Solving Single Machine Total Weighted Tardiness Problem with Unequal Release Date Using Neurohybrid Particle Swarm...

Area of Science:

• Operations Research
• Computational Intelligence
• Manufacturing Systems

Background:

• The single machine total weighted tardiness problem (SMTWT) with unequal release dates is a complex scheduling challenge.
• Existing optimization methods may struggle with the computational demands of this problem.

Purpose of the Study:

• To develop and evaluate a novel neurohybrid optimization system for the SMTWT problem.
• To enhance solution quality and efficiency by integrating multiple metaheuristic algorithms.

Main Methods:

• A neurohybrid system was developed, utilizing particle swarm optimization (PSO) as the primary optimizer.
• Genetic algorithms (GA) and simulated annealing (SA) were employed as sub-hybrid components and local search tools.
• The neurodominance rule (NDR) was incorporated to further refine solutions by analyzing sequential job tardiness.

Main Results:

• The integrated neurohybrid-PSO system demonstrated improved performance in solving the SMTWT problem.
• The combination of PSO, GA, and SA, along with NDR, effectively addressed the complexities of unequal release dates and weighted tardiness.

Conclusions:

• The proposed neurohybrid-PSO solution system offers a robust and effective approach for the single machine total weighted tardiness problem.
• Hybrid metaheuristic strategies show significant potential for optimizing complex scheduling problems in manufacturing and operations research.