高鐵運具選擇模式之更新、預測與政策模擬之研究 高鐵運具選擇模式之更新、預測與政策模擬之研究

A Study of the Updating, Prediction, and Policy Simulation for the Mode Choice Model of HSR

鄭少勳(Shao-Hsun Cheng)

指導教授 ： 楊志文

國立臺中科技大學/商管學群/事業經營研究所/碩士(2008年)

摘要

運具市場佔有率預測之探討，主要是從旅運者的個體選擇行為推估，然而運具選擇模式的建構，會耗費相當大的時間和成本。本研究目的即透過模式更新解決這方面的問題，利用轉移尺度法更新模式中的替選方案特定常數和尺度因子，不同於原始的轉移尺度法需要新的個體資料，修正後的轉移尺度法只需要知道運具的市佔率即可更新模式，較原始轉移尺度法更為方便。此外，在預測市場佔有率時，由於新運具高速鐵路的加入，導致少了新運具的常數項，該如何指定高鐵的常數，也是本研究所探討的。其次本研究透過政策模擬的方式，針對班次、票價、油價、國道收費去做調整，來觀察政策與運具市場佔有率之間的關係。最後本研究將整個研究流程，透過VBA程式去設計一套使用者介面，讓使用者在操作變數及預測市場佔有率能更加的方便。
研究結果顯示，透過模式更新的程序，比未更新模式下所做的市場佔有率預測更加的準確。在預測高速鐵路新運具加入的市場佔有率，其常數項的指定以台鐵的常數代入，預測效果最佳。政策模擬的部分，根據對市場佔有率變動的幅度大小來看，依序是票價、國道收費、油價的策略較有效果，最差的是班次的調整。票價、油價和國道收費最能影響市場佔有率的變動，其中又以票價和國道收費較容易隨策略做調整。班次的增減對於市場佔有率影響甚小，表示目前的班次幾乎達到尖峰，再增加只會提高成本，對於市佔率反而沒有顯著提升。從市場佔有率的大小和變動的幅度來看，以高速鐵路、國道客運和小汽車為最，台鐵和航空的影響相當的微小，表示旅客多在高速鐵路、國道客運和小汽車三個運具之間做轉移。

關鍵字

高速鐵路 ； 運具選擇模式 ； 模式更新 ； 轉移尺度 ； 政策模擬

並列摘要

The forecasting of market share in intercity transportation is estimated mainly from the disaggregate behavior of travelers. However, the construction of mode choice model wastes a lot of research resource times and costs. The purpose of this research firstly adapts model updating method to solve the above problem. We use the approach of transfer scaling to update the alternative specific constants and scale parameter of mode choice. In different to the primitive transfer scaling, which needs to collect new disaggregate sample, the revised transfer scaling can updated the model just by market share of mode. In addition, because of the new mode introductory, the lack of the alternative specific constant of new mode can not predicting its market share. Hence how to specify the alternative specific alternative specific constant of high-speed rail is also to be discussed in this research. Secondly, this research has observed the relation between the policy and market share in adjusting the frequency, fare, oil price and highway toll by policy simulation. Finally, this research has used VBA program to design a user interface, in order to show the entire research procedure, and to make it convenient for the user to operate.
The result indicated that the updating model has the better predicting effect then original model. The way of specifying the constant of high-speed rail as rail constant has better results in predicting market share. According to the changes in market share, the policy of fare and highway toll has the most influences in policy simulation. According to the changes in market share, the policy of high-speed rail, bus, and car has the most influences in policy simulation, which indicates that the passengers are more likely to choose the high-speed rail, bus, and car.

並列關鍵字

High-Speed Rail ； mode choice model ； model updating ； transfer scaling ； policy simulation

參考文獻

段良雄、楊志文(民90)，更新方法、效用函數指定與數據型態對運具選擇模式更新之影響，運輸計劃季刊，第三十卷，第四期，763-794。

周榮昌、陳筱薇、劉祐興(民93)，城際旅運者運具選擇行為之研究，中國土木水利學刊，第十六卷，第二期，33-44。

Abdelwahab, W. M. (1991). Transferability of Intercity Disaggregate Mode Choice Models in Canada. Canadian Journal of Civil Engineering, 18, 20-26.

Atherton, T. J. & Ben-Akiva, M. (1976). Transferability and Updating of Disaggregate Travel Demand Models. Transportation Research Record, 610, 12-18.

Badoe, D. A. & Miller, E. J. (1995). Analysis of the Temporal Transferability of Disaggregate Work Trip Mode Choice Models. Transportation Research Record, 1493, 1-11.