For safety design of industrial vehicles, it is important to understand the optimum posture parameters such as the elbow-ground distance, the popliteal height and the reach angle. RULA and REBA, which are most commonly adopted to evaluate the risk associated with posture exhibit higher dependence on the mentioned parameters. The relative significance of these parameters is not known for drivers in industrial vehicle. The main objective of this study is to develop a model using an automated neural network search (ANS) approach for the prediction of RULA and REBA based on the coupled interactions of posture parameters. In the context of model development, field study that contains measurement of these posture parameters was utilized. RULA and REBA were assessed from these posture parameters using CATIA software. Further, the study also reveals the relative significance of these posture parameters and identifies the most optimum parameters for minimum risk to driver's health.