| Fashion retailers face the problem of optimizing the order, allocation, and replenishment to fulfill the size-specific demand. Merchandise is commonly pre-packed for easy handling in the supply chain. By packaging multiple quantities of Stock Keeping Units (SKUs) together, the flow efficiency in the supply chain can be improved. Meanwhile, extra opening costs are introduced. The decision maker needs to match the pack level supply to store level demand as well as to minimize the total costs involved.;This dissertation builds the foundation for the pack optimization problem, which originates from the practical need of the fashion apparel industry. It determines the optimal pack order quantities to satisfy each store's forecasted demand at size level. In the meantime, it minimizes the handling costs, and penalty costs due to mismatches between the supply and demand. In the literature, no formal research has been conducted for this topic. In this dissertation, the pack optimization problem is formulated as an integer programming problem. A dynamic programming model is also developed for a special case of the problem. Two heuristic methods are proposed for practical use. Computational experiments indicate that the hierarchical decomposition heuristic method outperforms other methods. |
