CUSTOMER ORDER SCHEDULING PROBLEM: REVIEW OF THE STATE OF THE ART

Usually, scheduling relates to the sequencing and timing of jobs in the production system to optimize some performance measures. The most used are makespan, total completion time, and tardiness. Such measures are related to the single job to be produced. However, in a customer-oriented production, the performance measures to optimize and evaluate are not related to the single job, but to customer orders. A customer order may include a set of jobs, and only the status of the complete order is relevant for the customer, and hence for the company. The customer order scheduling problem, then, differs from the standard scheduling problem only in the performance measures to be optimized. It considers indeed performance measures related to the customer orders, and not to a single job. The purpose of the thesis is to gain insight into the research domain, summarize the state of the art by tracking trends over time, and enlighten the gaps in the literature that could be a potential interest for future research on Customer Order Scheduling Problem. The literature has been focused mainly on single stage dedicated parallel machines, the most used methodology is heuristics, and the efficiency of the proposed algorithms was tested mostly with computational experiments. The AI integration with the metaheuristics and a cooperation with Industrials players could improve the quality of solutions and align the research to the real-world by introducing more constraints and multi-objective functions (today not much frequent).