ACCIDENT PREDICTION MODEL FOR HIGHWAYS WITH REST AREA BY USING POISSON AND BINOMIAL REGRESSION MODEL

Abstract

Currently, rest area positively impacts on road users in terms of business points and information providers; however, the important task of rest area is reducing the number of road accident cause of fatigue. In the past, truck drivers in Thailand usually parked their vehicles on the roadside or shoulder, which these harmful areas could generate road accidents when parked vehicles move out from the road shoulder to the travelled way. In Thailand, road accident situation is top 10 of the world. Research objectives consist of the determination of influencing factors that affect the road accident within upstream and downstream 16 km of rest areas in Thailand and built the accident prediction models. It was found out that main variables such as serviceability of rest areas and the number of the truck users in rest areas are significantly impacted into the accident prediction model. In summary, rest areas could help to reduce the number of road accidents.