THE USE OF GENETIC ALGORITHMS IN JOB SHOP SCHEDULING PROBLEMS
Loading...
Date
2024
Journal Title
Journal ISSN
Volume Title
Publisher
ECOLE NATIONALE SUPERIEURE DE TECHNOLOGIE ET D’INGENIERIE - ANNABA
Abstract
This dissertation investigates the optimization of job shop scheduling problems (JSSP) through the application of genetic algorithms (GAs). JSSP, a critical challenge in production management, involves sequencing jobs on machines to minimize the makespan, which is the total time required to complete all jobs. This research focuses on exploring various GA methodologies, including encoding schemes, selection strategies, and genetic operators.
The study begins with an in-depth analysis of the theoretical aspects of GAs, emphasizing the internal mechanisms that enhance their efficiency and performance. Each GA method is evaluated for its effectiveness in addressing the complexities of JSSP, with a detailed examination of their strengths and limitations. Through theoretical frameworks and illustrative examples, the dissertation compares different GA approaches, highlighting their suitability for optimizing job shop processes.
To provide practical insights, a simplified JSSP example is presented, and solved using multiple GA methods. The results are analyzed to determine the makespan and overall efficiency of each method. The findings indicate that hybrid GA approaches offer significant improvements in optimizing JSSP, achieving optimal or near-optimal solutions more consistently. This research underscores the value of GAs in complex scheduling environments, providing a foundation for further exploration and application of advanced optimization techniques in production management. The dissertation concludes with recommendations for future research directions and potential improvements in GA methodologies for JSSP.