Abstract
Commuting traffic associated with the “9 to 5” workday shaped the morning and evening peaks across the world. The COVID‐19 pandemic led to unprecedented changes in travel behaviour such as an increase in cyclists and telecommuting, where employees worked from home during lockdown periods. Transport modellers, planners and policy makers need to know whether the 9 to 5 has returned, or we have entered a “New‐normal” of more flexible working arrangements and increased cycling, key for delivering sustainability targets. In this research, the unsupervised machine learning technique k‐means clustering investigates temporal patterns across the day and week, comparing the pre‐ and post‐pandemic era across both motorised vehicles and bicycles. Results show that the total daily traffic flow has returned to pre‐pandemic volumes, but more spread across the day. Mondays and Fridays have less‐pronounced peaks compared to pre‐pandemic, having implications for air quality modelling and assessment, traffic management and transport planning. Meanwhile, cycling has increased in volume and the time‐of‐day people are travelling has changed. Policy makers need to consider whether the additional capacity on the road, brought about by reduced peak traffic, could be reallocated to make roads safer for and reduce delay to cyclists, contributing towards net zero goals.
Original language | English |
---|---|
Journal | IET Intelligent Transport Systems |
Early online date | 8 Sept 2024 |
DOIs | |
Publication status | E-pub ahead of print - 8 Sept 2024 |
Keywords
- demand forecasting
- bicycles
- road traffic
- pattern clustering