand others positions and be able to motivate a large variety of
people and viewpoints to work towards the same goal
• Problem solving – Leaders will always be presented with difficult
problems or issues. While a leader is not required to know every answer
to every problem, they need to be willing and able to acknowledge when
an answer is not readily available and work to seek out the solution.

24
 • Technical – Leaders need to have a solid understanding of the technical
aspects of the problems or groups they are leading. Leaders need to be
able to make sound, information based decisions.
• Influence – Leaders must be able to influence others. There will always
be some doubt/descent within a group. A leader must be able to find a
way to influence their constituents to rally toward the common goal(s).
As mentioned, leaders must be able to influence. There are many ways of
expressing influence over individuals or groups:

• Pressuring – Using one position of power to try to force another to

change an opinion or follow them. This method of influence should be
avoided as it can diminish group cohesiveness and respect
• Exchange – Quid-pro-quo, or the understanding that a favor may be
returned in the future. This method of influence must be used with
extreme caution as it can lead to unethical decisions
• Legitimating – Using existing rules or regulations to appeal to another
nature.
• Rational persuasion – Using facts and figures to appeal to another
• Consultation – Allowing a person or group to take part in goal setting
or decision making to attain “buy-in”
• Emotional appeals
• Personal appeals
• Friendliness

25
At the UN level, a UN Special Envoy has been created with a secretariat located
at the United Nations Economic Commission for Europe. This individual has
helped promote road safety to world leaders and many influencers. This is an
example of a global leader on road safety and a champion of road safety.

Many coalitions have formed across the world, such as the Global Youth
Coalition for Road Safety which was extremely active during the recent 2020
Stockholm Ministerial Conference. This is especially relevant since road crashes
are among the leading cause of death age 5-29.

Victims of road trauma often make coalitions, such as the Mothers Against
Drunk Driving (MADD) in the United States, or the NGO Global Alliance for Road
Safety at the global level.

Types of champions, coalitions and leaders include:

• Elected officials: May or may not be part of official duties, but all officials
can choose to have a role in public health by their actions/examples

• Advocacy Groups: Formal role to bring issues to public

• Industry: Lead by corporate safety policy

• International: Global organizations (e.g. UN, WHO), UN Special Envoy

• Celebrity: Bringing “star” power to a cause

Coalition building can be broken down into specific steps:

• Recognize problem—it is important to identify the problem for what it

is: e.g. drunk driving, lack of speed enforcement, lack of youth
involvement, poor infrastructure, unacceptable death and injury level for
young people, etc.

• Identify all players on the field: scan and categorize who is working and
is responsible for the problem.

• Determine extent or scale of problem: through statistical analysis,

qualitative surveys, research, community engagement, etc.

• Consider solutions in coordination with coalition members

• Consider consequences for each proposed action: understand that every

course of activity can lead to positive or negative outcomes. (SWOT
analysis: Strength, Weaknesses, Opportunities, Threats).

26
 • Inform coalition of the choice of action

• Work with policymakers and carry out the plan: preferably together and
not in opposition.

• Evaluate the choice of action(s): preferably with surveys, qualitative and

quantitative analysis, etc.

Champions, leaders and coalitions can emerge at any level. This includes those
taking the course through individual actions, such as ensuring you and your
family wears seat belts and does not practice distracting driving, to how you
perform your job as a road safety professional. At the government level, an
agency may declare a policy ensuring and encouraging road safety for its
employee, or a private-sector company may decide to provide specific road
crash avoidance training for its employees.

Anyone can help secure the needed leadership, resources, visibility, support and
commitment from partners. We need to all be road safety champions.

27
Ministerial and Executive committees as depicted in figure 10 for Traffic Safety
has been established along with the General Secretary, by royal decree no. 636
dated 23/10/1438H. The role of ministerial is an essential work to manage the
traffic safety at high level. Ministerial committee consists of ministers and high
leading members. Two committees including executive and technical
committee are leading the traffic safety plans in professional way to achieve the
strategic goals which help to reduce the fatalities and injuries accidents. The
chart structure clarifies work follow and the tasks according to committee type.
Champions provide enthusiasm and support for the safety programs and act at
the heart of the relationship and cohesion between all stakeholders.

