Real-time prediction model of arrival time for floating transit vehicle

According to the real-time GPS data of floating transit vehicle,the effects of road section average speeds,bus stations and traffic lights were considered,and a novel prediction model of bus arrival time was proposed.The precision of prediction model was improved by estimating arrival time to the nearest downstream station and judging the effectiveness of GPS data on raod section.The prediction model was tested and verified by using the transit vehicle GPS data of Chongqing City.Calculation result indicates that the arrival time of floating transit vehicle to the downstream station can be real-time predicted by using the model,the prediction precision is better than the existing methods,the prediction error is less than 9% in peak periods and is about 6% in non-peak periods,and the prediction model has good adaptability to the various of traffic conditions.3 tabs,8 figs,12 refs.