Study On The Arrangement Of Commuter Activities

With the growth of population and number of vehicles, traffic jam is becoming a more and more serious problem. And the cause of traffic jam is that supply doesn't equate demand. Many studies imply that transportation demand management and proper transportation planning are effective ways in order to solve the problem. It is essential to forecast the demand for both ways, which is also the function of transportation demand forecast.The four step method plays the main role in travel demand forecast. In the four step method, we count and analyse personal traffic activities according to traffic areas which brings about models in traffic areas. In the process of transportation planning, we need to launch a large-scale traffic study which costs substantial human and financial resources if we choose the four step method. Besides, it is a method that research transportation demand from the macro perspective without considering personal travel behavior.The discrete models emerged when planners realized the necessity of developing complex models. It models travel behavior with individual as a unit. As a result, it precisely describes the decision-making process of an individual or a family.In the 1970s Oxford university scholars developed activity-based travel demand forecast method. In the third act of the international conference held in Australia in 1977, this method got approval and then it had wide public concern. Up to now, the method based on activities has become the most promising way as a method of traffic demand forecast. In the method based on activities, we consider that people travel to participate in economic activities and their fundamental aim is to travel in a different time and place for the completion of different activities. Therefore, activity is the basic of travel demand, and it affects the departure time, travel mode and the destination. When the activity based method is used to forecast travel demand, the activity variables are firstly analysed and then the travel variable is modeled by using the activity variables as parametric variables. Consequently, activity based method is capable of delineating the inherence relationship between traveler activity schedual and travel demand.Based on the third urban resident trip survey of bejing city, the commuter family attributes, the individual attributes and the commute activity attributes are analysed. It is found that the number of male commuters is more than female commuters. And the number of stops is more for male commuters comparing to the female commuters. Travel mode relates to depature time, number of stops and activity type closely to different extent.Relying on the result of above analysis, (1) a joint model of commute start time choice and number of stops is developed. The maximum likelihood function is defined and the calibration program is achived with the help of software of STATA. The result implies that both departure time and number of stops relate to the job beginning time variable and the travel mode variable. Positive relationship is also noted between departure time and number of stops. (2) The activity type choice model is developed. It is noted that age, job duration, number of children under six, travel mode and travel distance can significantly affect activity type choice. Afterwards, how to serve the traffic policy using the activity type model is discussed. (3) The commute duration model is developed using the AFT method. Assuming the basic risk function keeps to Weibull distribution or Exponent distribution, both models are calibrated and analised. The results imply that Weibull distribution can describe commute duration in a better way. At last, variables that impact commute duration are analysed.Based on urban resident trip survey, to do research on commute activity plays an important role in relieving traffic congestion in the city, improving the quality of commuters'travel, forecasting traffic demand, traffic planning and making transport strategy.