Identifying Meteorological Determinants Associated with Roadway Crash Severity in Dhaka Metropolitan Area of Bangladesh: An Econometric Analysis over a Decade

This study investigates the impact of urban meteorological factors on road crash severity in Dhaka, Bangladesh. Using police crash data, and meteorological data from NASA POWER database for years 2011–2022, a generalized ordered logit model was used to analyze crash severity, and interpreted using odds ratio, log odds ratio, predicted probabilities and marginal effects. The results show that land surface temperature (LST), relative humidity, precipitation, surface pressure, and wind speed have significant association with crash severity. Relative humidity, surface pressure and LST exhibited positive relation with higher severity levels of crashes, whereas precipitation had a negative relation. We recommend three actions to lessen the severity of crashes during inclement weather based on the findings: (i) weather-responsive transport safety policies, which incorporate real-time weather data into intelligent transport systems; (ii) law enforcement-oriented policy implications, which include using automated speed cameras and red-light violation cameras to improve compliance consistency and updating driver training courses to include modules on risk perception across various environmental conditions; and (iii) infrastructure and vehicle-related policy implications, which include designing road geometries and surface conditions to prevent the effects of adverse weather conditions and utilizing safety equipment, such as electronic stability control and anti-lock braking systems.