Figure 15: Structure of Ministerial Committee

28
 Figure 16: Champion, Leaders and coalition in KSA

Champion provides enthusiasm and support for a road safety program, and
responsible for securing for leadership buy-in, financial resources, public and
institutional visibility for program, continuous support, multisectoral /partners
commitment.

Leader recognized a need for change, acted on the need, and inspired others
to follow. Anyone with drive, dedication, and a good idea can lead.

Coalition should give participants the opportunity to describe their safety

concerns and current programs. This may advance the discussion of critical
safety issues, identify opportunities, and forge an agreement on how to proceed

29
V. Communications and Outreach
The primary goal of any communications and outreach campaign for road
safety is to influence public opinion in a way that optimizes a given desired
outcome.

Mass media (traditional media) can still play an important role, but increasingly
new media (social media and the internet), along side telecoms (business-to-
consumer messaging) plays a role in influencing campaign outcomes.

Basic outreach types may include:

• Advertising: Formation of messaging that has been vetted (possibly by

research) in print or media

• Public meetings: Local, city-wide, state-wide, region and global

• Safety fairs: highlight technology or intervention (with or without public

participation)

• PSA: Public service announcement: targeted for larger audience to

highlight new law, technology, future safety development, etc.

• Choosing an outreach strategy depends on goal, agency, statutory

responsibility, budget, etc.

Recognition vs. recall: two critical and frequently measured principles are at the
heart of campaigns. Recognition is the retrieval method of identifying correct
information among alternatives. Recall involves reproducing information stored
in memory with the fewest possible cues. It is the aim of many campaigns to
trigger the shortest path for a consumer to remember the core message.

Improving transportation safety involves changing the behavior and attitude of

individuals who use the transportation system rather than purely modifications
to the roadway and surrounding environment. One way to change behavior and
attitude is by utilizing effective communication approaches. Communication
approaches, such as communicating new knowledge, are targeted and tested
over long period of time and are considered develop interactive methods which
consider concrete steps to avoid danger. Other successful approaches include:

• Campaigns with careful pre-testing and delineation of a target group

that receives the messages;

• Longer-term programs that deliver a message in sufficient intensity over

time;
30
 • Education programs built around behavioral change models, using
interactive methods to teach skills to resist social influences through role
playing;

• Public information campaigns implemented in conjunction with other

ongoing prevention activities; and

• Fear messages combined with concrete steps to avoid the danger.

Dissimilarly, research has also shown that there are ineffective communication
approaches. Ineffective communication approaches, such as ineffective public
information and education, are typically associated with:

• Passive messaging communicated by signs, pamphlets, brochures and

buttons;

• Slogans that give simple exhortations for people to behave in certain

ways to avoid undesirable outcomes;

• Education programs that are lecture oriented and informative only;

• Short-term programs with low-intensity messages; and

• Use of extreme fear or scare techniques, especially when directed at

adolescents and without clear instruction of steps necessary to avoid the
danger.

Another form of effective communications is public outreach and public

involvement. Public involvement allows the citizens of the town or city to
provide their thoughts and meaningful inputs into the decision-making process.
Meetings can be held with the public as an opportunity to educate them on
road safety issues and give planners and stakeholders’ feedback on strategies
and countermeasures. Additionally, public involvement promotes ownership of
policies and draws on the public’s knowledge and views of road safety which
can result in decisions that reflect community values. Another benefit of
effective public involvement is it builds the public’s trust which results in
acceptance and support of new safety transportation policies.

Duration of a campaign is based on the goals. It can be seasonal (such as around

holidays) or continuous (all-year round for example). Coordination with other
sectors is vital for road safety campaigns. For example, an anti-speed campaign
which encourages the public to drive more slowly can be coordinated with a
police-enforcement campaign simultaneously.

31
The use of celebrities in road safety campaigns is found in many countries, but
effectiveness is mixed. In Australia and New Zealand for example, some of the
most effective campaigns emphasize both personal responsibility,
consequences of breaking the law, and the science of road safety. This is in part
due to a culture in these countries of individual responsibility both to protect
oneself as well as ensuring the well-being of the general society. This is also
part of the reason “traffic safety culture” influences road user behavior which in
turn influences the actions that lead or do not lead to crash severity.

