Optimizing environmental sensing on public transport to cover broader demographics in Amsterdam

Abstract Urban senrsing systems often have uneven spatial coverage, with sensors concentrated in few areas, reinforcing social and environmental inequalities. This study explores how public transport vehicles can serve as mobile environmental sensors not only to increase the spatial coverage of sensing strategies but also to create a fairer and adaptive monitoring system. Using Amsterdam as a case study, we combine GPS locations of public transit vehicle data with 100 × 100 m sociodemographic data to assess and optimize fair sensing coverage. Our method selects the minimum number of vehicles to balance demographic fairness, spatial reach, and temporal frequency. Results show that six vehicles operating during one day can ‘fairly’ represent 17.6% of residents within the applied direct-sensing range, compared with 0.2% for the six static PM₂.₅ sensors deployed in Amsterdam. By redistributing mobile sensors, cities can move from unequal monitoring toward an equitable, city-wide environmental sensing framework.