The previous research available to predict travel speed is wide and has been extensively studied. What currently is missing from the previous work is to estimate the travel speed when different non-recurrent events occur, such as car accidents and road maintenance work. This research implements a machine learning model to predict the average speed on a road segment with and without road accidents. The model would assist in (1) planning the most efficient route which could reduce CO2 emissions and travel time (2) the drivers in traffic could get an estimate of when the traffic will open up again (3) the authorities could take safety measures if drivers are expected to be stuck for too long. In our work, we conducted a review to determine some of the optimal machine learning models to predict on time series data. What we found by comparing GRU (Gated Recurrent Unit) and LSTM (Long Short Term Memory) on travel speed data over a road in Sweden provided by the Swedish Transport Administration, is that there is no major difference in performance between the LSTM and GRU algorithms to predict the average travel speed. We also study the impact of using weather, date and accident related parameters on the model’s predictions. What we found is that we obtained much better results when including the weather data. Furthermore, the inclusion of road events vaguely hints that it could improve performance, but can not be verified due to the low number of road accidents in our dataset.

Road network of a region is of a paramount importance in the overall development. Management of road traffic is a key factor for the city authority and reducing the road traffic congestion is a significant challenge in this perspective. In this thesis, a Wireless Sensor Network (WSN) based road-traffic monitoring scheme with dynamic mathematical traffic model is presented that will not necessarily include all adjacent intersections of a block; rather the important major intersections of a city. The objective of this scheme is to reduce the congestion by re-routing the vehicles to better performing road-segments by informing the down-stream drivers through broadcasting the congestion information in a dedicated radio channel. The dynamic model can provide with the instantaneous status of the traffic of the road-network. The scheme is a WSN based multi-hop relay network with hierarchical architecture and composed of ordinary nodes, Cluster-Head nodes, Base Stations, Gateway nodes and Monitoring and Control Centers (MCC) etc. Through collecting the traffic information, MCC will check the congestion status and in defining the congestion, threshold factors have been used in this model. For the congested situation of a road-segment, a cost function has been defined as a performance indicator and estimated using the weight factors (importance) of these selected intersections. This thesis considered a traffic network with twelve major intersections of a city with four major directions. Traffic arrivals in these intersections are assumed to follow Poisson distribution. Model was simulated in Matlab with traffic generated through Poisson Random Number Generator and cost function was estimated for the congestion status of the road-segments over a simulation period of 1440 minutes starting from midnight. For optimization purpose we adopted two different approaches; in the first approach, performance of the scheme was evaluated for all threshold factor values iteratively one at a time, applying a threshold factor value to define threshold capacities of all the road segments; traffic was generated and relative cost has been estimated following the model specifications with the purpose of congestion avoidance. In the second approach, different values of threshold factor have been used for different road segments for determining the optimum set-up, and exhaustive search technique has been applied with a smaller configuration in order to keep computations reachable. Simulation results show the capacity of this scheme to improve the traffic performance by reducing the congestion level with low congestion costs.

Road network of a region is of a paramount importance in the overall development. Management of road traffic is a key factor for the city authority and reducing the road traffic congestion is a significant challenge in this perspective. In this thesis, a Wireless Sensor Network (WSN) based road-traffic monitoring scheme with dynamic mathematical traffic model is presented that will not necessarily include all adjacent intersections of a block; rather the important major intersections of a city. The objective of this scheme is to reduce the congestion by re-routing the vehicles to better performing road-segments by informing the down-stream drivers through broadcasting the congestion information in a dedicated radio channel. The dynamic model can provide with the instantaneous status of the traffic of the road-network. The scheme is a WSN based multi-hop relay network with hierarchical architecture and composed of ordinary nodes, Cluster-Head nodes, Base Stations, Gateway nodes and Monitoring and Control Centers (MCC) etc. Through collecting the traffic information, MCC will check the congestion status and in defining the congestion, threshold factors have been used in this model. For the congested situation of a road-segment, a cost function has been defined as a performance indicator and estimated using the weight factors (importance) of these selected intersections. This thesis considered a traffic network with twelve major intersections of a city with four major directions. Traffic arrivals in these intersections are assumed to follow Poisson distribution. Model was simulated in Matlab with traffic generated through Poisson Random Number Generator and cost function was estimated for the congestion status of the road-segments over a simulation period of 1440 minutes starting from midnight. For optimization purpose we adopted two different approaches; in the first approach, performance of the scheme was evaluated for all threshold factor values iteratively one at a time, applying a threshold factor value to define threshold capacities of all the road segments; traffic was generated and relative cost has been estimated following the model specifications with the purpose of congestion avoidance. In the second approach, different values of threshold factor have been used for different road segments for determining the optimum set-up, and exhaustive search technique has been applied with a smaller configuration in order to keep computations reachable. Simulation results show the capacity of this scheme to improve the traffic performance by reducing the congestion level with low congestion costs.