**1. Introduction**

Since the 1960s, with the beginning of industrialization, Korea experienced a rapid urbanization [1–3]. As such, the transportation infrastructure has also developed steadily, greatly improving the mobility of citizens. However, in metropolitan areas, such as the capital city and its surrounding regions, the expansion of urban areas and overpopulation caused increased traffic and consequently other social problems, including traffic congestion and air pollution. To minimize such social costs, various policies to facilitate public transportation and taxis have been actively promoted [4]. Despite the efforts, the current services are short in providing an equal level of mobility to all residents, requiring enhancement. In particular, in Seoul Metropolitan City, despite its excellent public transportation infrastructure, there is a constant inconvenience of users with a need for enhancing problems caused by congestion during rush hours and a shortage of ridesharing (including taxi) supply late at night on the outskirts of the city and in hilly areas [5].

Against this backdrop and with the recent introduction of the mobility as a service(MaaS) concept, efforts have been made to improve individuals' mobility and accessibility through the integrated

use of shared and cutting-edge modes of transportation [6–8]. MaaS consists of various means of transportation, including public transportation and a mobile platform-based ridesharing service. However, the ridesharing service is usually owned by the private sector, oriented toward profitability. Therefore, the provision of such a service tends to be concentrated in areas with high demand. Even if the ridesharing service were to expand, those users living in areas with low demand, especially suburban areas, are expected to face inconvenient services, such as longer waiting times due to the lower availability of vehicles. Especially during late-night hours when public transportation no longer operates, there are only limited options available, thus leading to further exacerbation of the imbalance between supply and demand. Therefore, a solution for this problem is needed. In a previous study [9] on experiences with taxi service refusal, respondents answered that they had di fficulties getting taxi services or were refused during late-night hours (48.7%) when their destinations were either remote areas or outside of the city boundary (32.8%). From the perspective of users, this is recognized as a serious problem. The purpose of this study was to develop a dynamic pricing scheme to attract ridesharing (including taxi) drivers to mobility-disadvantaged regions during late-night hours when public transportation services are either reduced or come to a halt.

In the meantime, with the recent Fourth Industrial Revolution, the overall social paradigm has changed. A hyperintelligent society based on cutting-edge technologies, such as artificial intelligence, is being implemented. In the transportation sector, studies related to the introduction of big dataand artificial intelligence-based technology are growing [10]. Indeed, overseas ridesharing platforms, namely, Uber and Lyft, optimize supply and demand, as well as profitability, through AI-based dynamic pricing to retain the number of vehicles needed to respond to high demand in certain regions. Among the various subcategories of artificial intelligence, reinforcement learning has advantages in exploring unknown areas and identifying optimal outcomes through repeated exploration, unlike supervised learning, which is based on already available data. In contrast to other countries, there are not enough data related to ridesharing services in Korea and, accordingly, reinforcement learning would be more suitable. Consequently, this study conducted a reinforcement learning method using Seoul's taxi data to determine regionally appropriate levels of dynamic ridesharing fare rates with the purpose of improving the supply of ridesharing services in the mobility-disadvantaged regions in Seoul.

The paper is organized as follows: Section 2 reviews studies on the spatial fairness issues of transportation services and dynamic pricing applications for ridesharing services. Section 3 discusses the analysis framework and reinforcement learning method applied in this study. Subsequently, the results of the analysis are presented and discussed in Section 4, while the conclusions are drawn in Section 5.
