Fuzzy inference model for assessing occupational risks in construction sites

Abstract

Occupational risk assessment is a key measure to reach safety in construction industries. The assessment process is involved with many parameters which are difficult to assess, due to inadequate data or imprecise information. So, traditional quantitative approaches fail, frequently, to assess risk levels and to identify adequate preventive measures. A Takagi-Sugeno type fuzzy inference system is developed in this article to overcome these lacunas. In the model formulation process, the risk factors and controlling factors for accidental injuries are considered as input parameters. Safety levels of each type of injury prone body parts are evaluated by using analytical hierarchy process. Subtractive clustering technique is used to reduce the number of rules and thereby an initial fuzzy inference system is generated. Finally, the initial model is updated by tuning all the parameters corresponding to the input variables using a hybrid learning process. The developed methodology has been applied to few selected construction sites in India. The derived results validate the applicability of the developed model for assessing risks in construction sites and also identifies the pertinent progress of existing safety strategies.