Adaptive and various learning-based algorithm for supply chain network equilibrium problems
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916691891453952 |
|---|---|
| author | He, Sheng-Xue |
| author_facet | He, Sheng-Xue |
| contents | This paper proposes a novel nonlinear programming model to capture the equilibrium state of complex supply chain networks. The model, equivalent to a variational inequality model, relaxes traditional strict assumptions to accommodate real-world complexities, such as nonlinear, non-convex, and non-smooth relationships between production, consumption, and pricing. To efficiently solve this challenging problem, we introduce a novel heuristic algorithm, the adaptive and various learning-based algorithm (AVLA), inspired by group learning behaviors. AVLA simulates individual learning processes within different subgroups at various stages, employing a success history-based parameter adaptation mechanism to reduce manual tuning. Extensive computational experiments on 29 benchmark problems and 5 supply chain networks demonstrate AVLA's superior performance compared to 19 state of the art algorithms. AVLA consistently achieves the best results in terms of both average and best objective function values, making it a powerful tool for addressing complex supply chain network equilibrium problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_11499 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adaptive and various learning-based algorithm for supply chain network equilibrium problems He, Sheng-Xue Optimization and Control 90 I.2.0 This paper proposes a novel nonlinear programming model to capture the equilibrium state of complex supply chain networks. The model, equivalent to a variational inequality model, relaxes traditional strict assumptions to accommodate real-world complexities, such as nonlinear, non-convex, and non-smooth relationships between production, consumption, and pricing. To efficiently solve this challenging problem, we introduce a novel heuristic algorithm, the adaptive and various learning-based algorithm (AVLA), inspired by group learning behaviors. AVLA simulates individual learning processes within different subgroups at various stages, employing a success history-based parameter adaptation mechanism to reduce manual tuning. Extensive computational experiments on 29 benchmark problems and 5 supply chain networks demonstrate AVLA's superior performance compared to 19 state of the art algorithms. AVLA consistently achieves the best results in terms of both average and best objective function values, making it a powerful tool for addressing complex supply chain network equilibrium problems. |
| title | Adaptive and various learning-based algorithm for supply chain network equilibrium problems |
| topic | Optimization and Control 90 I.2.0 |
| url | https://arxiv.org/abs/2504.11499 |