Prediction of the Urban Travel Demand by Activity-Based Method: Case study: 3 Regions in City of Tehran

The first travel demand forecasting models were created on an aggregate basis since 1950s. These models included four independent modeling steps: trip generation, trip distribution, modal split and traffic assignment. However, after observing the shortcomings of this process in analysis of strategies and incorrect results of demand forecast, development of disaggregate methods which have a behavioral basis was welcomed by planners and decision makers of transportation industry since 1970s. After testing many methods such as travel chain, activity based methods emerged in which the activity is considered as the factor causing the travel. Based on this, in most of the large cities of the world and especially in the United States of America, activity based models were created and the results were evaluated and implemented. Nowadays, many of the large cities of the country are facing problems related to management of traffic demand. In this paper, the required information was drawn from the questioning of inhabitants of Tehran performed by Tehran Comprehensive Transportation & Traffic Studies Company. By applying the activity based structure to the existing data bank, the activity based data base was prepared and by univariable, two variable and multivariable analyses of this data bank, the factors affecting the utility functions of daily activity pattern, time and vehicle of the primary tours were identified. The three independent, nested and joint decision making structures were created and analyzed by logit models. Considering the analyses performed, the independent structure was evaluated as more suitable for the sample of the study and by using that, a general knowledge of the daily activity planning of individuals was provided.