OPTIMIZATION OF PRODUCTION PROCESS USING

ARTIFICIAL INTELLIGENCE (AI)

Prepared by

Muhammad Talha Sarour Shah

Student ID: 23858125

DATE:
 OPTIMIZATION OF PRODUCTION PROCESS USING
ARTIFICIAL INTELLIGENCE (AI)

A proposal submitted in partial fulfilment of the requirements for the

Degree of

MSc Advanced Design and Manufacturing

University of Northampton

Student Name

Muhammad Talha Sarour Shah

Student ID: 23858125

Under the guidance of Supervisor

Dr Faris Elasha

Senior Lecturer in Mechanical Engineering and Design,

. University Of Northampton

Date of Submission
 INTRODUCTION
Optimization of production processes using Artificial Intelligence (AI) is transforming industries by
enhancing efficiency, reducing cost and improving product quality. AI driven tools like machine learning,
neural networks, and reinforcement learning analyze vast amount of real time data from production
lines to identify inefficiencies, predict equipment failures and optimize resource allocation. Unlike
traditional methods, AI continuously adapts to changing conditions, enabling dynamic scheduling,
minimizing waste and automating decision making. By integrating AI into manufacturing, businesses can
achieve smarter, faster and more sustainable operations, ultimately gaining a competitive edge in
today’s fast paced industrial landscape.

Strategies and Insight

Nowadays, businesses are using artificial intelligence (AI) more and more to improve their production
processes. By optimizing production, companies can work more efficiently, save money, and make better
[Link] tools like machine learning, predictive analytics, and automation help organizations analyze
large amounts of data, find patterns, and make smart choices to make operations smoother. This
method includes different parts of making products, like keeping track of inventory and organizing how
things are delivered, and uses current information to adjust to what people want to buy. When
businesses use AI in making things, they can reduce waste, forecast when things need fixing, and manage
resources better, which helps them make more products and money in the end. As industries keep
changing, it's important to know how to use AI strategies to make production better for growth and
staying ahead of the competition. This study will look at the ways, tools, and examples that show how AI
can improve production processes.

Background and History

In the past few years, AI has boosted manufacturing by making progress in technology, data analysis, and
machine learning. Early Development (1960sto1980s) 1. Operations Research: Back in the middle of the
20th century, people came up with ways like linear programming and simulation to tackle tricky logistics
and manufacturing issues. 2. Computeraided manufacturing (CAM): The 1970s and 1980s saw the arrival
of computeraided design (CAD) and CAM systems. These systems helped smooth out the manufacturing
process setting the stage for future AI. The rise of artificial intelligence (1980sto1990s1. Expert systems:
Companies started using expert systems in the 1980s to capture human knowledge in software for
production-related decisions.
Data-driven approach: As power consumption grew, companies began gathering extensive data from
production processes, setting the stage for intelligent vehicle use in the future. The rise of machine
learning (1990s to 2010s). Progress in machine learning: Machine learning algorithms like neural
networks and decision trees gained significance in the 1990s and early 2000s. People began applying
these methods to boost production by forecasting machine maintenance needs and assessing product
quality. Big Data: The emergence of big data in the 2000s enabled companies to utilize vast amounts of
operational data helping them boost processes through AI. Integration and Industry 4.0 (2010s-
Present). Industry4.0. In the early 2010s, the concept of Industry 4.0 aimed to combine IoT AI, and data
analysis in manufacturing. This led to intelligent factories where AI enhances supply chain management
and implements production changes in real-time.

