Reliability of Variable Message Signs Information: A Field Experiment Study in Chongqing
Jianwei Du,Gang Ren,Wei Liu,Guangchuan Yang,Haojie Li,Dawei Li
DOI: https://doi.org/10.1016/j.trf.2021.11.002
IF: 4.349
2021-01-01
Transportation Research Part F Traffic Psychology and Behaviour
Abstract:A variable message sign (VMS), which is a key component of intelligent transportation systems, has been frequently used in the management of urban roads and motorways to provide drivers with real-time traffic condition information about a road section or area. Nevertheless, there is a lack of unified regulations for VMS design, which makes it difficult to accurately determine the understandability and legibility of VMS information. In practice, inappropriate designs of VMSs are common, such as overload of VMS information and excessive number of phases, particularly on the urban roads of China. Building on our earlier findings obtained by surveys, in this study, field tests were conducted to assess the reliability of VMS information in terms of the visual perception characteristics of drivers. For a full visual perception, the information displayed on a VMS must be reliable, i.e., it should be easy for drivers to detect, obtain, and understand the messages they require. Two basic theories-information theory and visual perception theory-were introduced to build a quantification model of VMS information, select evaluation indices, and design the experimental process. A total of 24 drivers participated in the field experiments and questionnaires. Five indices-information obtainment rate, periodic validity, driver subjective scoring and the corresponding UI (represents information intelligibility), and CE (represents information legibility)-were analyzed to evaluate their relationships to VMS information reliability. The results confirmed that the amount of information, and the scrolling period considering redundant distance, significantly affect the reliability of VMSs information. For static VMSs, the information obtainment rate and the subjective scoring decrease with increasing amount of information. The recollection accuracy of the drivers significantly declines when the amount of information shown increases to 90 bits, corresponding to an information obtainment rate of less than 0.8 and UI equal to 0. For dynamic VMSs, the information reliability deteriorates as the scrolling period shortens and the number of phases increases. Unreasonable reliability is found when the periodic validity is less than 0.9, i.e., the actual scrolling period is more than 10% less than the calculated one, corresponding to CE equal to 0. The reliability of information was evaluated by combining the subjective scoring of the drivers and a data-based statistical analysis and considering driving safety. Accordingly, it was recommended that 90 bit is the maximum amount of information to be shown on a VMS and that the preferred scrolling period of each phase of a dynamic VMS is 5 s. The results of this study support the objective of providing reliable information to drivers by addressing the problems related to the amount of information and presentation time for VMSs. These findings provide a basis for determining the thresholds for VMS information to promote practical and user-friendly designs of VMSs on urban roads.