London’s public transport authority, Transport for London (TFL), releases hourly Santander Cycle trip logs dating back to 2015. The dataset, split into 144 weekly CSVs, records every ride from 800 stations, totaling 9.2 million station‑hours. Developers can download and convert the files to Parquet for efficient analysis.

Researchers harness the data to probe how major tube strikes alter cycling behaviour. By mapping stations within 400 m of affected Underground lines and aggregating trips to hexagonal H3 cells, they create a binary treatment variable indicating strike exposure. The log‑transformed per‑station trip count serves as the outcome metric.

After filtering cells with at least one tube stop within 500 m to satisfy the positivity assumption, the final panel contains 62 H3 cells and 66,039 daily observations. Each row records total trips, station count, weather, and holiday flags, enabling regression models that isolate the strike effect from seasonal noise using difference in log‑rate estimates.

By converting raw CSVs to a tidy Parquet format, joining spatial coordinates to Uber’s H3 grid, and applying a rigorous causal framework, the study demonstrates that tube disruptions trigger measurable upticks in bike usage. The methodology offers a template for urban planners seeking data‑driven insights into multimodal transport resilience for policy and resource allocation. This analysis.