A computational study on a time-sensitive multiobjective flexible job shop scheduling problem

---

Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Endüstri Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2017

Öğrenci: CANER OĞUZKAN

Danışman: BAHAR ÇAVDAR

Özet:

In this thesis we focus on a time-sensitive flexible job shop scheduling problem. In time-sensitive systems, the main concern is more on the total completion time, which includes time spent during the computation-only phase and the implementation of the solution, rather than finding a solution which will take the least time to implement. However, the conventional solution approaches do not directly address the computation time. In this study, we employ a Computation-Implementation Parallezation (CIP) approach, which was originally introduced for routing problems, in order to find more time-efficient solutions in a multiobjective flexible job shop scheduling problem. This is the first study to implement the CIP approach on a multi-objective problem. Moreover, we implement the CIP approach on a problem where there are precedence relations between the operations. Therefore, our results provide a further understanding of the benefits of CIP under the tradeoff between different objectives, and also the limitations of embedding the computation time into the implementation in a more challenging problem setting. We perform extensive computational experiments on many different scenarios based on different instance sizes, flexibility of the processing tools, processing times of operations and compare a base solution method with its CIP implementation. Our results show that CIP approach can provide considerable improvement in the solution quality without increasing the computation-only time or we can find better solutions using the same computation-only time in a multiobjective flexible job shop scheduling problem.