PREDICTION OF THE NUMBER OF ROAD TRAFFIC ACCIDENTS OCCURRENCE ON THE LOKOJA-ABUJA-KADUNA EXPRESSWAY USING GREY-ARTIFICIAL NEURAL NETWORK MODEL

Globally, Road Traffic Accidents (RTAs) rank among the leading causes of death, a situation exacerbated in developing countries where road networks are the primary transport means. Consequently, developing reliable models to predict RTAs has become an urgent international priority for reducing loss of life and property. The conventional grey GM(1,1) prediction model, despite its proven usefulness across various domains, is prone to overestimation and underestimation, limiting its dependability. This paper uses an integrated Grey-Artificial Neural Network Model (GANNM) to enhance predictive accuracy for RTAs on Nigeria's Lokoja-Abuja-Kaduna expressway from 85.97% to 96.02%. Additionally, the results demonstrate that the GANNM resolves the estimation inaccuracies of the grey GM(1,1) approach, yielding critical data to support Nigerian policymakers in crafting effective road management strategies to enhance safety on this highway.