Trip Generation Based on Land Use Characteristics: A Review of the Techniques Used in Recent Years

The inadequate transportation facilities and growing population reflect the necessity of travel demand forecasting in developing nations. Forecasting travel demand serves as the foundation for planning and policy development that helps a country's economy grow. The trip generation process is the first step in the travel demand modelling process.. The accuracy of modelling trip generation is heavily reliant on the accuracy of two stages- data collection stage and generation of the model which depends on different modelling techniques used. The first and most important step in Trip Generation is data collection. Data from household trips is critical for both managing the existing transportation network and planning and designing future facilities. For many decades, household travel surveys (HTS) have been used as a time-consuming and costly method of data collection. New technologies emerge as alternatives to HTS as time passes, but HTS remains the most commonly used technique in developing countries. Data analysis techniques are the second important step in Trip generation modelling. Unlike existing modelling techniques, i.e., regression and category analysis, new modelling trip generation techniques involving machine learning have been developed in developing countries in recent years. A brief overview of data collection and modelling techniques used in developing and developed countries is provided as a comparative study.