Prediction of occupational risk in the shipbuilding industry using multivariable linear regression and genetic algorithm analysis

In this research, an effective approach based on Multivariable Linear Regression (MVLR) and Genetic Algorithm (GA) methods has been applied to study the effect of working conditions on occupational injury, using data of occupational accidents accumulated by ship repair yards. The work aims at the development of a calculating model that will use soft computing techniques to assess the occupational risk in the working place of shipyards using occupational accidents data. For each accident the following parameters have been considered as the model's input features: day and time, individual's specialty, type of incident, dangerous situation and dangerous actions involved. Reported accident data were used as the training data for the MVLR model to map the relationship between the working conditions and occupational risk. With the fitness function based on this model, genetic algorithms were used for the prediction of occupational risk taking into consideration the severity and the frequency of occupational accidents data accumulated by ship repair yards. The working parameters' values for minimum occupational risk were obtained using GAs. By comparing the predicted values with the reported data, it was demonstrated that the proposed model is a useful and efficient method for predicting the risk of occupational injury.