Analysis of factors affecting travel behavior by data mining: A case study of Kahramanmaraş

Transportation sector, which plays an instrumental role in societal, social, and economic frameworks, is intrinsically linked to the well-being and growth of societies globally. The sector's intricate network facilitates the movement of goods and people, consequently promoting accessibility to markets, employment opportunities, capital investments, and the spirit of healthy competition. This necessitates the development of the transportation sector, with an emphasis on effective planning and management to strike a balance between transportation demand and supply. Transportation demand and supply are inherently tied to the unique characteristics, preferences, and behaviors of travelers. Consequently, understanding the profiles and travel habits of these individuals is a prerequisite step in streamlining transportation planning. This study aims to decipher and interpret the underlying factors influencing travelers' transportation habits, utilizing data from the Household Travel Survey of Kahramanmaras Metropolitan Municipality Transportation Master Plan. Data mining, a powerful tool employed to extract valuable information from vast datasets, is the primary method used in this study. By deploying data mining techniques to analyse transportation data, it is possible to establish associations between numerous variables. These variables include travelers' ages, genders, car ownership status, economic conditions, starting and ending geographical points of their trips, and the types of transportation modes employed. For the purpose of this study, association rule analysis was conducted at two different levels: the household level and the individual level. These analyses were performed using the Orange software, a renowned data mining tool. The minimum support and confidence values of the generated association rules were calibrated to be appropriate for the specific dataset in use. The number of association rules obtained varied depending on the number of trips made. Upon examining the content of the rules, it was evident that the travelers' transportation mode preferences were associated with various factors. These factors spanned from demographic characteristics like age and gender, ownership of vehicles, to socio-economic characteristics such as economic status, and geographical characteristics such as the location of starting and ending points. The results derived from this study offer significant insights that can substantially contribute to the transportation planning process of Kahramanmaras Metropolitan Municipality. Besides, it serves as a valuable exemplar demonstrating the application of data mining methods for analytical studies in the transportation sector. The study underscores the potential of data mining as a powerful tool for understanding travel behaviour, offering transportation planners a scientifically grounded framework to guide decision-making and strategic planning. This study's innovative approach to understand travel behaviour through data mining not only contributes to the literature, but also lays the groundwork for future research in this area. It provides a comprehensive methodology that can be replicated in different contexts and geographical locations, further advancing our understanding of travel behavior and informing transportation policy worldwide. By identifying the intricate web of factors that influence travel behavior, this research contributes to a more sustainable, efficient, and inclusive transportation system. The findings have far-reaching implications, offering policymakers and urban planners invaluable insights into travel behavior, which can be leveraged to enhance transportation infrastructure, promote sustainable travel modes, and optimize the allocation of resources. Ultimately, the study paves the way for a more data-driven, evidence-based approach to transportation planning and management.