The Role of AI in Production Process Optimization

AI is used in manufacturing for various processes like design, production, quality control, and
maintence. One important aspect is the use of AI-powered design tools to improve product
development and performance. Research by Rane, Choudhary, and Rane [9], shows how 4D/5D/6D
Printing speeds up prototype development and how IoT improves product functionality by monitoring
in real-time and making data-driven design enhancements.
AI is essential in production for optimizing processes, maintaining equipment, and managing the
supply chain. AI-driven robotics and automation are changing factory operations, making
manufacturing more precise and [Link] helps improve productivity by analyzing data to enhance
production and resource use. Predictive maintenance, thanks to AI algorithms, decreases downtime
and extends equipment life, saving money .AI is also excellent for quality control, ensuring products
meet high standards and lowering the risk of defect. Moreover, AI improves decision-making by
providing up-to-date insights using data analysis, helping manufacturers improve inventory
management, predict demand, and allocate resources efficiently. Also, working together with human
operators, collaborative robots, or cobots, create safer and more comfortable workspaces, showing
how AI and human labor can work together seamlessly.
 Justification of the problem
Industries in different sectors are constantly working on improving their production processes. Many
industries still use old-fashioned ways like manual monitoring and linear modeling, which can lead to
problems like high costs and not using resources efficiently. Because of competition and the need for
better products, industries have to find ways to make production better and waste less. But figuring
out where things are slowing down, predicting changes in demand, and using resources in the best
way are still big challenges for industries trying to be the best they can be. Artificial Intelligence (AI)
can help solve these issues by allowing for data-driven decision-making, predictive maintenance, and
process automation. AI can analyze large amounts of data and provide immediate insights, leading to
better production processes with fewer mistakes and improved accuracy. Using AI in production can
reduce downtime, enhance product quality, and boost efficiency. New research shows that using AI to
improve operations can lower costs by 10-20% and increase productivity by 15-30% (McKinsey, 2020).
As AI gets better, more businesses are likely to start using it for things like predicting trends,
forecasting demand, controlling processes, and ensuring quality. If companies don't start using AI,
they could end up lagging behind rivals who are using it to make their production processes more
efficient (Deloitte, 2021) Today, industries are under pressure to do better in terms of productivity,
costs, and quality.
Old-fashioned ways of making things, which rely a lot on people watching and deciding things, often
cause problems like delays and inefficiencies. Some common problems are machines not working,
not managing supplies well, wasting materials, and not always keeping up quality.
Modern production is so complicated that it's hard for people to always watch and fix things right
away to keep things running smoothly. Artificial intelligence (AI) has the potential to improve
production processes by analyzing data in real-time, predicting maintenance needs, and making
decisions autonomously. AI systems can analyze large amounts of data from machines, find
inefficiencies, and offer suggestions for improvements that humans might miss. For instance, AI can
reduce equipment downtime by predicting maintenance requirements, identify defects early, and
adjust production schedules to meet demand. Studying this field is important because using Al to
make improvements can help businesses work better, save money, and produce less waste.
Businesses that don't use Al in their processes could end up losing to competitors who are more
tech-savvy and flexible.
A report by McKinsey says that companies in manufacturing that use Al can make their work 20%
more productive and cut maintenance costs by 10-40%.
LITERATURE REVIEW

