在實務上，貨物運送的單位成本常隨數量的增加而遞減，其成本函數曲線為凹形，而此類問題可定式為含凹形節線成本之最小成本網路流動問題，但此問題屬於 NP-hard問題，故難在有限時間內求得大型問題的最佳解。新近的和諧搜尋演算法目前在各領域的問題求解上效果頗佳，但尚未發現有應用於含凹形節線成本最小成本網路流動問題，緣此，本研究以和諧搜尋演算法為基礎，並結合粒子群演算法、螞蟻族群演算法、門檻值接受法與凹形成本網路啟發解法之特點，以節線及路徑為基礎發展一混合式全域搜尋法，以有效求解含凹形節線成本之最小成本網路流動問題。為測試本研究演算法在不同規模及參數的網路問題之求解績效，本研究設計一隨機網路產生器產生大量隨機網路，並測試遺傳演算法、門檻值接受法、大洪水法、類螞蟻族群演算法及粒子群演算法，以評估本研究演算法之求解績效。測試結果顯示本研究演算法求解品質良好，可提供實務界求解此類網路運送問題之參考。

In practice, the unit cost for transporting freight usually decreases as the amount of freight increases. Hence, in actual operations the transportation cost function can usually be formulated as a concave cost function, causing the transshipment problem an NP-hard problem. The harmony search (HS), a global search algorithm, has led to good results in many applications. Since there has not yet been any research applying HS to minimum concave cost network flow problems, we employ HS, coupled with the techniques of PSO, ACS and TA, to develop a global search algorithms for efficiently solving minimum concave cost network flow problems. Finally, to evaluatealgorithms we designed a network generator to create a sufficient number of problem instances. To evaluate our algorithm, we also tested the recently designed TA, GDA, GA, ACS and PSO that solve minimum concave cost network flow problems. The results show that the developed algorithms performed well in the tests.