Classification and prediction of occupational injury risk using soft computing techniques: An Italian study

This study was aimed both at assessing the risk of occupational injury, considering the probability and consequences of injuries that may occur, and at identifying any general factors which may affect the incidence of injuries. The analysis in this paper refers to occupational injuries which occurred in a 5-year period (2002–2006) in an Italian region. Any activities to prevent occupational risks and injuries need to involve an adequate information technology system.Classification and prediction are necessarily involved in this type of data analysis in order to obtain models which describe important categories of data, or to predict the trend of future data. Neuro-fuzzy networks are considered a powerful tool, especially when dealing with these problems of data mining. The aim of this paper was to show through a real application the flexibility and advantages of using the neuro-fuzzy network, a typical soft computing tool, for an occupational injury study. Using these innovative techniques to analyse injury data coming from an Italian region, this study aims to:•obtain a classification of input data according to their importance and/or influence on the trend in injuries;•assess how a variation in one or more pieces of input data can effect occupational injury and subsequently carry out a sensitivity analysis concerning the frequency of the injuries and their consequences;•verify the possibility of comparing past data with the predicted trend in injuries, using the neuro-fuzzy networks trained with the past data.