FORECASTING TRIP ATTRACTION BASED ON COMMERCIAL LAND USE CHARACTERISTICS

Pretina George , Gymmy Joseph Kattor

Abstract: Trip Attraction is fundamental in planning of transportation facilities. It is important to the traffic engineer and urban planner in considering the impact of new development such as office complex, shopping center and residential development. This paper focuses its study on the formulation of a trip attraction model using multiple regression technique for the commercial land use in medium sized towns of Kerala. Questionnaire surveys are conducted for the eight selected commercial nodes from the three medium sized towns Tirur, Perinthalmanna, and Ponnani in Kerala. Socioeconomic surveys are conducted for the selected towns for obtaining the origin-destination data. The correlation and regression analysis are performed based on these surveyed data. The study showed that the multiple regression model with the three independent variables namely the number of commercial establishments, percentage of shops in the commercial node and percentage of banks in the commercial node with the R2 and Adjusted R2 value of 0.991 and 0.9844 respectively gives the better estimate of trip attraction. After model formulation, the model accuracy was also tested by examining the assumptions of multiple regression. The model is found to be accurate during the assumption examination process. This model would be very useful for estimating the trips attracted to a new or existing commercial center in any medium sized towns in Kerala, and thus aid to assess the traffic impact of the commercial center on the geometric design of roadways in the surrounding area.

Keywords: correlation, multicollinearity, multiple regressions, trip attraction

DOI: https://doi.org/10.15623/ijret.2013.0209072