We investigate the problem of shelf space allocation in distribution centres, considering product segregation and simultaneous categorization across horizontal and vertical shelves. Additionally, our model accounts for products with different storage conditions and limited storage compatibility. It is applicable to items requiring specific temperature, humidity, illumination, and chemical compatibility. The problem is formulated to optimize the convenience and efficiency of order pickers in the distribution centre. We propose a new “mushroom picker” heuristic using two variations of the internal sorting rule. To improve computational efficiency, thirteen parameters were introduced to reduce the solution space, ensuring high-quality solutions without relying on randomness or exhaustive search. When compared to the optimal solution obtained from a commercial solver, our heuristics achieved a solution quality of at least 98%–99%. Optimal solutions were found for 19 and 18 out of 25 instances using heuristics H1 and H2, respectively.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Mushroom Picker Heuristics for Shelf Space Allocation in Distribution Centre with Multi-criteria Product Segregation and Categorization

  • Kateryna Czerniachowska,
  • Krzysztof Lutosławski

摘要

We investigate the problem of shelf space allocation in distribution centres, considering product segregation and simultaneous categorization across horizontal and vertical shelves. Additionally, our model accounts for products with different storage conditions and limited storage compatibility. It is applicable to items requiring specific temperature, humidity, illumination, and chemical compatibility. The problem is formulated to optimize the convenience and efficiency of order pickers in the distribution centre. We propose a new “mushroom picker” heuristic using two variations of the internal sorting rule. To improve computational efficiency, thirteen parameters were introduced to reduce the solution space, ensuring high-quality solutions without relying on randomness or exhaustive search. When compared to the optimal solution obtained from a commercial solver, our heuristics achieved a solution quality of at least 98%–99%. Optimal solutions were found for 19 and 18 out of 25 instances using heuristics H1 and H2, respectively.