Creating a Multi-Iterative-Priority-Rule for the Job Shop Scheduling Problem with Focus on Tardy Jobs via Genetic Programming

Genetic programming is used to create Priority Rules (PR) for the Job Shop Scheduling Problem with the aim of minimizing the weighted sum of tardy jobs. Four types of structures are used for the PR: Normal PR, Iterative-PR (IPR), Multi-PR (MPR), and Multi-Iterative-PR (MIPR). These are then compared among one another and with classical PR like shortest-processing-time. A modern metaheuristic based on local search using disjunct graphs and critical paths is used to solve the static problem as a benchmark. The results show that all types provide better results than classical PR and that with and without time limit the types from best to worst are: MIPR, MPR, IPR, and PR. The gaps to the metaheuristic are also reported.