ABSTRACT
Road Transportation is the most accessible and widely used means of transportation in Nigeria, this is due to its relatively low cost and availability. Despite its far and penetrating reach to rural and urban centres and its ability to bring about interconnectivity between people of different races and ethnicity, road transportation comes with its own demerits, infamous of which is road traffic crashes. These crashes have maimed many individuals, causing permanent disabilities and leaving inglorious scars in the hearts of others, lives of both the old and young have been brought to an abrupt end, aborting many dreams and shattering many homes. All this has not been without the associated physical, emotional, and psychological trauma. In Nigeria, two lives are lost every four hours in crashes involving over 20,000 vehicles yearly, this scourge if not checked results in over 2000 deaths yearly. This research analysed road traffic crashes in Edo State, Nigeria and developed crash predictive models.
Data on the total number of crashes that occurred between year 2004 to year 2018 in Edo State, the number of people injured and the resultant number of people killed in these crashes were obtained from the Sector Command of the Federal Road Safety Corps (FRSC) situated in Ikpoba Hill, Benin City, Edo State. Crash predictive models were developed using ARIMA (Autoregressive Integrated Moving Average) and ARIMAX (Autoregressive Integrated Moving Average with Explanatory Variables) modelling techniques to forecast the likelihood of future crash incidents in a bid to make road safety planning more proactive than reactive.
The result showed that the ARIMAX models developed outperformed the corresponding ARIMA models when compared using Bayesian Information Criterion, Mean Square Error, Mean Absolute Percentage Error, and Coefficient of Determination as the decision criterion. For the period under review, 5,211 crashes were recorded, 14,252 injuries and 3,216 deaths. The research also revealed that human factor was responsible for 81% of the total crashes experienced in the state, with crash severity index ranging from 0.41 to 1.09 in 2010 and 2013 respectively. Pareto Analysis revealed that 5 causative factors; SPV, LOC, OTHERS, DGD and BFL led to 80% of the crashes recorded during this period. The recommendations in this research if accepted and applied will be useful to the Federal Road Safety Corps, Edo State Government and the Nigerian Government at large in bringing road crashes to a barest minimum.
