Characteristic and Prediction Model for Operating Speed Model in Urban Underground Road

Abstract:

Real car experiments on eight urban underground roads in Shanghai Municipality were conducted in this paper. A total of one hundred and fifty six measured data were used to fit four presented characteristic and prediction models. The stepwise linear regression method was employed to predict the proper 85th percentile operation speed on urban underground road. Four regression models were used to investigate the relations between operating speed and corresponding roadway factors including lane width, slope grade, number of the lanes and speed limit, etc. The prediction made by using the presented models can meet real car experimental results on a ninth urban underground road very well, which indicates the applicability and reasonability of the models. In addition, the speed limit was found to have very small effect on the operating speed in underground roads. The models put forward in this paper are believed to be useful for operating speed prediction for urban underground road with four to eight lanes at a design speed of 40 to 80 km·h-1.