Prediction of Road Traffic Accidents Outcomes on Selected Routes in Ibadan Environs Using Time-Series Models

Abstract

Road traffic accidents has emerged as a major public concern due to the associated burdens of injuries, loss of lives and properties. The objective of this paper is to use time-series technique applying autoregressive integrated moving average and seasonal autoregressive integrated moving over models to predict accidents outcomes within and around Ibadan metropolis for appropriate mitigation measures. Data on daily recorded cases of road traffic accidents on several specified routes between 2019 and 2021 were collected from the Federal Road Safety Corps, Ibadan Zonal Office. The RTAs outcomes were grouped into three: injuries, fatalities and vehicle occupants. The training set consisted of data from December 2019 to June 2021, whereas the testing set was composed of data from July 2021 to December 2021. During the study, 427 fatalities and 2,245 injuries were caused by traffic accidents involving a total of 5,577 vehicle occupants. By the end of 2025, the study projected a 172% rise in the death rate and a 38% increase in the number of vehicle occupants involved in RTAs in the study area, and approximately 32% of those occupants suffering injuries. There was no discernible seasonal pattern in the trends of injury and fatality rates. There is an immediate need for a road traffic accident prevention and evaluation program in the study region given the projected number of injuries and fatalities