Though technology has shown a rapid development, manual material handling (MMH) tasks are still usual activities in most industries. According to recent surveys, MMH tasks remain one of the main reasons for the emergence of occupational low back disorders (LBDs). It is critical to be able to discriminate as accurately as possible between MMH jobs that place workers at high versus low risk of LBDs. In this study, the risk of occupational LBDs has been classified by support vector machines (SVMs) considering both trunk motion variables and workplace variables, which have been extensively used to identify risk of LBDs. The LBDs‐SVM model has outperformed the existing models in terms of accuracy, which is equal to 88.5% for correct classification of high risk when 10‐cross validation is applied. In other words, the proposed model has correctly classified an average of 29.2 cases out of 33 high‐risk cases, which is critical to be determined compared to low‐risk cases. The results obtained in this study indicate that SVM is a better classifier than the other existing methods in the literature to classify LBDs risks.