Analysing and visualising bike-sharing demand with outliers

TL;DR

This paper presents a methodology for detecting demand outliers in bike-sharing data using clustering and functional depth analysis, enhancing forecast reliability and urban mobility planning.

Contribution

It introduces a general outlier detection approach combining clustering and functional depth analysis, with visualizations and managerial insights for bike-sharing demand analysis.

Findings

Effective outlier detection improves demand forecast accuracy.

Visualizations reveal demand patterns and anomalies.

Method is applicable beyond the Washington D.C. dataset.

Abstract

Bike-sharing is a popular component of sustainable urban mobility. It requires anticipatory planning, e.g. of station locations and inventory, to balance expected demand and capacity. However, external factors such as extreme weather or glitches in public transport, can cause demand to deviate from baseline levels. Identifying such outliers keeps historic data reliable and improves forecasts. In this paper we show how outliers can be identified by clustering stations and applying a functional depth analysis. We apply our analysis techniques to the Washington D.C. Capital Bikeshare data set as the running example throughout the paper, but our methodology is general by design. Furthermore, we offer an array of meaningful visualisations to communicate findings and highlight patterns in demand. Last but not least, we formulate managerial recommendations on how to use both the demand forecast and the identified outliers in the bike-sharing planning process.