Multinomial Logit Approach to Modelling Crash Severity on Selected Highways in Ondo State Nigeria.
Olajide Olaniyi Ipindola
Department of Civil Engineering
University of Ibadan, Nigeria
Email: olajideipindola@gmail.com
ABSTRACT
Reducing the frequency and severity of road crashes has been the major goal of transportation agencies in developed and developing countries, considering their impact on the economy. Road transportation plays a
major role in the socioeconomic development of a developing country like Nigeria with over 80% of her freight transported by road. However, the safety of these public infrastructures, particularly two-lane single carriage
highways remain a public health and social concern. Investigation of significant causal factors of crashes is a critical step to improve safety of these highway facilities. This study employed multinomial logit model to
identify the significant factors influencing crash severity on selected two-lane single carriage highways in Ondo state Nigeria using a Four-year crash data extracted from Federal Road Safety Corps database. The
model identified 85th percentile speed of vehicles, speed-related cause, involvement of heavy vehicle, involvement of private car, single-vehicle crash and two-vehicle crash as significant factors influencing the
severity of crashes on two-lane highways in Nigeria. Road safety agencies, engineers and transportation planners will find the model developed in this study useful for remediating crashes and improving the safety
and sustainability of two-lane highways. Keywords: Multinomial Logit Model, Crash Severity, Two-lane single carriage highway
