An intelligent matching recommendation algorithm for a manufacturing capacity sharing platform with fairness concerns

Lei Xie, Jianghua Zhang, Qingchun Meng, Yan Jin, Weibo Liu*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)
27 Downloads (Pure)

Abstract

A supply and demand mismatch, or imbalance of the amount of supplies in the market, is always an issue and can happen all the time. Capacity sharing is an effective way to address this problem, and the capacity sharing platform facilitates the optimal matching between multiple capacity buyers and sellers. In the context of Industry 4.0, many industries are adopting intelligent algorithms to assist in decision-making. This paper presents an optimal or near-optimal matching algorithm to cope with a large volume of capacity sharing problems. The fairness of the matching solution is captured by including three objectives from platform, sellers and buyers. In this paper, a 2-dimensional crossover and an order-first mutation are developed and employed with genetic algorithms (GA), including GA and NSGA-II. Additionally, a novel repair mechanism is proposed by considering various constraints to transform infeasible solutions into feasible ones. Two matching schemes are studied based on whether orders from buyers can be split or not. The results show that both algorithms based on traditional GA and NSGA-II are effective for different schemes. In addition, it is found that GA has better performance in the case of “more sellers” and NSGA-II shows better performance in the “more buyers” case.
Original languageEnglish
Number of pages19
JournalInternational Journal of Production Research
Early online date15 Dec 2022
DOIs
Publication statusEarly online date - 15 Dec 2022

Keywords

  • Capacity Sharing
  • Genetic algorithm
  • Multiple objectives
  • match recommendation
  • fairness concern

Fingerprint

Dive into the research topics of 'An intelligent matching recommendation algorithm for a manufacturing capacity sharing platform with fairness concerns'. Together they form a unique fingerprint.

Cite this