While RWIS stations are the most adopted technology that transportation authorities use to monitor their vast road network, they can only be located at select areas due to budgetary constraints. It is therefore indispensable to fill those large spatial gaps that exist between stationary RWIS stations to promote safer driving conditions and lower WRM activities cost. Furthermore, most stationary and mobile RWIS nowadays are equipped with cameras that provide users with a direct view of the road conditions being covered; however, checking the road conditions via these cameras is still being done manually, which hinders the full utilization of these rich image-based road condition data for optimizing maintenance services and improving the travel of the general public.  

Since what is commonly accessible to highway maintenance personnel is stationary RWIS, further investigation is required to infer RSC at unmonitored areas (e.g., between RWIS stations) during inclement weather events using stationary RWIS alone with help of road conditions profiles constructed via mobile RWIS. Therefore, the primary objective of this project is to develop a systematic yet transferrable method for estimating RSCs between different pairs of existing stationary RWIS stations using Big Data and advanced modelling techniques; namely, kriging and deep learning.