Trafikverket, the agency in charge of state road maintenance in Sweden, have a number of so-called Road Weather Information Systems (RWIS). The main purpose of the stations is to provide winter road maintenance workers with information to decide when roads need to be plowed and/or salted. Each RWIS have a number of sensors which make road weather-related measurements every 30 minutes. One of the sensors is dug into the road which can cause traffic disturbances and be costly for Trafikverket. Other RWIS sensors fail occasionally. This project aims at modelling a set of RWIS sensors using supervised machine learning algorithms. The sensors that are of interest to model are: Optic Eye, Track Ice Road Sensor (TIRS) and DST111. Optic Eye measures precipitation type and precipitation amount. Both TIRS and DST111 measure road surface temperature. The difference between TIRS and DST111 is that the former is dug into the road, and DST111 measures road surface temperature from a distance via infrared laser. Any supervised learning algorithm trained to model a given measurement made by a sensor, may only train on measurements made by the other sensors as input features. Measurements made by TIRS may not be used as input in modelling other sensors, since it is desired to see if TIRS can be removed. The following input features may also be used for training: road friction, road surface condition and timestamp. Scikit-learn was used as machine learning software in this project. An experimental approach was chosen to achieve the project results: A pre-determined set of supervised algorithms were compared using different amount of top relevant input features and different hyperparameter settings. Prior to achieving the results, a data preparation process was conducted. Observations with suspected or definitive errors were removed in this process. During the data preparation process, the timestamp feature was transformed into two new features: month and hour. The results in this project show that precipitation type was best modelled using Classification And Regression Tree (CART) on Scikit-learn default settings, achieving a performance score of Macro-F1test = 0.46 and accuracy = 0.84 using road surface condition, road friction, DST111 road surface temperature, hour and month as input features. Precipitation amount was best modelled using k-Nearest Neighbor (kNN); with k = 64 and road friction used as the only input feature, a performance score of MSEtest = 0.31 was attained. TIRS road surface temperature was best modelled with Multi-Layer Perceptron (MLP) using 64 hidden nodes and DST111 road surface temperature, road surface condition, road friction, month, hour and precipitation type as input features, with which a performance score of MSEtest = 0.88 was achieved. DST111 road surface temperature was best modelled using Random forest on Scikit-learn default settings with road surface condition, road friction, month, precipitation type and hour as input features, achieving a performance score of MSEtest = 10.16.

Because winter maintenance is so costly, Utah Department of Transportation (UDOT) personnel asked researchers at Brigham Young University to determine whether asphalt or concrete pavements require more winter maintenance. Differing thermal properties suggest that, for the same environmental conditions, asphalt and concrete pavements will have different temperature profiles. Climatological data from 22 environmental sensor stations (ESSs) near asphalt roads and nine ESSs near concrete roads were used to 1) determine which pavement type has higher surface temperatures in winter and 2) compare the subsurface temperatures under asphalt and concrete pavements to determine the pavement type below which more freeze-thaw cycles of the underlying soil occur. Twelve continuous months of climatological data, primarily from the 2009 calendar year, were acquired from the road weather information system operated by UDOT, and erroneous data were removed from the data set. To predict pavement surface temperature, a multiple linear regression was performed with input parameters of pavement type, time period, and air temperature. Similarly, a multiple linear regression was performed to predict the number of subsurface freeze-thaw cycles, based on month, latitude, elevation, and pavement type. A finite-difference model was created to model surface temperatures of asphalt and concrete pavements based on air temperature and incoming radiation. The statistical analysis predicting pavement surface temperatures showed that, for near-freezing conditions, asphalt is better in the afternoon, and concrete is better for other times of the day, but that neither pavement type is better, on average. Asphalt and concrete are equally likely to collect snow or ice on their surfaces, and both pavements are expected to require equal amounts of winter maintenance, on average. Finite-difference analysis results confirmed that, for times of low incident radiation (night), concrete reaches higher temperatures than asphalt, and for times of high incident radiation (day), asphalt reaches higher temperatures than concrete. The regression equation predicting the number of subsurface freeze-thaw cycles provided estimates that did not correlate well with measured values. Consequently, an entirely different analysis must be conducted with different input variables. Data that were not available for this research but are likely necessary in estimating the number of freeze-thaw cycles under the pavement include pavement layer thicknesses, layer types, and layer moisture contents.