Research On Combined Model Of Traffic Distribution And Assignment Based On Fuzzy Travel Demand

Traffic demand forecasting process is the basis of traffic planning, and theforecasting results will directly as an analysis basis for the decision maker to makedecisions. In the traditional four-stage travel demand forecasting method, everyforecasting stage is independent of each other, so the predict process will requiredlots of work, but also some discrepancies with the actual situation. The combinedprediction model can combined those closely predicted stages into a single processto study, so the combined model shortening the forecasting process and reducingthe prediction error. Currently, all the combined models are assume that traffictravel demand is accurate, without taking into account the travel fuzzy uncertainty,it will leading to a large error between the prediction results and the actual value,so the prediction results will result in higher decis ion-making risk. In this context,constructing combined model under the fuzzy travel demand is particularlyimportant.Firstly, analysis of the uncertainty of travel demand, inc lude the fuzzy anduncertain factors and the transitivity of uncertainty in traffic demand forecastingprocess.Secondly, through the analyzed of trip generation of fuzzy uncertainty andfactors in the traditiona l trip generation forecasting model based on fuzzy settheory, considering the factors that affecting trip generation input ambiguity, thenconstruct fuzzy trip generation forecastin g model and a numerical examp leillustrates the valid ity of the model.Finally, based on a comb ination of traditional Logit model, taking intoaccount the fuzzy trip generation, the traditiona l combined model is to beimproved and proposed the new model co mbined traffic distribution and trafficassignment based on the fuzzy trip generation. Then, at a certain confidence leve l,changed the programming model into a general problem, combined with thetraditiona l combination of model algor ithm, gives the new mo del a specificalgorithm and design an effective example of fuzzy combination of travel demandmodel, and analys is the effectiveness of the algorithm verification, the new modelpredictions to some extent, improve the prediction reliability, reduce the ris k ofuncertainty decisions.