Parti bölmeli esnek atölye çizelgeleme problemi için kısıt programlama tabanlı çözüm yaklaşımı

This thesis studies the Flexible Job Shop Scheduling and Lot Streaming Problem (FJSSP-LS) with additional manufacturing resources other than machines. The related work in the literature reveals the influence of lot streaming for time-based objective functions. However, no study simultaneously considers the setup and transport resource restrictions when making scheduling decisions, which eventually limits the applicability of the resulting schedule in practice. Therefore, this thesis considers the FJSSP-LS with setup and transport activities and specific resources required by them. The aim is to achieve a schedule that minimizes the makespan. This problem is among the most constrained scheduling problems. Therefore, we propose a novel Constraint Programming (CP) model enriched by an efficient variable and value ordering strategy specifically designed for the problem. We also present a CP-based iterative improvement method, CP-based Large Neighborhood Search (CP-based LNS), that focuses on exploring the large neighborhoods through the CP model. We generate small-, medium-, and large-sized instances for the problem under study. The computational study shows that the proposed CP-based LNS improves the objective function value for the medium- and large-sized instances by 4.68 percent on average compared to the CP model.