Analyzing drivers' preferences and choices for the content and format of variable message signs (VMS)

Background: Recent advance in variable message signs (VMS) technology has made it viable to provide spatio-temporal information on traffic and network conditions to drivers. There is a debate whether VMS diverts drivers' attention away from the road and may cause unnecessary distraction in their driving tasks due to inconsistent VMS contents and formats. There are also other external factors such as weather conditions, visibility and time of day that may affect the integrity and reliability of the VMS. In China, only about 23% drivers were persuaded by VMS to follow route diversion. Objective: In order to capture the full benefits of VMS, the aim of this paper is therefore to identify the factors affecting VMS by examining what kinds of VMS contents, formats and their interactions are more preferable to drivers, specifically in China. Methods: A revealed preference (RP) questionnaire and stated preference (SP) survey consisting of 1154 samples from private and taxi drivers was conducted and analyzed using discrete choice model. Results: The results revealed that the information showed by amber-on-black on text format, white-on-blue on graph format or the suggested route diversion information showed by single line are preferred by drivers in fog weather. In addition, highly educated drivers or drivers with no occupation are more prone to the qualitative delay time on a text-graph format in fog weather. In normal weather, drivers with working trip purpose are mostly preferred to receive the information on a congested traffic condition with a reason on a text-only format. However, the congested traffic condition along with the information on the apparent causes shown by red-on black or green-on-black on a text-only format was least preferred by drivers. Regarding current and adjacent road traffic information, drivers prefer to receive the suggested route diversion on a graph-only format in fog weather and the qualitative delay time on a text-graph format in normal weather. Irrespective to weather conditions, male drivers incline to the qualitative delay time on a text-graph format. Conclusions: The findings of this study could assist traffic authorities to design the most acceptable VMS for displaying traffic information for the purpose of improving road traffic efficiency and provide the theory evidence for the design of in-vehicle personalized information service system.