• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.6
• /
• pp.118-133
• /
• 2020

Research interest in the Mobility-as-a-Service (MaaS) concept for enhancing users' mobility experience is increasing. In particular, dynamic pricing techniques based on reinforcement learning have emerged since adjusting prices based on the demand is expected to help mobility services, such as taxi and car-sharing services, to gain more profit. This paper provides a simulation framework that considers more practical factors, such as demand density per location, preferred prices, the distance between users and drivers, and distance to the destination that critically affect the probability of matching between the users and the mobility service providers (e.g., drivers). The aforementioned new practical features are reflected on a data structure referred to as the Matching Matrix. Using an efficient algorithm of computing the probability of matching between the users and drivers and given a set of precisely identified high-demand locations using HDBSCAN, this study developed a better reward function that can gear the reinforcement learning process towards finding more realistic dynamic pricing policies.

• Journal of Korean Society of Transportation
• /
• v.23 no.2
• /
• pp.151-160
• /
• 2005

In case of public transit such as bus system, the probability concept is used to evaluate the Level-of-Service of the operations. And each levels could be classified according to the linear probability value. (TCQSM: Transit Capacity and Quality of Service Manual-2nd Edition, TRB, Washington DC., 2003) In this case, the drivers or passengers wouldn't think that the service level isn't equivalent to the linear probability value. Thus the linear probability value doesn't exactly reflect the service level. This study shows the problems of using the linear probability value in classifying the service level through the case of evaluation of bus operation's punctuality, presented in TCQSM. To make up for the problems of such case, two methodologies are presented in this study. The method of determining Level-of-Service criteria using probability density of headway variation's distribution, presented in this paper, adequately reflects passenger's expected waiting time. According to the application result to real bus operation data, it tis better than the method of TCQSM to evaluate the reliability of bus operations. However further research about the relations between utility difference and passenger feeling of service level in necessary to apply the method that uses the utility function. It remains as the limitation of this paper.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.8B
• /
• pp.695-705
• /
• 2012

Due to recent advances in wireless devices and the automotive industry, Vehicular Ad hoc Networks (VANETs) have emerged as a very promising technology for transferring data collected on the road by moving cars. The delivered data may contain emergency information which affects the safety of passengers and drivers as well as the traffic congestion, and the routing protocols have thus a significant impact on the performance of VANETs. In this paper, we study the impact of movement direction of the participating cars which forward data packets on the performance of data delivery and present a new approach which extends the contention-based forwarding (CBF). The proposed reliable CBF (R-CBF) increases the reliability of data deliver on the traffic lights installed roads and reduces the overall propagation delay without routing loops or interruption of data forwarding that may be caused by changes of relative positions of involving cars in routing. Simulation demonstrates that the R-CBF diminishes propagation delay by 38% in comparison to G-SRMB which forwards data to moving cars in the backward direction and eliminates unnecessary retransmissions.