考慮碳排放因素之城市物流中心選址問題

Abstract

因應溫室效應所帶來的環境衝擊日益迫切，歐洲各國越來越積極減少溫室氣體，如開徵二氧化碳稅，以達成京都議定書 (Kyoto Protocol) 中之各國溫室氣體減量目標；以及推動之排放交易制度，以符合歐盟溫室氣體排放交易指令 (Directive 2003/87/EC, Directive) 之規定，執行各國溫室氣體排放源之排放量核配規劃工作。 近年來，尤其在人口密集的大城市，汽車廢氣污染呈加劇之勢，成為城市的主要污染物 (Taniguchi et al., 2001)。美國空氣污染的50%來自運輸工具，日本也占到20%。運輸對環境帶來的負面衝擊，致使城市貨物運輸在永續發展上面臨許多挑戰。 本研究建置城市物流中心選址及規模決定 (即泊位數) 最佳化模式 (Taniguchi et al., 1999)，最小化物流活動之總成本，相較於過去的物流中心選址研究，本研究之特色分述如下： 1.於路網考慮城市物流的特性。 2.考慮碳排放因素。 本研究之數學架構為二階層規劃問題，二階層規劃問題可視為長期性規劃面的模型，因此利用上層規劃者長期的觀點來做決策，且也必須符合下層使用者短期性的目標。 在規劃者問題屬於非線性混合型整數規劃，使用者問題屬於非線性規劃。非線性規劃是具有非線性約束條件或目標函數的數學規劃，而非線性整數規劃可分成線性部分和整數部分。在整數規劃中，僅部分變數限制為整數，則稱為混合型整數規劃。 最後，針對情境一及情境二，與範例分析比較。並對碳排放成本轉換係數及道路容量作敏感度分析，探討係數對目標值 (總成本) 的影響。以及對後續研究者提出未來研究方向建議。

Due to the environmental impact by greenhouse effect increase, Europe is active in reducing greenhouse gases, such as the introduction of a carbon tax in order to achieve the Kyoto Protocol in the countries to greenhouse gas reduction targets; and promote emissions trading system, executive to comply with the provisions of the EU greenhouse gas emissions trading instructions (Directive 2003/87/EC, Directive), executive of national greenhouse gas emission sources allocation planning. In recent years, especially in densely populated cities, automobile exhaust pollution is exacerbating by the trend, becomes the city's major pollutants (Taniguchi, et al., 2001). 50% of U.S. air pollution from transport, and Japan accounted for 20%. The adverse impacts of transport on the environment, resulting in the city freight transport in the face many challenges to sustainable development. In this research, the model building a city logistics center location and size decision (berth number) optimization model to minimize the logistics activities total costs (Taniguchi, et al, 1999), compared to the past study, the characteristics of this research are described as follows: 1. The road network to consider the characteristics of city logistics. 2. Considering the carbon emission factors. The mathematical framework of the research is bi-level programming, bi-level programming can be regarded as a model of long-term planning, so use the long-term view of the upper level to make decisions, and must also be consistent with the lower level’s goals. Upper level problem is mixed integer nonlinear programming, and lower level problem is nonlinear programming. Nonlinear programming is with nonlinear constraints or nonlinear objective function, and mixed integer nonlinear programming can be divided into a linear part and the integer part. In integer programming, the only part of the variable limit is an integer, known as mixed integer programming. Finally, Scenario analyzed and compared. And the carbon emissions cost of conversion factors and road capacity for sensitivity analysis to investigate the influence of the coefficient on the target value (total cost). As well as follow-up researchers recommend future research directions.