Modal Split Model Using Multiple Linear Regression Analysis

Abstract

Several modal split models have been created around the world to forecast which mode of transportation will be selected by the trip -maker from among a variety of available modes of transportation. This modeling is essential from a planning standpoint, as transportation systems typically receive significant investment. In this study, the main purpose was to develop a mode choice model using multiple linear regressions for Ramadi city in Iraq. The study area was divided into traffic analysis zones (TAZ) to facilitate data collection. The data was collected through a home interview of the trip makers in their home units through a questionnaire designed for this purpose. The result showed that the most influential factors on the mode choice for the general trips model using multiple linear regressions are car ownership, age, and trip cost. This model gave a good correlation coefficient of 0.829 meaning that the independent variables explain 82.9 of variance in the dependent variable (type of mode), which will help transport planners in developing policies and solutions for future.