本研究探討隨機實體配送網路設計問題，考量隨機需求不確定性、運輸成本規模經濟及供給過剩與不足等實務營運情況。以兩階段隨機規劃方法，構建問題之數學模型，目標為尋求總成本最小，主要決策為求取最佳之設施位址、運輸配送路網、運送流量及設施點供給過剩或不足的配置。為使模型能應用於實際大型問題，本研究以樣本平均近似法之概念為基礎，發展一求解演算法。最後設計一案例，對所提模型及演算法進行測試。結果顯示，模型具實用性及可行性；演算法具運算效率，且能確保求解的品質。

This study addressed the stochastic physical distribution network design problem, considering demand randomness, economies of scale in transportation costs, and supply shortage/surplus at facilities in order to reflect the practical operations. The problem is formulated as a two-stage stochastic program. The objective function is to minimize the total cost and the decisions are to determine the optimal facility location, distribution network, flow allocation, and supply shortage/surplus at facilities. In order to enable the application in the real-size large-scale problem, a solution algorithm based on the Sample Average Approximation technique was developed to solve the proposed model. An illustrative example was created to test the proposed model and solution algorithm. The results show that the model is practically implementable and the algorithm is efficient while ensuring the quality of solution.