應用兩階段中央化資料包絡分析分配航線資源

Abstract

對定期航商而言，經營航線的各項海運成本、人力資源、船舶數量等，屬於航商的資源。定期航商為了追求企業最大利潤，除了設法增加營收外，降低營運成本也是關鍵。許多航商的經營策略從早期以自有船投入航線，直到近年以策略聯盟、交換與購買艙位等方式經營，均是要將營運成本降到最低。而在生產過程中的產出項，可以分為兩部分，意欲產出(desirable ouput)與非意欲產出(undesirable output)，意欲產出為生產者想要或需要的產出，非意欲產出則不受生產者青睞，為生產過程中產生的汙染、噪音等產品。近年來，溫室效應造成全球氣候變遷問題逐年加重，海運業方面，國際海事組織(International Maritime Organization, IMO)也針對溫室氣體排放進行研究，並擬定相關排放指標，有鑑於國際環保法規日趨嚴格，定期航商為符合法規、客戶及利害關係人的要求，發展綠色航運已成為定期航運之趨勢。因此本研究將航線碳排量視為非意欲產出，並納入模型中進行模式建構。 航運業營運績效研究多以財務績效與公司整體營運為主，少有以公司內部航線做為分析對象，且針對航商內部資源進行分配的文獻相當稀少，近年來許多文獻，透過資料包絡分析(Data Envelopment Analysis, DEA)探討非意欲產出的狀況，但大多針對汙染性較大的行業，鮮少針對海運業進行研究。傳統DEA模型，評估各決策單位的效率，並未考量整體決策單位的效率情況，而中央化資料包絡分析(Centralized Data Envelopment Analysis, CDEA)則以中央決策者(Decision Maker)的觀點，對整體決策單位進行評估，以達到整體決策單位的目標，也符合本研究針對定期航商內部航線進行資源分配的特性。 本研究依據定期航運業現況，配合其生產不可儲存之特性，透過運輸業績效評估架構，將定期航運業航線資源分配建構於兩階段網路模型(Two-stage Network Structures)上，以中央化DEA的中央決策者觀點，對於某定期航商進行分析，透過數學模型建構，形成兩階段中央化資料包絡分析(Two-stage Centralized Data Envelopment Analysis)，並對航商營運之亞洲區間航線進行模式驗證，了解航商在現有資源不變，並達到最適意欲產出與非意欲產出最低排放的情況下(航線碳排量)，航商對各航線資源的最適分配情況。 研究結果發現透過此模式，可有效減少非意欲產出(航線碳排量)排放，並可增加航商航線收入與載運里程，而航商可因應內部資源應用方式，選擇採用月資料或季資料作為航線資源分配之參考，並可因應航線營運狀況，對航線資源進行增減，在進行分配。

With regard to liner shipping, each of the shipping costs, human resources and the vessels are belonged to their management resources. In the recent year, shipping companies which join in Strategic Alliance can help to decrease their operating costs. In the produce process, many products are produced. The products are included the desirable outputs and undesirable output. Desirable outputs are intended for producers want or need output. Undesirable outputs are not favored by producers. International Maritime Organization (IMO) had the research in 2009, which studied about the carbon emissions of the industries in the world. The shipping industry was accounted for 3.3% of the carbon emissions in the world. While conventional Data Envelopment Analysis models set targets separately for each DMU, in this study we consider that there is a centralized decision maker who owns or supervises all the operating units. In this study, we combine the Network Data Envelopment Analysis and Centralized Data Envelopment Analysis to develop the Two-stage Centralized Data Envelopment Analysis. We use this model to evaluate the Lines of Shipping Company for resource allocation of a given resource. Resource allocation is an important issue in the management of corporations. It refers to the process of allocating limited resources to different parts of an organization in order to satisfy the overall goals. In the model it is assumed that lines of shipping company produce two types of outputs: desirable outputs (i.e. revenue, TEU- kilometer) and undesirable outputs (i.e. carbon emissions of shipping lines).We divide the model to two stage.The first stage’s inputs are included the salaries, containers reposition, slots capacities, fuel costs, stevedoring expenses and other costs.The first stages’ output are included ship calls and capacity-kilometers. These measures are the outputs from the first stage and are used as inputs to the second stage. The second stage’s outputs are included revenues, TEU- kilometers and the carbon emissions of shipping lines. This research is to adopt the data of 2013 fourth quarter in Asia line services from one of Taiwan’s marine carriers as the object of study. From the results of this study, the model can decrease the undesirable output (i.e. carbon emissions of shipping lines) effectively and increase the desirable outputs (i.e. revenue, TEU- kilometer). It can help shipping company to how to allocate the limited resources to get more revenue.