行前交通資訊對城際旅行者運具選擇行為影響之研究

A Study on the Effects of Pre-trip Traffic Information on Intercity Traveler's Mode Choice Behaviors

行前交通資訊（pre-trip traffic information）可能是影響旅行者運具選擇行為之重要因素之一，然而，傳統個體運具選擇模式所考慮的解釋變數，大抵以旅行者個人之社經變數、運輸工具之屬性及服務水準等變數為主，鮮少將行前交通資訊影響旅行者對運具選擇行為之隱藏性變數（latent variable）納入考量，因此，本研究以「隱藏性變數模式」（latent variable models）主要的研究方法「線性結構關係模式」（Linear Structure Relation,LISREL）進行相關研究。實證分析結果顯示，旅行者對行前交通資訊的感知，以及對運具選擇的態度等隱藏性變數，確實對旅行者的運具選擇行為有所影響，惟進一步將這些隱藏性變數納入個體運具選擇模式後，對整體運具選擇模式之解釋能力提升效果有限，究其原因，影響城際旅行者對運具選擇之因素，除了行前交通資訊以外，尚有其他因素存在，例如運具的舒適性、安全性、便利性等。

Pre-trip traffic information is one of the crucial factors that affect intercity traveler's mode choice behaviors. However, state-of-the-art mode choice methods estimate intercity transportation market shares by only considering users' socioeconomic characteristics and transportation modes' attributes and levels of service; without specifically capturing the effects of traffic information on user's mode choice behaviors. Therefore, the present research employed latent variable models to investigate the value of pre-trip traffic information on intercity traveler's mode choice behaviors. Among the family of latent variable models, linear structure relation (LISREL) model is the main approach in social sciences, and it was applied to the investigation of the present research topic. The empirical study results based on questionnaire surveys indicated that certain latent variables concerning perception to pre-trip traffic information and attitude towards mode choice did affect traveler's mode choice decisions. Nevertheless, if we further incorporate the latent variables identified in the LISREL model into disaggregate mode choice models, it did not significantly improve models' performance. It is concluded that, besides latent variables, there exist other factors (e.g., comfort, safety, and convenience, etc.) jointly contributing to the final outcomes of intercity traveler's mode choices.