Identifying potential market of advanced ride-hailing services with taxi-to-taxi transfer using a latent class choice model

Abstract

This study examines people’s preferences for advanced ride-hailing-based taxi platform services aimed at addressing late-night taxi shortages. These new services are defined as a combination of taxi-to-taxi transfers, ride-pooling, pre-booking, and car type. A stated-choice experiment is designed to assess people’s willingness to use various hypothetical services. The respondents are limited to those using taxis at night in Gangnam, Seoul, with a sample size of 1000 individuals. A latent class choice model is estimated using stated choice data to identify the potential market for new ride-hailing services. The analysis reveals that the market can be categorized into three subgroups, with significant heterogeneity in preferences for new services among these subgroups. The potential market for new services is associated with individuals’ socio-demographics, such as gender, main activity time, and frequency of taxi usage. Among the two groups likely to choose alternative taxis, one group is sensitive to fares and shows relatively low resistance to short transfer waiting times and advance reservations, suggesting that an appropriate fare discount could increase their likelihood of using the new services. The other group prioritizes guaranteed dispatch over fare discounts and exhibits relatively high resistance to time-related factors, indicating that a strategy focused on providing high-quality services with enhanced time flexibility could be effective. Notably, the two groups show distinct differences in their preferences for ride-pooling. To the best of our knowledge, this study is the first to identify the latent market for new taxi services, including taxi-to-taxi transfers. The results provide insights into the factors affecting taxi service decisions, including taxi-to-taxi transfers and ride-pooling.