1. The paper discusses how artificial intelligence ai and machine learning ml serves as key
drivers for process optimization within the industry 4.0 framework. It underscores the fact
that efficiency, and decision making can be facilitated with technology, but not replaced.
People are crucial to the effective functioning of AI and ML as they give the data context,
creativity, and governance. They also say the research underscores the ability of machines
and humans to work as a team, with the best results following from a balance between
machine learning and human intuition — rather than leaving decisions up to one method or
 the other. Ultimately, it is for this reason that the book argues for a view of AI as part of an
integrated whole with workforce as against adopting 'either' or 'notion at all'[1]
2. This paper attempts to analyze the possibilities of using AI-powered techniques for real-time
process monitoring and quality control in manufacturing community. They highlight AI for
analyzing production process data to detect defects, anticipate failures and leverage
workflow optimization. Manufacturers can benefit from better performance, less waste and
higher quality products with the implementation of machine learning algorithms into their IoT
(Internet of Things) devices. The paper said case studies provided in the study illustrate large
efficiency improvements and cost savings through AI, which together amount to a reinvention
of legacy manufacturing practices. The bottom line from BDO is that manufacturers need to
introduce AI into how they work — otherwise the currently-whirling field will leave them cast
off in its wake.[2]
3. At MOXAI, we work on improving manufacturing processes with model agnostic explainable
AI and data-driven process tuning. This aims to boot production efficiency and quality without
involving actual models of machine learning. Transparency & Tractability — MOXAI creates a
window into the decision-making process of AI, enabling visibility and trust leading to insights
in order for manufacturers to properly-manage their operations. It brings together cutting
edge analytics and real-world applications, making it easier for businesses to adapt and
evolve in a continuous loop [3].
4. Here, the paper gives a brief but detailed overview of how artificial intelligence (AI) arises for
process optimization of additive manufacturing. It discusses different AI techniques such as
machine learning, and neural networks proving how they can help to make design, process
parameters, and quality control better. In this review, the authors organize and describe
existing research, outline major challenges in deployment, and report future outlooks toward
AI integration in additive manufacturing. These are some good and efficient applications of AI
that provide assistance in even maintaining the cost, an infographic to give you a better view
about how fast-paced this field is becoming and what all work can be done with the help of
Artificial Intelligence as well it shows potential that AI has in order to improve efficiency, cut
costs and drive innovation, overall.[4]
5. This paper introduces a new programmable Auto ML methodology for SME (small and
medium- sized enterprises) manufacturers, to optimize the production process. It describes
how Auto ML can help with the operationalization of data science and advanced analytics by
automating model selection, hyperparameter tuning, and feature engineering. The
methodology seeks to increase production efficiency, lower operational costs and offer
administrative support to the decisions at SMEs level that rarely have resources for large size
of data science functionalities. In this way, the study shows that SMEs could use data-driven
insights to effectively enhance their manufacturing processes if provided with accessible tools
and frameworks.[5]
6. In this paper, the ways by which manufacturers are deploying Artificial Intelligence (AI) is
discussed. It emphasizes more on certain applications like predictive maintenance, quality
control, supply chain optimization as well as process automation. With the help of artificial
intelligence (AI), manufacturers can increase efficiencies, minimize downtime, and deliver
better product quality. The paper also examines some of the issues affecting the uptake of AI,
like governance of data and labor up skilling. Namely, it highlights what a difference AI can
make in manufacturing, allowing firms to keep up with the pace of an always-in-flux market.
[6]
 7. This paper investigates the performance of artificial intelligence (AI) techniques in modelling
and simulating production systems based upon optimization requirements. In the area of
intelligent manufacturing, as one of the methods to improve modeling and analysis
capabilities for complex manufacturing process is among the AI such as machine learning,
genetic algorithms The study shows how these tools help in making better decisions,
increased operational efficiency and lowering production costs. It also makes clear that AI will
need to be combined with conventional simulation techniques if the results are to improve.
Overall, the paper demonstrates how AI can revolutionize production system optimization
making it more efficient and able to adapt more easily to variations in demand.[7]

8. This paper presents an extensive literature review on the application of Artificial intelligence
(AI) in manufacturing operations. The authors break the use of AI into four main buckets:
Predictive maintenance Quality control Process optimization and supply chain management.
By analyzing and comparing these articles, the review showed that AI could be used to
improve efficiency, decrease costs, and increase product quality. The story also breaks down
obstacles, such as integrating data into processes, changes with the workforce and
implementing technology. In sum, the paper highlights how AI may disrupt current
manufacturing practices and outlines future directions and research opportunities in modern

manufacturing enabled by AI.[8]

Methodology
The next thing we need to do in our research, is to analyze the articles and books written about
Artificial Intelligence, process optimization, Industry 4.0, and process operator. This will help us
gather more evidence to support our conclusions. For this research the methodology will employ a
mixed method approach combining both qualitative and quantitative techniques to explore the
optimization of production process using AI.
Methodology is broken down into the following stages:
1. Literature Review: Will be conducted to examine existing AI techniques, framework, and their
applications in production system.
2. Case Studies: From various industries such as automotive and electronics, will be analyzed to
understand real world AI implementations in production environments.
3. Industry Experts Thoughts: will provide qualitative insights into challenges and best practices
in AI adoption.
4. ML Models and Optimization Algorithms: for the quantitative aspects machine learning
models and optimization algorithms will be developed and tested through simulations.
5. Simulations: These simulations will allow for experimentation with different AI driven
approaches to process optimization, focusing on production efficiency, cost reduction and
waste minimization. The AI models will be validated by comparing their performance with
traditional optimization techniques and if feasible a real world validation in an industrial
setting will be conducted to assess their practical impact.
 OBEJECTIVES
1. Automate Repetitive Tasks: Use AI to deliver faster results by eliminating human error and
speeding up repeatable processes.

2. Predictive Maintenance: By utilizing AI powered predictive maintenance to reduce

equipment downtime and improve the longevity of machinery.

