From Counting Stations to City-Wide Estimates: Data-Driven Bicycle Volume Extrapolation

TL;DR

This study develops a machine learning approach to estimate city-wide bicycle volumes using sparse counting station data combined with various public datasets, aiding urban planning and cycling advocacy.

Contribution

It introduces a data-driven method that extrapolates bicycle volumes across an entire city using diverse public data sources and machine learning, notably XGBoost, improving accuracy with short-term counts.

Findings

XGBoost outperforms other models in prediction accuracy.

Crowdsourced cycling and infrastructure data are most influential.

Short-term counts significantly reduce prediction error.

Abstract

Shifting to cycling in urban areas reduces greenhouse gas emissions and improves public health. Street-level bicycle volume information would aid cities in planning targeted infrastructure improvements to encourage cycling and provide civil society with evidence to advocate for cyclists' needs. Yet, the data currently available to cities and citizens often only comes from sparsely located counting stations. This paper extrapolates bicycle volume beyond these few locations to estimate bicycle volume for the entire city of Berlin. We predict daily and average annual daily street-level bicycle volumes using machine-learning techniques and various public data sources. These include app-based crowdsourced data, infrastructure, bike-sharing, motorized traffic, socioeconomic indicators, weather, and holiday data. Our analysis reveals that the best-performing model is XGBoost, and crowdsourced cycling and infrastructure data are most important for the prediction. We further simulate how collecting short-term counts at predicted locations improves performance. By providing ten days of such sample counts for each predicted location to the model, we are able to halve the error and greatly reduce the variability in performance among predicted locations.