Current urban traffic congestion costs are increasing on account of the population growth of cities and increasing numbers of vehicles. Many cities are adopting intelligent transportation systems (ITSs) to improve traffic efficiency. ITSs can be used for monitoring traffic congestion using detectors, such as calculating an estimated time of arrival or suggesting a detour route. In this paper, we propose an urban traffic flow prediction system using a multifactor pattern recognition model, which combines Gaussian mixture model clustering with an artificial neural network. This system forecasts traffic flow by combining road geographical factors and environmental factors with traffic flow properties from ITS detectors. Experimental results demonstrate that the proposed model produces more reliable predictions compared with existing methods.Index Terms-Intelligent transportation system (ITS), traffic flow prediction, pattern recognition, artificial neural network (ANN), Gaussian mixture model (GMM) clustering. 1524-9050

It is important to guarantee the safety of vehicle to minimize the damage to the driver in case of the accidents. In order to inspect and later enhance the safety of vehicle, the owner of the vehicle usually have a vehicle safety inspection. The Korea Transportation Safety Authority (KOTSA) issues the Comprehensive Performance Inspection Certificate after vehicle inspection. The certificate only specify the legal inspection criteria for safety and measured values of the safety parameters, however, as ordinary driver in lack of expert knowledge about the vehicle is difficult to understand the contents of the Certificate. Thus, in this paper, the authors try to give the information about the inspection results in easier way to understand. This information not only guarantees the owner of the vehicle to better understand the inspection results, but it also gives the opportunity to the driver to deal with the specific problem listed in the results. The methods in this paper are to transform the vehicle inspection data into the non parametric distribution to easily represent the values to the index later on. Also, example indexes are presented to the actually inspected vehicle based on the reference distribution to show the better assessment of the developed method. †