Our society needs to continuously perform transports of people and goods toensure that business is kept going. Every disturbance in the transportation ofpeople or goods affects the commerce and may result in economical losses forcompanies and society. Severe traffic accidents cause personal tragedies forpeople involved as well as huge costs for the society. Therefore the roadauthorities continuously try to improve the traffic safety. Traffic safety may beimproved by reduced speeds, crash safe cars, tires with better road grip andimproved road maintenance. The environmental effects from roadmaintenance when spreading de-icing chemicals need to be considered, i.e.how much chemicals should be used to maximize traffic safety and minimizethe environmental effects. Knowledge about the current and upcoming roadcondition can improve the road maintenance and hence improve traffic safety.This thesis deals with sensors and models that give information about the roadcondition.The performance and reliability of existing surface mounted sensors wereexamined by laboratory experiments. Further research involved field studies tocollect data used to develop surface status models based on road weather dataand camera images. Field studies have also been performed to find best usageof non intrusive IR technology.The research presented here showed that no single sensor give enoughinformation by itself to safely describe the road condition. However, the resultsindicated that among the traditional road surface mounted sensors only theactive freezing point sensor gave reliable freezing point results. Furtherresearch aimed to find a model that could classify the road condition indifferent road classes from existing road weather sensor data and road images.The result was a model that accurately could distinguish between the roadconditions dry, wet, snowy and icy. These road conditions are clearly dissimilarand are therefore used as the definition of the road classes used in this thesis.Finally, results from research regarding remote sensing IR technology showedthat it significantly improves knowledge of the road temperature and statuscompared to data from surface mounted sensors. / Vårt samhälle bygger på att det finns effektiva transporter av människor ochvaror för att säkerställa att samhällets funktioner fungerar och att företagenkan genomföra sina affärer. Störningar i transporterna av människor och varorpåverkar handeln och kan leda till ekonomiska förluster för både företag ochvårt samhälle. Allvarliga trafikolyckor orsakar personliga tragedier för deinblandade samt stora kostnader för samhället. Det är med denna bakgrundsom vägmyndigheterna kontinuerligt arbetar med att förbättratrafiksäkerheten. Trafiksäkerheten kan förbättras genom att minskahastigheterna, se till att bilarna blir krocksäkra, krav på däck med bättreväggrepp och ett bättre vägunderhåll. Miljöeffekterna från vinterväghållningdär avisningsmedel sprids på vägarna måste beaktas, d.v.s. hur mycketkemikalier bör användas för att maximera trafiksäkerheten och minimeramiljöpåverkan. Denna avhandling handlar om sensorer och modeller som gerinformation om väglaget. En kunskap om aktuellt och kommande väglag kanförbättra väghållningen och därmed öka trafiksäkerheten.I avhandlingen har prestanda och tillförlitlighet hos befintliga vägmonteradesensorer granskats i laboratorieexperiment. Data från fältstudier har använtsför att utveckla modeller som kan ge information om vägytans status baseratpå meteorologiska mätdata och kamerabilder. Det har också genomförtsfältstudier för att utforska den fördelaktigaste användningen av beröringsfriinfraröd sensorteknik.Den forskning som presenteras här visar att ingen enskild givare ger tillräckliginformation för att säkert beskriva väglaget. Från de traditionella ytmonteradesensorerna drogs slutsatsen att den aktiva fryspunktsgivaren gav de mesttillförlitliga fryspunktsresultaten. Det vidare arbetet handlade om att hitta enmodell som skulle kunna klassificera vägförhållanden i olika vägklassergenom att utnyttja information från befintliga sensorer och kamerabilder.Detta arbete resulterade i en modell som tillförlitligt kan särskilja väglagentorr, våt, snöig och isig. Dessa väglag är väsentligt olika och har därför valtssom väglagsklasser i denna avhandling. Under en säsong genomfördes ävenfältförsök med beröringsfri infraröd mätteknik där det visade sig att denberöringsfria teknologin förbättrar kunskapen om vägbanans temperatur och vägbanans status.

