Loading...
Transformation of Automotive Preventive Maintenance to Predictive and Condition Based Maintenance
Al Bayaa, Laila Ayman
Al Bayaa, Laila Ayman
Date
2025-11
Author
Advisor
Type
Thesis
Degree
Citations
Altmetric:
Files
35.232-2025.68a Laila Ayman Al Bayaa.pdf
Adobe PDF, 2.73 MB
- Embargoed until 2027-03-23
Description
A Master of Science thesis in Engineering Systems Management by Laila Ayman Al Bayaa entitled, “Transformation of Automotive Preventive Maintenance to Predictive and Condition Based Maintenance”, submitted in November 2025. Thesis advisor is Dr. Mahmoud Awad and thesis co-advisor is Dr, Abdulrahim Shamayleh. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).
Abstract
Maintenance is an integral part of operations as it has impacts on performance, availability, and cost. To ensure effective maintenance, many companies opt for preventive maintenance as it lowers the chances of sudden failures. However, preventive maintenance has drawbacks of its own as it leads to the early replacement of parts during their useful life which leads to increased waste and extra costs. This is especially true in the automotive industry when it comes to engine oil and brake systems as these are the most frequently replaced parts in vehicles. Unnecessary replacement of oil and brake not only has cost implications but sustainability and health implications as it increases waste and exposes maintenance workers to harmful fumes. The objective of this study is to investigate the use of condition-based and predictive maintenance approaches for oil and brake systems in automotive fleets. Predictive and condition-based maintenance approaches leverage the use of data, sensors, and machine learning to forecast potential equipment or machinery failures, enabling timely maintenance interventions to avert costly unplanned downtime. Despite its advantages, predictive and condition-based maintenance are not widely accessible in key industries such as the automotive industry. The objective of this study is to propose a ML predictive maintenance method for automotive components. To test the proposed method performance, a case study with public transport buses in the emirate of Dubai was conducted. For brake systems, driving, and environmental data was used to predict the brake pads and discs wear out and schedule replacement visits. Decision Tree models successfully predicted disc wear with 96.1% accuracy while Nueral Networks were able to interpret the complexity of nonlinear relationships of pad wear and achieved a 90% accuracy. For engine oil, a correlation was found between viscosity and soot and hence soot will be used to evaluate the condition of the oil and decide whether to replace or run. Together, these initiatives are expected to save 9 million AED and save 175 tons of CO2 emission in the first year of adoption.