Data Mining Approach to Pre-Screening of Manual Material Handlers Based on Hand Grip Strength

The general incidence of musculoskeletal disorders (MSD) is one the major causes of mobility and concerns about financial and physical losses. The more objective pre-employment selection and employee placement test methods can reduce the work related injuries and Employers in general prefer for post-offer pre-placement (POPP) testing as a tool for evaluating worker's physical proficiency towards essential functions within a job. This functions as an effective screening technique intended to identify vulnerable workers who may be at greater risk of developing future musculoskeletal injuries. In this study, data mining approaches to estimate the proper work group for a pre-defined specialized job is used. In the first phase, the cluster analysis for the grouping the young workers according to their physical parameters was applied and according to the characteristic of each cluster, particularly with increasing handgrip strength values, the clusters were coined as very poor, poor, fair, good and excellent. In the second phase, supervised learning methods are utilized to classify the worker based on the anthropometric variables.