隨著企業市場規模擴大，供應鏈產銷整合問題將變得更為複雜，一般企業的資訊系統對供應鏈產銷整合多在資料處理的範疇，通常不牽涉到營運規劃的部分，對企業管理者制訂產銷計畫的決策支援有限。本研究在供應鏈網路環境下，以供應鏈整體營運資源整合分配的觀點，運用最佳化理論與啟發式演算法技術，設計並比較三種產銷規劃方法以解決產銷問題。本研究的貢獻包括：(1) 在多規劃期多產品的四階供應鏈下，考量生產技術限制與配送路徑選擇等特性建立一個以利潤極大化為績效目標的完整決策模型；(2)考慮到完整模型有NP-Complete 的限制，本研究提出了啟發式演算法，或是將完整模型拆解為兩階段模型以作為替代方案，之後並以不同的實驗情境比較來驗證演算法、完整模型與兩階段模型的品質與效率。由測試的情境規劃結果可知，本研究所提出的規劃演算法除了能夠節省時間，亦能維持一定的求解品質，在實務應用上應可輔助管理者進行決策。

From the perspective of integrated supply chain resource allocation, the paper proposes three methods to solve the problem of integrated planning in the four-echelon supply chain network. The first one is a full MILP model which considers both production and distribution constraints. The second one is a two-stage MILP model which allocates the production demands to manufacturing sites in stage one and a distribution model is used to provide the distribution plans in stage two. The third one is a heuristic algorithm to find feasible plans efficiently. Several scenarios are tried out to evaluate the effectiveness and efficiency of the proposed methods. The result reveals that the heuristic algorithm is more efficient than the two MILP models.