In Saudi Arabia, the communication is an essential part of Saudi's traffic safety

plan, ministerial committee has established a media committee incorporating
with RSC to manage the campaign for road traffic safety. Ministry of media leads
the media plan that is organizing the campaign and programs based on goals
and targets to promote the road & traffic safety. Media plan (Vision, Messages,
and objectives) as follow:

Vision
We strive for traffic safety in the Kingdom to be the safest in the world

Message
Establishing the safety principle of the traffic environment

Objectives:
- Raise the rate of traffic safety for the Kingdom internationally.
- Raise awareness of traffic regulations.
- Promote the concept of traffic safety through awareness
programs and events.
- Unifying media campaigns between relevant sectors, government
agencies and the private sector.
- Motivating the partners in the Traffic Safety Committee to
participate in the awareness campaigns.

32
VI. Multiple Disciplines in Road Safety
Multidisciplinary teams can achieve road safety goals through collaboration
between partners. Partnerships can lead to effective programs, projects,
initiatives, and investments. Collaboration through partnerships is KEY
especially within agencies. Many agencies play a major role in road safety
program management and provide different knowledge, expertise, and
approaches to problem solving.

These agencies are each tasked with specific focus areas and seek to address
safety issues for various modes of travel. The following chart shows these
various focus areas and the agencies tasked with addressing these areas in USA.

Figure 17: (Carter et al., 2017)

33
The Five Pillar UN Plan for Road Safety is based on a fundamental principle of
multi-sectoral interventions:

• Road Safety Management

• Safer Infrastructure and Mobility
• Safe Vehicles
• Safer Road Users
• Post-Crash Care

These have been largely adopted on the belief that at some level, in every
country of the world at least some aspect of each pillar can be improved upon
and must be improved upon in order to meet the goals of the last Decade of
Action.

This is also based on calculations by WHO and other global leading data
providers that road crashes are vastly under-reported across the world, from
less than 10% under-reporting in many high-income countries to over 500%
under-reporting in many low income countries, particularly in Africa. Given this
problem, data collection remains a priority in order to effectively allocate
funding proportionally to each pillar in any given state or country.

One rule GENERALLY is constant: in any given road network, a SMALL

PERCENTAGE OF THE NETWORK CAN BE RESPONSIBLE FOR A HIGH
PERCENTAGE OF THE SOCIAL COSTS OF CRASHES. As the example given in
class, 20% of roads can account for 79% of traffic and 87% of the social cost of

34
crash.

Quality or Quan ty?

Figure 18: Targeting the network

This is important to understand in order to prioritize and target interventions in

a way that makes sense to save lives and efficiently distribute funding. Tools like
road assessment programs, road safety audit and road safety inspection
become important to help effectively identify where the prioritization should
go.

In more complex multi-sectoral projects, involvement can include the lead

agency, the public works department, enforcement, road infrastructure
improvement, monitoring and evaluation, education and safety campaigns,
and post-impact care. Bringing together multiple sectors is a good way to boost
capacity and ensure that all relevant parts of the system work together in
partnership. This will not always be the case in every project, but it is increasingly
recognized that small, stand-alone projects may not always be the most
efficient way to build capacity or produce results.

Two of the best documents on the above subjects are:

• Bliss/Breen (2009): Country Guidelines for the Conduct of Road Safety

Management Capacity Reviews and the Specification of Lead
Agency Reforms, Investment Strategies and Safe System Projects (
35
 • ITF/OECD (2008): Towards Zero: Ambitious Road Safety Targets and
the Safe System Approach

A comprehensive road safety strategy should be driven by good quality data

and encompass the “4 E’s” which are all tied together through effective
partnerships.

Figure 19: Multi-discipline in KSA

36
Example: Hot spot process that give a good example how partnership work
together with good governance.

Figure 20: Flowchart of the process of hotspot

On the other hand, 4’Es strategy can be worked though the traffic safety
committee at city level.

