EURO 2025 Leeds
Abstract Submission

2991. Constraint Programming Approach for Customer Order Scheduling Problem with Sequence-Dependent Setup Times

Invited abstract in session TB-12: Emerging trends, challenges and innovations in scheduling and project management, stream Scheduling and Project Management.

Tuesday, 10:30-12:00
Room: Clarendon SR 1.02

Authors (first author is the speaker)

1. Funda Guner
Industrial Engineering, Ankara Yıldırım Beyazıt University
2. John H. Drake
Computing and Mathematical Sciences, University of Leicester

Abstract

The customer order scheduling (COS) problem is a fundamental scheduling problem where customer orders consist of one or more jobs that must be processed and delivered together. Efficient scheduling of customer orders is crucial for minimizing delays and optimizing resource utilization. In this study, we consider sequence-dependent setup times, where a setup is required before processing each job unless it is the first job in an order and is the same as the last job of the previous order.

We focus on minimizing the total completion time by exploring mixed-integer linear programming (MILP) and constraint programming (CP). MILP is a well-established mathematical modeling approach, while CP is known for efficiently handling complex constraints and combinatorial structures.

We compare the performance of MILP and CP to evaluate their effectiveness in solving the COS problem with sequence-dependent setup times. CP is particularly effective in solving highly constrained instances, while MILP can provide optimal solutions but struggles with scalability for larger instances.

This study contributes to the literature by offering a direct comparison of MILP and CP for COS with sequence-dependent setup times. CP models are easier to modify, making them suitable for dynamic environments. Incorporating setup times adds complexity, requiring efficient scheduling methods. Future research could explore hybrid methods that combine MILP and CP to enhance performance.

Keywords

Status: accepted

Back to the list of papers