The study emphasises the importance of understanding individuals' mode choice between train and car to reduce CO2 emissions. It addresses two significant gaps in the existing literature, focusing on the impact of detailed attitudes towards mode use over time and the impact of train travel time components on actual mode use. The study reveals that environmental awareness and cost-consciousness drive increased train usage and reduced car usage, while status-sensitivity has no significant influence on mode use over time. Additionally, travelers place varying importance on different train travel components, with transfer time significantly impacting the modal share of the train. Furthermore, the perceived valuation of train travel time components emerges as a more reliable indicator of individuals' mode choice between train and car than the actual door-to-door train travel time. To encourage a shift from car to train use, the study emphasizes the need to encourage travelers to consider cost and environmental factors related to mode use, as well as the need to minimize transfer time and enhance the appreciation of transfer time

Efficient repositioning strategies for idle vehicles in ride-sourcing systems help reduce passengers' waiting time and drivers' operational costs, which help platforms attract more passengers and drivers. In this paper, we propose a decentralized repositioning strategy for drivers, in which drivers make individual decisions on where they reposition themselves, based on their own experiences. In comparison to an existing centralized repositioning strategy, in which drivers comply with reposition instructions provided by the platform, we examine the effects of the decentralized strategy on service rate, passengers' waiting time and drivers' net income. We compare the reposition strategies under different supply and demand levels and different demand spatial distribution dispersion rates. We also explore the influence of platform information on drivers' decision-making process.
We found that decentralized repositioning strategies have better performance in reducing waiting time, while the centralized strategy is better at increasing driver income and service rate. We also found that when platform information is accessible, the system has the best performance when 20% to 60% proportion of drivers utilize platform information when making decisions.

Compared to traditional public transport, ride-hailing makes it possible for people to get a more comfortable and faster riding experience with a higher fare. Ride-sharing fall in between the two, offering a discount at the price level of ride-hailing, yet operates with more detours and less comfortable experience. In this study, with different price levels for ride-hailing and discount rates for ride-sharing, we would like to examine the system performance of co-existence of ride-hailing, ride-sharing and public transport services. We would also like to search for an optimal solution for the ride-hailing & ride-sharing company to maximize its profit. We apply ExMAS, an open-source agent-based model for ride-sharing simulation, to simulate passengers' and vehicles' behavior on a microscopic level, and acquire numbers of results. Based on our model, in the case of Amsterdam, when price level is 1.1 euro/km and discount rate is 0.4, the company could enjoy maximum profit and market share. It is also found that, when price level gets higher more people opt for the competitive mode instead, resulting in the overall profit falling significantly.