%0 Journal Article
%T A Neural Network Based Diagnostic System for Classification of Industrial Carrying Jobs With Respect of Low and High Musculoskeletal Injury Risk
%A Rohit Sharma
%A Ranjit Singh
%J International Journal of Biometric and Bioinformatics
%D 2012
%I Computer Science Journals
%X Even with many years of research efforts, Safety professionals and ergonomists have not yetbeen established the occupational exposure limits of different risk factors for development ofMusculoskeletal disorders (MSDs). One of the main problems in setting such guidelines is toaccurately assess the association between exposures and possible occupational disorders ordiseases and predict the outcome of any variable. The task of an industrial ergonomist iscomplicated because the potential risk factors that may contribute to the onset of the MSDsinteract in a complex way, and require an analyst to apply elaborate data measurement andcollection techniques for a realistic job analysis. This makes it difficult to discriminate wellbetween the jobs that place workers at high or low risk of MSDs. This paper describes a newapproach for the development of artificial neural networks applied to classifying the risk of MSDsfor industrial carrying jobs. The data set used in this research was collected from Foundry andSugar industries workers using the physiological variables.The main objective of this study was toto develop an artificial neural network based diagnostic system which can classify industrial jobsaccording to the potential risk for physiological stressors due to workplace design. The neuralnetwork obtained can be used by the ergonomist as a diagnostic system, enabling jobs to beclassified into two categories (low-risk and high-risk) according to the associated likelihood ofcausing MSDs. This system provides a higher proportion of correct classifications than otherprevious models. So, the system can be used as an expert system which, when properly trained,will classify carrying load by male and female industry workers into two categories of low risk andhigh risk work, based on the available characteristics factors.
%K Musculoskeletal Injuries
%K Physiological Risk
%K Artificial Neural Network
%U http://cscjournals.org/csc/manuscript/Journals/IJBB/volume6/Issue1/IJBB-146.pdf