Introduction to Engineering

Data Analysis

Eng’r. Ma Liezl Gallardo, Ph.D., PIE

Importance of Data Analysis in Engineering
Informed Decision- Optimization and Innovation and Predictive
Making Efficiency Design Maintenance

Data analysis empowers Identifying patterns and Data analysis fuels Analyzing sensor data can
engineers to make insights can lead to innovation by enabling predict equipment
informed decisions based process optimization, engineers to understand failures, allowing for
on objective evidence and improved efficiency, and customer needs, test new proactive maintenance
trends. reduced costs. concepts, and create and preventing downtime.
better products.
Types of Engineering Data
Numerical Data Categorical Data
Quantitative measurements, such as temperature, pressure, Qualitative data that can be grouped into categories, such
and speed. as material type or manufacturing process.

Time Series Data Geospatial Data

Data collected over time, such as sensor readings or market Data that is linked to a specific location, such as GPS
trends. coordinates or geological information.
Data Collection Techniques

Sensors Cameras
Measuring physical parameters, Capturing visual information,
such as temperature, pressure, or such as images or videos for
vibration. analysis.

Surveys and Databases and APIs

Questionnaires
Accessing existing data from
Gathering user feedback, various sources, such as company
preferences, and opinions. records or public databases.
Fundamentals of Data
Preprocessing
1 Data Cleaning: Removing errors, inconsistencies, and
missing values.

2 Data Transformation: Converting data into a suitable

format for analysis, such as normalization or
standardization.

3 Data Scaling: Adjusting the range of data values to

ensure equal contribution of different variables.
Exploratory Data Analysis
(EDA)
Understanding Data Structure
Examining the data's organization, relationships, and
distributions.

Identifying Patterns and Trends

Revealing hidden patterns, outliers, and potential
correlations.

Formulating Hypotheses
Generating hypotheses about the data and potential
relationships.
Visualizing Engineering Data
Descriptive Statistics

1 2
Mean Median
Average value of a dataset. Middle value in a sorted dataset.

3 4
Standard Deviation Quartiles
Measure of data dispersion Dividing data into four equal
around the mean. parts, revealing distribution
patterns.
Inferential Statistics

Hypothesis Testing
1 Evaluating hypotheses about the population based on sample data.

Confidence Intervals
2
Estimating a range of plausible values for a population parameter.

Statistical Inference
3 Drawing conclusions about the population based on
sample data.
Regression Analysis
Linear Regression
1
Modeling the relationship between variables with a straight line.

Polynomial Regression
2
Modeling the relationship between variables with a curved line.

Multiple Regression
3 Modeling the relationship between multiple independent
variables and a dependent variable.
Classification Techniques
Decision Trees Support Vector Machines Logistic Regression
(SVMs)
Tree-like structures for classifying Predicting the probability of
data based on a series of rules. Finding the optimal hyperplane to belonging to a specific class based on
separate data points into different input variables.
classes.
Clustering Methods

K-Means Clustering Hierarchical Clustering

Partitioning data into k clusters based on distance from Building a hierarchy of clusters by iteratively merging or
cluster centroids. splitting data points.
Time Series Analysis
1 Trend Analysis 2 Seasonality Analysis
Identifying the overall Detecting recurring
direction of the time series patterns at regular
data. intervals, such as daily or
monthly fluctuations.

3 Forecasting
Predicting future values of the time series based on historical
data.
Optimization Techniques
1 Linear Programming
Optimizing a linear objective function subject to linear
constraints.

2 Genetic Algorithms
Using evolutionary principles to find optimal solutions
to complex problems.

3 Simulated Annealing
A metaheuristic optimization technique inspired by the
annealing process in metallurgy.
Decision-Making with Data
Data-Driven Insights
Identifying actionable insights from data analysis.

Risk Assessment and Mitigation

Using data to assess potential risks and develop mitigation strategies.

Informed Decision-Making
Making decisions based on data-driven insights, leading to better outcomes.
Ethics and Bias in Data Analysis
Fairness and Equity Privacy and Confidentiality Transparency and
Accountability
Ensuring that data analysis results Protecting sensitive data and
are unbiased and fair to all respecting the privacy of Providing clear explanations and
stakeholders. individuals. justifications for data analysis
processes and results.
Tools and Software for Engineering Data
Analysis
Python R MATLAB Tableau

A powerful and versatile A statistical programming A technical computing A data visualization

language with extensive language widely used for environment for software for creating
data analysis libraries. data analysis and numerical computation, interactive dashboards
visualization. visualization, and and reports.
programming.
Case Studies in Engineering Data Analysis

Wind Turbine Optimization Smart City Planning

Data analysis is used to optimize wind turbine performance, Data analysis helps create smarter and more sustainable
maximizing energy generation and reducing costs. cities by optimizing traffic flow, improving public
transportation, and managing resources.
Best Practices for Effective
Data Analysis
Define Clear Ensure Data Quality
Objectives
Clean and preprocess data
Start with a well-defined to ensure accuracy and
goal for the analysis. reliability.

Choose Appropriate Communicate Results

Techniques Clearly
Select the right statistical Present findings in a way
and analytical tools for the that is understandable and
task. actionable.
Conclusion and Future Directions
Engineering data analysis is a transformative field that empowers engineers to make better decisions, optimize processes,
and drive innovation. As data collection and analysis capabilities continue to advance, we can expect even more exciting
applications and breakthroughs in the future.