Developing Models for the Requirement of Travelling by Public Transportation which was Varied by Time

Abstract

This study aimed to, 1) Analyze factors which influenced the number of students who used public transportation at Suranaree University of Technology. 2) Present the process of developing models of forecasting the quantity of people who used public transportation within any 15 minute period at Suranaree University of Technology. The study area is divided into 2 zones. Zone A is a school building and a group of office buildings. Zone B is a group of student dorms. It uses multiple regression analysis. Poisson regression analysis Negative binomial regression analysis in model development and test the accuracy of each model to allow the most accurate model to be identified and selected. And to review the accuracy of the model to select the most accurate model. The results show that the model using Negative Binomial Regression Analysis for model development is the most accurate. The factors which affected the numbers of people who used public transportation of zone A, such as 15 – 30 minutes after changing class, the numbers of student who were going to study their next class within next 15 – 30 minutes and periods where students do not have any class for more than 2 hours. The variables which influenced the quantity of people who used public transportation of zone B, such as, 15 – 30 minutes after changing class, the total number of students who registered to learn within each hour, the number of students who were going to their next class within 15 – 30 minutes and periods where students have no class for more than 2 hours.