TY - JOUR T1 - P153 Predicting and determining factors of occupational accidents severity rate (ASR) using artificial neural networks (ANN); a case study in construction industry JF - Occupational and Environmental Medicine JO - Occup Environ Med SP - A171 LP - A172 DO - 10.1136/oemed-2016-103951.470 VL - 73 IS - Suppl 1 AU - Ahmad Soltanzadeh AU - Iraj Mohammadfam AU - Abbas Moghimbeigi Y1 - 2016/09/01 UR - http://oem.bmj.com/content/73/Suppl_1/A171.3.abstract N2 - The severity of accidents is an important index for occupational accident analysis and modelling. Accident severity rate (ASR) as in the construction industry may be due to various factors. This study aimed to determine the factors of accident severity rate (ASR) in the construction industry and introduces a model to predict ASR for construction accidents. This study was carried out in 13 large construction sites and analysed and modelled ASR of construction accidents that occurred from 2009 to 2013. Pearson χ2 coefficient and artificial neural networks (ANN) were the models of choice for the study. Findings of both models showed that some individual factors (IFs), organisational factors (OFs), HSE training factors (HTFs) and risk management system factors (RMSFs) could be predictive and related factors of ASR in the construction industry. The results indicated that Pearson coefficient and ANN are reliable tools which could be used for occupational accidents ASR factors’ modelling in many industries. ER -