Optimization of a Prospective Carpooling Service in the Regional Unit of Thessaloniki with the Use of Conjoint Analysis and Market Simulation

Abstract

In the Regional Unit of Thessaloniki, severe traffic problems are observed during peak hours and especially in its urban center. It is also observed that the occupancy of the cars that serve the demand is rather low, i.e., 1.44 travelers per vehicle. This phenomenon, combined with the unattractiveness of public transport, exacerbates the Unit’s traffic problem. One way to alleviate the problem is to create carpooling schemes, a practice in which the driver of a vehicle accepts other passengers. In this paper, a Conjoint Analysis questionnaire survey is presented, which was conducted in order to determine whether Greek travelers are receptive to carpooling, as well as what factors influence their choices, such as cost, waiting time, and driving behavior of the car driver, etc. In the survey, 84 people participated. The sample seems to be low, but in the case of Choice-Based Conjoint (CBC) analysis, it is sufficient to derive statistically significant results. The main objective is to optimize such a prospective carpooling service to make it more attractive to the users. The attribute with the highest importance was found to be “Reliability of driver / Safety” with 54.3%, followed by “Route cost sharing” with 20.0%, “Number of passengers in vehicle” with 16.6%, and “Time of the route” with 9.1%. The “Reliability of driver/Safety” attribute was found to be of paramount importance, a fact that was verified with three (3) scenarios of market simulation.