英文摘要 |
The practices of ridesharing in USA, EU and Japan have proven to be very effective in reducing high commuting cost and large fuel consumption during peak hours of commuting trips in a metropolis. However, ridesharing has never been successful in Taiwan's major metropolises because there is little promotion and almost no incentive for commuters to share rides. On the other hand, taxi-sharing, as an alternative of ridesharing, has been attractive to commuters, especially for the users of public transit systems. At major bus transfer stations or rail and MRT stations, taxi drivers often pick up passengers who are willing to share their taxi fares for nearby destinations. To extend the taxi-sharing experiences and to improve the efficiency of taxi dispatching, we implement a heuristic matching algorithm under the platform of commercial vehicle operations to match travelers who have made advance reservations departing from or arriving at different locations dynamically. Using the questionnaire survey data collected from their 17-day taxi-sharing experiences, we analyze the data by using multivariate analysis methods to understand users' degree of satisfaction regarding taxi-sharing and to explore the internal and external factors affecting the practices. Next, we apply a Logit model to predict taxi-sharing trips transferred from various modes and to evaluate the effectiveness of the practices. In the case study, we find that most of the taxi-sharing trips are transferred from public transit systems. In addition, taxi-sharing is not as attractive as auto or motorcycle for the majority of commuters in Taipei at present unless incentive measures are provided. |