The Research On Preactive Traffic Management Plan

The number of motor vehicles is increasing with the continuous development of urban and limited road resources. Urban traffic environment is getting worse and traffic congestion has become in many urban development needed to solve the problem.Traffic congestion leads traffic participants to losses of time and economic, at the same time,low speed can produce more car exhaust and noise emissions to make urban environmental quality wores. To deal with these problems, more and more traffic management measures have been proposed. Predictive traffic management plan is a more effective method to solve urban traffic congestion, the method uses the current traffic data to predict the future traffic situation accurately and can propose the traffic management plan in advance to be as a strategies to solve traffic congestion for the traffic management department.Predictive traffic management belongs to the subsystems of ITS in thi paper. With the help of dynamic traffic package-DYNAMEQ which is a the med- simulation package combined with dynamic traffic assignment theory and traffic management concepts to bulid a simulation model,we can preditc and simulate the furture traffic conditions and make traffic organization and optimization to the congested area in advance. This paper studies the content of the following two aspects:Given the difficculty in obtaining time-vary OD matrix and to deal with the deficiency problem in solving the open-source, mesoscopic dynamic traffic assignment package NEXTA-DTALite's OD-estimation model, we first make a conversion of the OD matrix estimation model to construct a new objective function. Then we use open-source linear programming program GLPK to solve OD estimation model. By using the open-source characteristic of CSV file format of reading and writing, we integrate NEXTA-DTALite and CLPK of seamless data exchanging. We propose the OD-estimation algorithm by the combination of NEXTA-DTALite and GLPK to solve the dimension explosion problem. Also the method enhances the of open-soure domain of the NEXTA-DTALite package in solving the OD-estimation model.we use the OD matrix which obtained in above algorithm as the initial matrix in DYNAMEQ.Second, we use the DYNAMEQ platform to build a regional road network model in Changsha and to analyze the detailed simulation results and choose the important MOE to find the reason of traffic congestion after the model calibration.By changing the key intersection's signal timing and compared the optimized sumulation results and the before results to find out the optimization of the MOE and the effectiveness of the traffic management plan.