The increasing in road traffic crashes rate is one of the major concerns for authorities in the Middle East. Traffic collisions may result in injury, death and property damage. Road traffic accidents involving a motor vehicle with another vehicle, animal or pedestrian are becoming increasingly common in Saudi Arabia. According to Population Characteristics surveys, the total number of fatalities due to the crashes in the Kingdom of Saudi Arabia during 2017 was 10961. The number of accidents during the period 1971-1994 have increased tremendously (i.e., 30 times) while the injuries and fatalities also increased significantly.Accroding to a recent study, traffic accidents are frequent among young people aged less than 20 years in urban areas. About 4 people are getting injured hourly in the Kingdom due to the road traffic accidents. The consequences of the increasing rate of traffic crashes include significant social and economic welfare loss at the national level of the Kingdom. The severity of the traffic crashes is an important element to investigate and address the welfare loss. Therefore, the adoption of appropriate machine learning based tools to predict the traffic crash severity is expected to be an essential efforts for the Middle Eastern countries due to the scarcity of data and use of advanced prediction tools and techniques. The objective of this study is to predict the traffic crash severity using artificial intelligence AI techniques to find the most significant variables that affect the severity of the traffic crashes. The reason for using AI techniques rather than the other simple techniques such as regression analysis is the heterogeneity of traffic accident data which made it difficult for researchers to analyze and identify the hidden relationships between the accident factors. The output of this research can then be utilized by the authorities and decision makers to apply the proper actions and remedies required to enhance traffic safety.
|Effective start/end date||1/09/18 → 1/08/19|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.