3. Cost Reduction: Turn to AI for finding areas in which money can be saved from using
resources more efficiently and generating less waste.

4. Demand Forecasting: Better predict consumer demand using AI algorithms so you know
when to reach optimal stock and production scheduling levels.

5. Quality Control: Improve quality control through AI-based visual inspection systems that
can automatically spot defects and ensure product consistency.

AI Tools
In the research literature focused on using AI to improve the manufacturing process, various AI
tools and techniques are frequently used. These tools help you make decisions, increase efficiency,
and reduce costs. Here are some useful tools and techniques:

1. Machine Learning (ML) Algorithms

2. Genetic Algorithms (GA)
3. Neural Networks (NN) and Deep Learning
4. Optimization Techniques
5. Fuzzy Logic
6. Digital Twin Technology
7. AI-Powered Data Analytics Tools
8. Natural Language Processing (NLP)
9. Robotic Process Automation (RPA)
10. Multi-Agent Systems (MAS)

RESEARCH SCHEDULE
A tentative research plan is given in the table below:

SR Activities Date Duration

.
 1. Literature Survey Sep 1-Sept 25 25

2. Identifying AI tools Oct 1-Nov 1 30

3. Running simulation Nov 1-Nov 15 15

4. Research on Industries Nov 15-Jan 1 45

5. Final Report Writing Jan 1- Jan25 25

REFERENCES
[1] Artificial Intelligence and Machine Learning as a Process Optimization driver under
Industry 4.0 framework, the role of the people in the process by Federico Walas Mateo La
Universidad Nacional Arturo Jauretche GLOBCER’21, 2-4 June 2021.
[2] March 2024 Journal of Advances in Mathematics and Computer Science (AI-Driven
Real-time Quality Monitoring and Process Optimization for Enhanced Manufacturing
Performance) 39(4):81-89 Authors: Olanrewaju Morayo Okuyelu California State
University, Northridge.
[3] MOXAI – Manufacturing Optimization through Model-Agnostic Explainable AI and
DataDriven Process Tuning Clemens Heistracher1*, Anahid Wachsenegger2*, Axel
Weißenfeld2 , and Pedro Casas2 1 craftworks GmbH, Vienna, Austria
[Link]@[Link] 2 AIT Austrian Institute of Technology, Vienna, Austria.
[4] Optimization with artifcial intelligence in additive manufacturing: a systematic review
Francesco Ciccone1 · Antonio Bacciaglia1 · Alessandro Ceruti1 Received: 12 August 2022 /
Accepted:
12 April 2023 / Published online: 12 May 2023 © The Author(s) 2023.
[5] Yarens J. Cruz, Alberto Villalonga , Fernando Castano, Marcelino Rivas , Rodolfo E. Haber a a
Centro de Automatica ´ y Robotica, ´ CSIC-Universidad Polit´ecnica de Madrid, Madrid 28500, Spain b
Centro de Estudios de Fabricacion ´ Avanzada y Sostenible, Universidad de Matanzas, Matanzas
40100, Cuba.
[6] Key Findings and Insights from Manufacturers Kathryn Wengel Executive Vice President and
Chief Technical Operations & Risk Officer, Johnson & Johnson Chair of the Board, National
Association of Manufacturers.
[7] ARTIFICIAL INTELLIGENCE TOOLS IN SIMULATION AND OPTIMIZATION OF PRODUCTION
SYSTEMS Peter SKORIK, Milan GREGOR, Andrej STEFANIK BANKS, J. – CARSON, J.S. –
NELSON, B.L. – NICOL, D.M.: Discrete Event Systems Simulation. 3rd edition. Englewood
Cliffs, 2000.
[8] A review of artificial intelligence applications in manufacturing operations Siby Jose
Plathottam | Arin Rzonca | Rishi Lakhnori | Chukwunwike O. Iloeje Energy Systems and
Infrastructure Analysis Division, Argonne National Laboratory, Lemont, Illinois, USA.
[9] Rane N, Choudhary S, Rane J. Enhanced product design and development using Artificial
Intelligence (AI), Virtual Reality (VR), Augmented Reality (AR), 4D/5D/6D Printing, Internet of
Things (IoT), and blockchain.