Adaptive multiple crossover genetic algorithm to solve workforce scheduling and routing problem

Martínez-Gavara, Anna; Algethami, Haneen; Landa-Silva, Dario

Abstract

The Workforce Scheduling and Routing Problem refers to the assignment of personnel to visits, across various geographical locations. Solving this problem demands tackling numerous scheduling and routing constraints while aiming to minimise the operational cost. One of the main obstacles in designing a genetic algorithm for this problem is selecting the best set of operators that enable better performance in a Genetic Algorithm (GA). This paper presents an adaptive multiple crossover genetic algorithm to tackle the combined setting of scheduling and routing problems. A mix of problem-specific and traditional crossovers are evaluated by using an online learning process to measure the operator's effectiveness. Best performing operators are given high application rates and low rates are given to the worse performing ones. Application rates are dynamically adjusted according to the learning outcomes in a non-stationary environment. Experimental results show that the combined performances of all the operators works better than using one operator in isolation. This study makes a contribution to advance our understanding of how to make effective use of crossover operators on this highly-constrained optimisation problem.

Citation

Martínez-Gavara, A., Algethami, H., & Landa-Silva, D. (2018). Adaptive multiple crossover genetic algorithm to solve workforce scheduling and routing problem. Journal of Heuristics, 25(4-5), 753-792. https://doi.org/10.1007/s10732-018-9385-x

• Journal Article Type: Article
• Acceptance Date: Jun 22, 2018
• Online Publication Date: Aug 2, 2018
• Publication Date: Oct 1, 2018
• Deposit Date: Jul 20, 2018
• Publicly Available Date: Aug 3, 2019
• Journal: Journal of Heuristics
• Print ISSN: 1381-1231
• Electronic ISSN: 1572-9397
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• Volume: 25
• Issue: 4-5
• Pages: 753-792
• DOI: https://doi.org/10.1007/s10732-018-9385-x
• Public URL: https://nottingham-repository.worktribe.com/output/940269
• Publisher URL: https://link.springer.com/article/10.1007/s10732-018-9385-x
• Contract Date: Jul 20, 2018