A Bayesian network model for analysis of causes and consequences of accidents

Accidents are chains of events that lead to identifiable injuries and illnesses. Among the various mechanisms of trauma, traffic accidents have the highest mortality rate. This study examines historical data for analysis of causes and consequences of accidents in King of Saudi Arabia in period 2016-2020. To collect the necessary data, the researchers utilized the Saudi open data portal, a National e-Government Portal. The data on consequence, type, seasons, location, and gender were extracted from the database. In order to analyze the collected data, GeNIe academic software was employed to conduct Bayesian network analysis. In total, 106513 accidents occurred in the Kingdom of Saudi Arabia in period 2016-2020. Among them, 40287 cases were with the consequences of damage and injuries and 9492 with the consequences of deaths. In sensitivity analysis of accident types, the worst variations in accident consequences were related to accident types of collapsed, drowning, and car accident. So that those could increase probability of death consequence by 5, 4, and 3 percent, respectively. In sensitivity analysis of location, for west and east regions, the probability of deaths consequences decreased by 1. In sensitivity analysis of season, the probability of deaths decreased by 1 percent in autumn season. In sensitivity analysis of gender, female could decrease the probabilities of deaths by 4 percent. Other factors could not make variation in the probability of deaths. These findings show most important accident types in association with death consequences are collapsed, drowning, and car accident. Location of west and east of king of Saudi Arabia, seasons of autumn, and gender of female were also associated with decrease of death consequences.