Figure 21: Flowchart of the Supreme Committee for Traffic Safety

37
VII. Safety Program Delivery

The management functions of the road safety strategy pyramid become

important for ensuring a maximum outcome for road safety program delivery.
These include:

• Results focus
• Coordination
• Legislation
• Funding and resource allocation
• Promotion
• Monitoring and evaluation
• Research and knowledge transfer

Funding road safety also is a constant battle between the need to secure a
funding stream to adequately address the issue and the political will to deliver
that funding. There are many revenue options, including road users directly, the
private sector, insurance companies, vehicles manufactures, international
financial institutions, civil society, philanthropies, and national, state and local
government.

Cos ng Road Safety

Total Annual Cost to Society

Costs of Lost Short to Long-Term Human Cost/Effect

Administra ve Costs Property Damage
Economic Output Healthcare Costs on the Poor

Target
(% Reduc on in Costs)

Es mated Investment to Achieve

Target

Annual Road Safety

Expenditure

Figure 22: Flowchart Costing road safety

The cost to society of road crashes can be substantial. On average, it can cost
2-4% of GDP in developed countries and 10-12% in many low income countries

38
as discussed in previous modules. Substantial health care costs are involved in
the hospitalization and rehabiliation of crash vitims, as well as adding to the
social safety net in countries that have programs in place to support victims
(many do not in developing countries). Road crashes are a serious and suddent
event in life which can create MULTI-GENERATIONAL poverty for crash vicims:
lost wages, disability, strain on family finances when all family members cannot
work, prison time, or death of the main financial provider.

However, it must be very clear that providing safe and effective road safety is
PRIMARILY A GOVERNMENT FUNCTION since government SHOULD recognize
the social and fiancial costs of road crashes. The government is in almost every
circumstance the one entity to take responsibility for the social welfare of its
citizens, primarily in terms of health outcomes. The transit routes of a country,
heavily and primarily dependent on the road network, provide vital economic
and social outcomesand should be treated as important as any other top safety
and security priority.

39
It important to have insight on the process outcome criteria that explaining the
evaluating steps method. The following framework for evaluating collaborative
planning efforts considers both process and outcome criteria for evaluation.

Process Criteria
• The process includes representatives of all relevant interests.
• The process allows participants to decide on ground rules, objectives,
tasks, working groups, and discussion topics.
• The process engages participants, keeping them at the table, interested,
and learning.
• The process encourages challenges to assumptions and the status quo
and fosters creative thinking.
• The process incorporates many kinds of high quality information and
ensures agreement on its meaning.
• The process seeks consensus only after discussions have fully explored
issues and interests and significant effort has been made to find creative
responses to differences.

Outcome criteria
• The process produces a high-quality agreement.
• The process ends stalemates.
• The process results in learning and change that go beyond the
immediate group.
• The process creates social and political capital.
• The process sets in motion a cascade of changes in attitudes, behaviors
and actions, and new practices or institutions.
• The process results in institutions and practices that are more flexible
and networked, allowing the community to more readily adapt to change
and conflict.

As far back in 2002, the Royal Commission for Riyadh City issued what is
believed to be among the first traffic safety strategies in the Middle East. Further
work during the decade which followed produced strategies in the Eastern
Province and nationally. Results from the Eastern Province from 2012 onwards
demonstrated that with effective action and partnership working across a range
of road safety disciplines, very significant casualty reduction can be achieved.

Under the NTP, a key pillar is enhancing standards of living, and within this
Enhancing Traffic Safety has been identified as a key activity. The Ministerial
Committee for traffic Safety has developed an execution strategy for the activity
in three subsets:

1. Prevention of Road Traffic Crashes

2. Reduction of Injury Severity

40
 3. Enhancing Post Crash Care

Saudi Arabia has set a target to reach a rate of 23 road deaths per 100,000
population by 2020, but this target had already been surpassed by the third
quarter of 2019, with a rate of 17.67 per 100,000 population. With a superior
plan can be achieved our target as planned. In the following chart, we show the
number of deaths and serious injuries for a specific period from 2004 to 2019.

Figure 23: Shows the enhancement in deaths and injuries number

41