Improving Variable Speed Limit Systems: investigation of the impact of weather and pavement conditions on operating speed

A Variable Speed Limit System (VSLS) is an advanced ITS scheme that can be employed to increase the safety level of highway facilities by varying the speed limit according to downstream operational condition and/or current weather and pavement conditions. In a weather- based VSLS, the proposed speed limit is related to weather and pavement conditions, and therefore, the relationship between speed and weather variables has to be well defined. This research investigates the relationship between adverse weather conditions and operating speed within the framework of improving an existing VSLS in British Columbia (BC), Canada. The purpose is to propose recommendations for the weather-based module to better correlate weather parameters with operating speeds. A dataset of six weeks of traffic and weather data was collected at 36 sites located on three major corridors in BC. The data was analyzed using stepwise regression where the weather variables that affect operating speed were identified. A model was developed independently for each corridor due to the differences in their characteristics. The models were further modified by reducing the number of independent variables for practicality and ease of application. Pavement surface status was shown to be the most important variable in building speed-weather models followed by surface temperature. Level of grip and visibility, on the other hand, were shown to be significant in two models only while having no major impact in the third model.