|
Hybrid analytical and simulation optimization approach for production and distribution supply chain planning |
|---|---|
| รหัสดีโอไอ | |
| Creator | Jirasak Ji |
| Title | Hybrid analytical and simulation optimization approach for production and distribution supply chain planning |
| Contributor | Jirasak Ji, Navee Chiadamrong |
| Publisher | Asia-Pacific Journal of Science and Technology |
| Publication Year | 2562 |
| Journal Title | Asia-Pacific Journal of Science and Technology |
| Journal Vol. | 24 |
| Journal No. | 3 |
| Page no. | 1-18 |
| Keyword | Production-distribution planning, Hybrid analytical-simulation optimization, Makespan, Safety stock |
| URL Website | https://www.tci-thaijo.org/index.php/APST |
| Website title | https://www.tci-thaijo.org/index.php/APST/article/view/177464/151322 |
| ISSN | 2539-6293 |
| Abstract | Supply chain planning consists of designing an optimal and feasible production and distribution plan for the whole supply chain. Traditionally, two common methods of optimization are analytical and simulation-based optimization, and each of them has pros and cons. In this paper, both methods are combined to consolidate the strengths of each, also known as the hybrid analytical and simulation approach. A case study of a multi-period, multi-echelon, and multi-product production and distribution problem that maximizes the whole supply chain's profit is introduced, to demonstrate the proposed hybrid approach. The analytical model is solved to find the optimal production-distribution plan, and then the plan is inputted into a simulation model, where uncertainties are incorporated. The proposed algorithm is then applied to identify a feasible plan that meets makespan limitation and service level requirements. Safety stock is incorporated to satisfy the service level requirements and maximize the supply chain's profit. This procedure continues iteratively until the production-distribution plan is feasible and optimized. The results show that the proposed approach can solve for a near or possibly optimal as well as feasible solution with relatively fast computational time. |
