Traffic Speed Estimation through Data Fusion from Heterogeneous Sources for First Response Deployment

During emergencies, the efficiency of first response deployment is critical. Once the assignments are decided for the distribution of first responders, the deployment efficiency for the teams to arrive at the affected zone is determined by the response time. Knowing the condition on the road network could substantially reduce the response time, in other words, increasing the transport efficiency for the deployment. On the other hand, real-time traffic data acquisition has been the core and basis of all development of advanced traffic-management systems. For the goal of measuring reliable traffic speed, the traffic data sources should generally include spot speed data received from vehicle detectors, space speed data collected by probe vehicles, and historical data to generate traffic information for main arterials within urban areas. This paper describes the fusion technique to integrate active and passive data from spot and space data for the estimation of traffic speed in emergency scenarios based on entropy and optimal weight. This provides road-network information to decision makers in emergency response. Through the proposed fusion process, the collection, fusion, and analysis of traffic speed data are performed. The process is composed of three consecutive computational steps. The first step is data screening to reduce inaccuracy of outliers in the system. The second step is to transform data into the same basis of space mean speed, and to classify and individualize the data. Data can then be converted into probabilities for the production of entropy. The third step is to apply the optimal weight rule to generate weight allocation for different data sources. The encouraging results from the data processing and sensitivity analysis revealed the potential to apply the proposed data fusion process to decision making for emergency response.