The world loses a human live in every 24 second due to Road Traffic Accidents (RTAs). In Kenya approximately 3000 lives are lost annually due to RTAs. The interventions to improve road traffic safety (RTS) failed because they were not informed by any scientific research. In this paper we employed the multi-layer feed forward perceptron neural network model to classify the road traffic safety status (RTSS) as:-excellent, fair, poor or danger states which model’s output are. We considered the vehicle internal factors that contribute to RTAs as model’s inputs which included:-inside-vehicle-condition, entertainment, safety-awareness, passager’s (attention, criminal-history, health-history, movement inside vehicle, body posture, frequency of journey, drunkenness’, drug-influence, use-of-mobile-phone and load), luggagetype and the safetybelt. The model was trained, tested and validated with classical data collected from a sample of 1000 respondents from road traffic safety authority (RTSA) experts in Kenya.
The world loses a human live in every 24 second due to Road Traffic Accidents (RTAs). In Kenya approximately 3000 lives are lost annually due to RTAs. The interventions to improve road traffic safety (RTS) failed because they were not informed by any scientific research. In this paper we employed the multi-layer feed forward perceptron neural network model to classify the road traffic safety status (RTSS) as:-excellent, fair, poor or danger states which model’s output are. We considered the vehicle internal factors that contribute to RTAs as model’s inputs which included:-inside-vehicle-condition, entertainment, safety-awareness, passager’s (attention, criminal-history, health-history, movement inside vehicle, body posture, frequency of journey, drunkenness’, drug-influence, use-of-mobile-phone and load), luggage-type and the safetybelt.