14. A machine learning-based solution approach for solving the sustainable biomass supply chain network design problem
Contributed abstract in session FB-2: Machine learning, stream Machine learning.
Friday, 10:30 - 12:00Room: M228
Authors (first author is the speaker)
| 1. | Pinar Yunusoglu
|
| Industrial Engineering, İzmir Bakırçay University | |
| 2. | Fehmi Burcin Ozsoydan
|
| Industrial Engineering, Dokuz Eylul University | |
| 3. | Bilge Bilgen
|
| Industrial Engineering, Dokuz Eylul University |
Abstract
Sustainable supply chain management considers integrations of economic objectives and environmentally and socially responsible practices in each supply chain stage to minimize negative impacts on the environment and people. This study addresses a real-world sustainable biomass supply chain network design (BSCND) problem that handles strategic decisions (i.e., facility location) and tactical decisions (i.e., biomass sourcing and allocation, production planning, inventory management and the specific constraints related to BSCND problem) simultaneously. A machine learning-based solution approach based on a clustering methodology is developed to solve the BSCND problem. Initially, the k-means clustering algorithm is used to reduce the complexity of the problem. Later, an optimization model is solved to provide efficient operation and design of the supply chain. The proposed machine learning-based solution approach has achieved the optimal solution of the real-world BSCND problem in a reasonable computation time. Furthermore, the results of the computational experiments conducted on generated test instances indicate that the proposed machine learning-based solution approach is both effective and efficient in handling the computational complexity of the problem.
Keywords
- Logistics and supply chain management
- Machine learning
- Mixed integer programming
Status: accepted
Back to the list of papers