Covering the Northern part of the Province, Ontario’s low-volume roads provide a link from remote resource areas to markets. Thus, preserving this transportation asset from the two main sources of pavement deterioration, namely traffic loading and the environment is extremely critical to the movement of goods and to the economy. In particular, Northern Ontario’s secondary highways are challenged by a combination of heavy, low frequency traffic loading and a high number of freeze-thaw cycles for which most of these highways have not been structurally designed. Therefore they experience environmental damage and premature traffic-induced deterioration. To cope with this issue, the Ontario Ministry of Transportation places Spring Load Restrictions (SLR) every year during spring-thaw. For economic reasons, the duration of SLRs is usually fixed in advance and is not applied proactively or according to conditions in a particular year. This rigidity in the schedule needs to be addressed, as it can translate into economic losses either when the payload is unnecessarily restricted or when pavement deterioration occurs. While the traditional approaches are usually qualitative and rely on visual observations, engineering judgment and historical records to make SLR decisions, the latest approaches resort to climatic and deflection data to better assess the bearing capacity of the roadway. The main intent of this research was to examine how the use of a predictor for frost formation and thawing could improve the scheduling of load restrictions by tracking the frost-strengthening and thaw-weakening of the pavement structure. Based on field data captured in Northern Ontario, and on a preliminary analysis that found good correlation between frost thickness in the roadway and Road Weather Information Systems (RWIS) variables, more advanced frost and thaw predictors were developed as part of this research and are presented herein. The report outlines how the model was developed, details the calculation algorithms, and proposes an empirical methodology for a systematic site-specific calibration. This research also involved several experimental and numerical tools, including the use of a Portable Falling Weight Deflectometer (PFWD) to estimate pavement strength during spring thaw, and the use of the Mechanistic-Empirical Pavement Design Guide (MEPDG) software to simulate the impact of SLR on the performance of typical Northern Ontario low volume roads.

Covering the Northern part of the Province, Ontario’s low-volume roads provide a link from remote resource areas to markets. Thus, preserving this transportation asset from the two main sources of pavement deterioration, namely traffic loading and the environment is extremely critical to the movement of goods and to the economy. In particular, Northern Ontario’s secondary highways are challenged by a combination of heavy, low frequency traffic loading and a high number of freeze-thaw cycles for which most of these highways have not been structurally designed. Therefore they experience environmental damage and premature traffic-induced deterioration. To cope with this issue, the Ontario Ministry of Transportation places Spring Load Restrictions (SLR) every year during spring-thaw. For economic reasons, the duration of SLRs is usually fixed in advance and is not applied proactively or according to conditions in a particular year. This rigidity in the schedule needs to be addressed, as it can translate into economic losses either when the payload is unnecessarily restricted or when pavement deterioration occurs. While the traditional approaches are usually qualitative and rely on visual observations, engineering judgment and historical records to make SLR decisions, the latest approaches resort to climatic and deflection data to better assess the bearing capacity of the roadway. The main intent of this research was to examine how the use of a predictor for frost formation and thawing could improve the scheduling of load restrictions by tracking the frost-strengthening and thaw-weakening of the pavement structure. Based on field data captured in Northern Ontario, and on a preliminary analysis that found good correlation between frost thickness in the roadway and Road Weather Information Systems (RWIS) variables, more advanced frost and thaw predictors were developed as part of this research and are presented herein. The report outlines how the model was developed, details the calculation algorithms, and proposes an empirical methodology for a systematic site-specific calibration. This research also involved several experimental and numerical tools, including the use of a Portable Falling Weight Deflectometer (PFWD) to estimate pavement strength during spring thaw, and the use of the Mechanistic-Empirical Pavement Design Guide (MEPDG) software to simulate the impact of SLR on the performance of typical Northern Ontario low volume roads.

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.

NIRA is a company based in Linköping and operates within the automotive industry. NIRA's Tire Grip Indicator is one of their products used to create friction values during normal driving conditions. This data is, together with additional sensor information from the vehicle, such as environmental sensors, collected in NIRA's cloud environment RSI. Utilising hundred of thousand of vehicles, several products are created and offered to the market. This master thesis will focus on the winter maintenance industry business where NIRA provides refined information to support decision making and quality control for the winter maintenance operations both in Sweden and internationally. The master thesis will be conducted in close cooperation with several actors active in the winter maintenance business within the scope of a collaboration project.  The thesis explores how road surface information can be used as decision support for actors within road maintenance. The thesis starts with a literature study describing different perspectives of winter maintenance and its relation to decision support systems. Through the study, several decision support systems are used to analyse the road climate and the maintenance operations in the city of Gothenburg. The analysis takes the form of a case study where each day during the winter season is analysed. Several features are extracted from the cases and are used to classify each case as one of 6 categories. Finally, the cases are valued within three cost areas: traffic accidents, accessibility, and fuel consumption.  The findings of the thesis show that the different systems within the project fulfil different functions in the work for more efficient winter road maintenance operations, where the forecasting tools can be seen as tools for planning measures. At the same time, friction data from fleets of connected vehicles can help evaluate the outcome of maintenance actions. This applies especially to the possibility of implementing and evaluating preventive measures. The findings also show several potential benefits in using road surface forecasts and road friction data in terms of financial, socio-economic, and environmental perspectives.

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.