A conceptual framework for minimizing construction site accidents using task-personnel nexus matrix

Majority of the construction site accidents are due to the influence of human factors such as negligence, lack of awareness and lack of training. This work proposes a conceptual framework for minimizing construction site accidents using Task-Personnel Nexus Matrix (TPM) which selects the right worker for a particular task. Each construction site task is evaluated for its risk and the probability of human error associated with the task using Hazard Identification and Risk Assessment (HIRA) and Human Error Assessment and Reduction Technique (HEART) techniques respectively from which a combined metric of Risk cum Human Error (RHE) is proposed to be evaluated. Kohonen algorithm is proposed to classify the workers database into two or more clusters, each of which is trained using the Back Propagation Neural Network (BNN) to predict the performance of the worker knowing his basic characteristic. The high performer is proposed to be assigned to higher risk tasks which correspond to a higher value of RHE. It is seen that the proposed method is highly promising in reducing the construction site accidents and is easy to implement by predicting, evaluating and assigning the right person for a right job in the construction site.