摘要: |
Advanced Traveler Information Systems (ATIS) offer the promise of better informed travel decisions and more efficient use of transportation infrastructure. However, no firm consensus has emerged as to how travelers decide to access information, or how they actually incorporate this information into their travel choices. This paper is designed to complement existing research in this area with results from a large-scale survey of travelers in the Seattle area. The paper focuses on how characteristics of the traveler and of the trip itself influence these choices. The analysis reveals that Seattle-area travelers seek information on a narrow share of their daily travel--just over 3% of respondents recorded trip segments involved information usage, with radio traffic reports the most common source. The decision to seek information does not seem to be strongly influenced by traditional demographics such as sex or income, but is correlated with employment status, Internet usage, and experience with congestion. Trip characteristics are also influential, with travelers much more prone to seek information where (1) information is likely to be available for the route or mode in question, (2) the trip is arrival-time sensitive, and (3) there is a great deal of variability or uncertainty about the travel time. Thus, a typical traveler is about 20 times more likely to seek information on a peak-period, 25-mile commute trip to work than on an off-peak shopping trip. However, studies of ATIS usage should not limit themselves to commute trips, since other time-sensitive trips, such as those to the airport and to medical appointments, also exhibit above-average levels of information acquisition. If expanding ATIS use beyond this narrow segment of trips is a public policy goal, this will require increasing the benefits of ATIS and/or reducing the costs of access. In light of earlier research on travelers' ATIS enhancement priorities, this implies that improvements in fundamental data quality--in particular, more frequent updates and greater geographical coverage--will be most useful. These improvements would broaden the range of trips on which travelers would perceive information acquisition to have a favorable benefit-cost profile. Travelers' response to the information they receive is complex, and their reactions in real-world settings are not easily modeled. This paper does confirm results from earlier work, such as the fact that information acquisition rarely leads to a shift in travel modes. In contrast to some studies, however, it also indicates that changes in route are more common than changes in departure time and that personal characteristics, such as gender, do not have a strong influence on thi |