Journal of Korean Society of Transportation ISSN:1229-1366(Print) 2234-4217(Online) 대한교통학회지

In recent times, a population concentration in major cities has been observed, while simultaneously, rapid phenomena of low birth rates and aging are leading to a decrease in the population of local towns and villages in South Korea. This trend is increasing the proportion of areas suffering from transportation exclusion, marked by reduced mobility and accessibility for residents. To improve the mobility of residents in transportation-excluded areas, a public taxi service is being introduced. For sustainable and efficient transportation services, it is essential to understand demand and supply and develop appropriate service introduction strategies. This study utilizes operational data from public taxis in 32 city and county areas to analyze current operations and characteristics, and proposes operational strategies for service management plans. The findings indicate that the proportion of the elderly population is higher in county areas (31.5%) than in city areas (20.4%), with the highest utilization (10.9%) of the service relative to the beneficiary population in Yecheon County, which has a high elderly population ratio. Additionally, areas with long-distance travel tend to have more single-passenger rides, as users often travel to individual destinations. Using GIS Pro to analyze boarding and alighting points, demand concentration areas, and service zones, it was found that the main boarding and alighting points were residential areas (34%), downtown (18%), and community centers (14%). Some of the demand within identified clusters could not be included within a 5 or 10-minute influence radius from the center, due to geographical features and road structures. Furthermore, most areas showed a pattern of returning at various times after morning peak hours from 8 am to 10 am. Ultimately, based on the analysis, Decision tree model was proposed that categorizes clusters based on three months of demand data and decides on the service operation model for each cluster. Such a practical strategy approach using real data is expected to contribute to the efficient and customized planning and operation of transportation services in excluded areas.