探討多重資訊服務提供者(ISP)環境下路徑導引策略之動態路網績效

Abstract

The main function of Advanced Traffic Information System (ATIS), as a subsystem of the Intelligent Transportation System (ITS) is to provide real time or in-time traffic related information such as road condition, travel time information, to the subscribed road users or travelers, in particular the route guidance. The effectiveness of such system must rely on an overall construct of integrated framework under which complex interactions among the Information Providers, Road Users, and Information can be identified and recognized. However, very few studies recognized and handled this issue comprehensively. It is therefore the main focus of this proposed study to establish an integrated framework and to explore/evaluate the overall and specific network performance under more realistic multiple Information Service Provider (Multi-ISP) context. This study is in the non-cooperative scenario, where ISPs are assumed to act along with their own information provisions for route guidance strategy. A series of dynamic simulation experiments were performed on a selected real inter-city highway network by proper experiment design. Various level settings of traffic information penetration, ISP market share, and road users’ compliance rate in response to routing guidance formed 36 simulation experiment scenarios. And the associated dynamic simulations were conducted with a modified version of the simulation platform known as DynaTAIWAN. Average travel time was generated to be performance measure to evaluate overall network performance as well as users (hypothetically) receiving services to each ISP. Relationship between various combinations of ISP guidance markets and network performance were analyzed accordingly. The findings were that performance could be at either highest or lowest extreme when the most of the traffic information market shared by few ISPs under multi-ISP environment, exhibiting highest travel time variations with respect to various levels of information penetration and user compliance combinations. On the other hand, the variations were the lowest if ISPs shared the market equally. It may imply that under such equally-shared market scenario, the overall performance tend to be more stable, exhibiting less fluctuations across different combinations of information penetration and user compliance rate. Further analysis differentiated performance of different ISPs with various market share characteristics evaluated by the aggregately average travel times of users receiving guidance service respectively. High-share ISPs seemed to be benefited with better performance by dominance. Interestingly, the performance of singly-dominated ISP was slightly lower then that of its own when co-dominated the market with another ISP. And when the market was shaped toward equally shared, all ISPs enjoyed adequately good performance at medium to high levels of information penetration and user compliance.