Bike sharing systems: a new incentive rebalancing method based on spatial outliers detection

Abstract

Since its launch, Velib' (the Bike Sharing System -BSS-in Paris) has emerged in the Parisian landscape and has been a model for similar systems in many cities. A major problem with BSS is the stations' heterogeneity caused by the attractivity of some stations located in particular areas. In this paper, we focus on spatial outliers defined as stations having a behavior significantly different from their neighboring stations. First, we propose an improved version of Moran scatterplot to exploit the similarity between neighbors, and we test it on a real dataset issued from Velib' system to identify outliers. Then, we design a new method that globally improves the resources' availability in bike stations by adapting the users' trips to the resources' availability. Results show that with a partial collaboration of the users or a limitation to the rush hours, the proposed method enhances significantly the resources' availability in Velib' system.

A New User-Based Incentive Strategy for Improving Bike Sharing Systems’ Performance

“…The proposed method improves the overall BSS performance by minimizing the number of abnormal stations, also called spatial outliers stations. This paper extends our earlier work in which we analyzed the data provided by Velib and suggested two methods to enhance the performance of the Paris' BSS [6]. When analyzing the data, we took into account the spatial attributes of a station (i.e., latitude), its non-spatial attributes (i.e., capacity), and its dynamic attributes for different timestamps t (i.e., the number of available bikes).…”

“…This would improve the performance of Velib's service without the extra cost of moving some bikes (e.g., by truck) to rebalance the stations. Using a modified version of the Moran scatterplot, we were able to show that most of the stations were locally heterogeneous [6]. In other words, most of the outliers were surrounded by balanced stations.…”

“…Recall that an isolated station satisfies the following condition: its status is significantly different from the state of the neighboring stations in terms of the filling rate at a given time t. Thus, isolated stations are considered spatial outliers. Six spatial algorithms will be considered in this paper: Iterative r [53], Iterative z [53], Median [53], Z-score [52], Moran Scatterplot [50], and Modified Moran Scatterplot [6].…”

“…To assess the efficiency of our proposed strategy, we compare it with the following three scenarios: (1) no rebalancing strategy is applied, (2) the static rebalancing strategy [5] is used, and (3) the rebalancing strategy based on the rush hours is performed [6].…”

Learning incentivization strategy for resource rebalancing in shared services with a budget constraint

“…Others have proposed solving the rebalancing problem without any explicit incentive mechanism. Chahchoub et al [11] proposed the use of identifying outliers (nearly empty or nearly full bike stations). Users are proposed an alternative destination or starting points within a small area to ensure the stations are no longer outliers.…”