A fuzzy logic-based machine learning algorithm for product distribution in supply chains considering rival⇔s strategic decisions

Delgoshaei A., MohammadAzari M., Hanjani S. E., Fard F., Beigizadeh R., Aram A. K.

International Journal of Industrial Engineering : Theory Applications and Practice, vol.27, no.6, pp.933-958, 2020 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 27 Issue: 6
  • Publication Date: 2020
  • Journal Name: International Journal of Industrial Engineering : Theory Applications and Practice
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Compendex
  • Page Numbers: pp.933-958
  • Keywords: Fuzzy logic, Game theory, Machine learning, Multi-layer perceptron, Supply chain management
  • Ankara University Affiliated: Yes

Abstract

c INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERINGA new fuzzy-based machine learning method is addressed in this research to respond to rival’s strategy. This method aims to find the best production and product distribution strategy while rivals can take various market strategies that affect a market quota. A multi-period mixed-integer programming method is developed for scheduling a supply chain over a time horizon. The developed model is flexible enough to use in industries. To solve the problem, we developed a hybrid fuzzy-based multi-layer perceptron and simulated annealing algorithm. Its results are compared with branch and bound, hybrid Tabu Search and Simulated Annealing algorithms, hybrid ant colony optimization, and simulated annealing algorithms. A new measuring index is developed to evaluate the production strategies in dynamic market demands. The outcomes reveal that while product demands are considered stochastic, it may affect the market quota among suppliers. Comparing different game theories shows that the proposed method can successfully generate effective production strategies while rivals change their approach.