Predictive logit model of trip mode with fuzzy attribute variables

Based on the disaggregate model and fuzzy mathematics theory,the trip behaviors of residents in urban agglomeration were taken as study subject,the trip time and the trip cost were taken as influence factors,and the parameters were calibrated by the maximum likelihood estimation method.Through t test,hit rate test and fit goodness test,the trip time was fuzzed,the influence of trip cost was ignored,and a predicative logit model of trip mode with fuzzy attribute variables was established.The fuzzy parameters of trip times for rail transit and car were chosen as 0.1,0.3 and 0.5 respectively,the influences of trip mode and trip time on trip behavior for residents were analyzed.Analysis result shows that the average trip perception time ratio of rail transit and car is between 0.8 and 1.2,and the two trip perception times change in equal degree.When the fuzzy parameter of trip time for rail transit is 0.1 and the trip time of car is less than 70 min,all the residents will choose rail transit.When the fuzzy parameter of trip time for rail transit is 0.3 and the trip time of car is less than 67 min,residents still choose rail transit,but when the trip time of car is more than 67 min and the fuzzy parameter of trip time for car is 0.1 and 0.3 respectively,residents will choose car.When the fuzzy parameter of trip time for rail transit is 0.5 and the trip time of car is less than 58 min,residents still choose rail transit,while the trip time of car is more than 66 min,all the residents choose car.