A Travel Behaviour Model To Predict Travel Behaviour Pattern Of Urban Road User Using Rule-Based Approach

Abstract

World’s population has rapidly migrated to urban areas and prefers to use private cars instead of public transport, which results in high volume of car users in urban cities. In Malaysia, most research related to travel behavior is based on traditional data collection methods which include face-to-face meetings and questionnaires. However, traditional methods lead to human errors ending in inaccurate data as well as false, resulting in models. Survey-based data collection is perception-based as it does not include the actual travel data. In this research, we have proposed GPS method of data collection, which represents the actual travel time of the road user. Moreover, these methods give confined accuracy and necessitate respondent conformity apart from being expensive and a burden towards participants. The current situation mandates a model that predicts and visualizes travel behaviour via advanced methods such as collecting data via smartphone (GPS) applications. An extensive literature search using the keywords travel behaviour model, GLT data, machine learning algorithms and travel patterns was to propose a visualization and prediction model. The proposed model is based on the GLT data and questionnaires which will be preprocessed using data mining techniques such as the rule-based algorithm to study the raw data obtained via GPS dataset and SPSS analysis to verify and validate the questionnaire details obtained. The performance proposed model will be benchmarked with existing models that use traditional methods to analyze data from travelers. This concept paper provides further opportunity to visualize travel behaviour patterns and its demand to assist policymakers and route planners to improve urban lifestyles.