Title: An integrated methodology based on machine-learning algorithms for biomass supply chain optimisation

Authors: Duy Nguyen Duc; Narameth Nananukul

Addresses: School of Management Technology, Sirindhorn International Institute of Technology, Thammasat University, Pathum Thani 12120, Thailand ' School of Management Technology, Sirindhorn International Institute of Technology, Thammasat University, Pathum Thani 12120, Thailand

Abstract: This paper presents an integrated methodology for biomass supply chain planning, using a stochastic optimisation model and machine-learning algorithms. A methodology that integrates machine-learning algorithms with the optimisation process was proposed in order to generate solutions for large-scale supply chain optimisation problems. Models based on artificial neural network (ANN) and Bayesian network were developed by using the knowledge from previously-solved problems, to define good starting points for the search for solutions in the optimisation process. With this novel approach, the search space can be reduced and optimal solutions found with a shorter runtime. The applicability of the proposed approach was evaluated with a case study relating to biomass supply chain planning in the Central Vietnam region. The results from the proposed framework reveal that the optimal biomass plan for biomass supply chain can be determined with accuracy up to 96%, with a decrease in runtime by 37.19% on average.

Keywords: machine learning; optimisation; relaxation induced neighbourhood search; biomass supply chain planning; artificial neural network; Naïve Bayes; Bayesian network.

DOI: 10.1504/IJLSM.2023.133521

International Journal of Logistics Systems and Management, 2023 Vol.46 No.1, pp.47 - 75

Received: 04 Feb 2021
Accepted: 29 Apr 2021
Published online: 19 Sep 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article