Research On Method For Traffic Jam Forecasting And Leading-decision

With increase of vehicles, traffic congestions in the big cities of the world are more deteriorated, which increase travel time, the fuel fee of automobiles, traffic accidents, and deteriorate the environment, restrict the city traffic develops well, as a result, how to catabate traffic congestions drastically has been a key means to resolve the problem of city traffic. Predict the degree of traffic congestion in the time-space road network scientifically in order to master the traffic conditions of all sites at all time in the future to induce the traffic flow reasonably ,and more drivers select the shortest route for the catabation of the traffic congestion has important practical significance. The paper researched the routine traffic congestion, firstly introduced the time-space distribution characteristic of traffic flow, its object is to make traffic congestion degree of the road network be minimal, probed into the theory and method for prediction the traffic congestion degree of the time-space road network as well as the leading and decision method. The paper researched the following parts mainly:1. On the basis of reviewing the traffic congestion research literatures, expatiated the basic conception and attribute of traffic congestion, summarized the time-space distribution characteristic of traffic flow, probed into the feasibility to predict traffic congestion, analyzed the correlated factors of traffic congestion, such as the capabilities of sections and crosses, transfer and restriction of the traffic flow.2. Contrasts the difficulty to obtain and predict all the traffic flow parameters, selected all the parameters to estimate traffic congestion of the road network, build the index system and fuzzy integrated evaluation model to determine the traffic congestion degree of time-space road network.3. Built the travel time model for the situation when there is signal cross or nonsignal cross, puts forward the double sites and multi-sites models for short term traffic flow prediction, build the P-S-F model to predict traffic congestion degree of time-space road network, the model can be used to predict the traffic congestion degree of single section, single cross, all sections, all crosses and the road network, which provides scientific theory basis for inducing the crowded traffic flow.4. For traffic flow is uncertain, the paper built the fuzzy grey model to lead traffic congestions based on entroy weight analysis method, and the decision frame for leading traffic congestions based on equilibrium distributing theory of traffic flow.The innovations brought out in this paper are as follows:1. Selected all the parameters to estimate traffic congestion of time-space road network, and build its index system with analytic hierarchy process.2. Built the fuzzy integrated evaluation model to estimate traffic congestion degree of the time-space road network, which provides scientific bases to describe the traffic congestion degree of the time-space road network quantificationally.3. Built the P-S-F model to predict traffic congestion degree of time-space road network, the model can beused to predict the traffic congestion degree of single section, single cross, all sections, all crosses and the road network, which provides scientific theory basis to induce the crowded traffic flow.4. Built one fuzzy grey model to lead traffic congestions based on equilibrium distributing theory of traffic flow.5. Built the decision frame for leading traffic congestions based on equilibrium distributing theory of traffic flow.