For better supporting informed diversion decisions in case of congestion and incidents, it is important to provide drivers with detailed traffic information such as travel times on intended and alternated routes through Variable Message Signs (VMS). This paper used a stated preference (SP) approach to undertake a quantitative assessment of the potential effects of travel time information provided by VMS on en-route diversion behavior of Shanghai urban freeway drivers. Using SP data, three types of binary probit models were estimated: the model without random coefficient, the model with normal random coefficient, and the model with log-normal random coefficient. The two models with random coefficient accommodate heterogeneous taste preference for travel time savings (the difference between the regular freeway route travel time and the alternate route travel time) among drivers. It was found that drivers' en-route decision on diverting from the regular freeway route to the alternate local street route can be significantly influenced by VMS information. The impact of information is significantly influenced by the characteristics of driver, route, and VMS message of travel time. Travel time saving and drivers' driving age serve as positive factors in drivers' diversion behavior. Signal number on alternate local route, driving employer-provided public car, frequency of driving on freeway, and driver's being mid-age, serve as negative factors in diversion behavior. The findings have important implications for the operation of VMS-based advanced traveler information systems (ATIS).
Abstract