The spatial mismatch between demand and supply over time is a significant concern in a bike-share service. One primary strategy to fill the mismatch is to rebalance a shared bike fleet by vans or trucks. The more vans operators use to meet the demand, the more vehicle miles traveled (VMT) the system produces, offsetting the VMT reduction benefit that was at least a part of the motivation for the city to authorize the service. Another approach to the problem is an incentive-based approach to rebalancing a fleet. This approach incentivizes users to walk farther to get a bike from the oversupplied area (origin-based incentives) or to take a bike to the undersupplied area (destination-based incentives) by offering some reward, such as free bike-share use or a prize of some sort. This approach has proven to be successful in docked bike-share services, but the potential in the context of dock-less bike-share services is unknown. This dissertation examines the potential effect of an incentive-based approach to rebalancing a dock-less e-bike-share fleet on bike-share use and social benefits, focusing mainly on VMT reduction, using a e-bike-share service in the Sacramento area, CA.
This dissertation consists of four studies. The first and the second studies are self-standing. The third study is built on the second study, and the fourth and final study assembles an agent-based model from the models presented in the prior three studies. In the first study, I examine bike-share users’ willingness-to-walk to pick up a bike or drop off a bike at some distance from their origins or destinations if rewarded and identify characteristics influencing willingness-to-walk. I find that half of the respondents use bike-share if the available bike is located 8.9 minutes away. Estimates of willingness-to-walk farther than the mean distance for incentives at origins and destinations were 3.8 minutes and 4.2 minutes per dollar, respectively. Based on these results, I find the potential effectiveness of incentives as a strategy for spatially rebalancing bike-share fleets.
The second study examines the factors influencing mode substitution, defined here as the mode that is replaced when bike-share is used. I find that walking is the dominant mode substitution for trips of less than 1 mile for most trip purposes. Long trips and non-commute trips that start at non-commercial locations are likely to represent car substitution and some groups, such as women, non-membership holders, and those who have a private car, are more likely to report car substitution for any trip purpose.
The third study develops a framework for estimating vehicle miles reduced from the introduction of a dock-less e-bike-share service. I find that the daily car substitution rate, including both “private car” and “ride-haling,” was 28% on weekdays. Furthermore, I find that the dock-less bike-share service with a fleet size ranging between 950 and 1100 was responsible for an estimated VMT reduction of 2,131 vehicle miles per day in total and 0.79 miles per trip on average across the service region on weekdays.
The fourth and final study develops a simulator using an agent-based model to examine how an incentive-based approach helps reduce the spatial mismatch between demand and supply and to estimate impacts on VMT reduction and its social cost in the context of a dock-less e-bike-share service. I find that incentive strategies improve the willingness of bike-share users to go out of their way to pick up or drop off a bike, but the effect varies by fleet size and the size of the incentive budget. The number of trips per bike does not change significantly with rebalancing strategies, suggesting that operators must determine the fleet size carefully before entering the market. I estimate that introducing incentive strategies reduces VMT by 3-6% by increasing the number of bike-share trips and saves US$20-29 in social costs per day. Based on the first three studies, this study demonstrates the potential of the rebalancing strategy's incentive approach to increase the bike-share operation's efficiency and social benefits regarding VMT reduction.