Given the high degree of mechanization to which agricultural activities are being submitted, the objective of this research was to develop a fuzzy model able to evaluate and classify the insalubrity level in different work environments. The model developed has as input variables: wet bulb globe temperature index (WBGTI, °C), noise level (dBA), metabolic rate (W m-2) and rest time (%); and has as output variable the human well-being index (HWBI). The Mamdani inference method was used, and for the defuzzification, the Center of Gravity method was used. The system of rules was developed based on combinations of input variables. Four hundred rules were defined, with weights equal to one, considering that in the development of rules, a specialist was consulted. Field data were used in order to test the developed system. The results showed that the proposed model is a promising tool in determining the HWBI, presenting the ideal rest time ranging from 64.2% (chainsaw, near the operator's ear) to 25% (harvester, 20 meters away from the operator), and, given a pre-defined scenario of thermal and acoustic environment, it was possible to determine the degree of human well-being and the ideal rest time for each equipment evaluated.
human well-being; mathematical modeling; expert systems
