A memetic NSGA-II for the multi-objective flexible job shop scheduling problem with real-time energy tariffs

Sascha Christian Burmeister*, Daniela Guericke, Guido Schryen

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

3 Citations (Scopus)
42 Downloads (Pure)

Abstract

Rising costs for energy are increasingly becoming a vital factor for the production planning of manufacturing companies. Manufacturers face the challenge to react to dynamic energy prices and design energy cost efficient schedules in their production planning. In the literature, the energy cost-aware Flexible Job Shop Scheduling Problem addresses minimization of both makespan and energy costs. Recent studies provide multi-objective approaches to model the trade-off of minimizing makespan and energy costs. However, the literature is limited to coarse-grained time periods and does not consider dynamic tariffs where costs change at short intervals, so that production schedules may fall short on energy costs. We aim to close this research gap by considering frequently changing real-time energy tariffs. We propose a multi-objective memetic algorithm based on the non-dominated sorting genetic algorithm (NSGA-II) with both makespan and energy cost minimization as the objectives. We evaluate our approach by conducting computational experiments using prominent FJSP-benchmark instances from the literature, which we supplement with empiric dynamic energy prices. We show results on method performance and compare the memetic NSGA-II with the results of an exact state-of-the-art solver. To investigate the trade-off between a short makespan and low energy costs, we present solutions on the approximated Pareto front and discuss our results.
Original languageEnglish
Number of pages41
JournalFlexible services and manufacturing journal
Early online date14 Nov 2023
DOIs
Publication statusE-pub ahead of print/First online - 14 Nov 2023

Fingerprint

Dive into the research topics of 'A memetic NSGA-II for the multi-objective flexible job shop scheduling problem with real-time energy tariffs'. Together they form a unique fingerprint.

Cite this