Determination of Optimal Temperature and Humidity Values in Dairy Farms Using Fuzzy Logic Model

Abstract

Abstract Climatic conditions are important environmental factors that impact the morphological and physiological characteristics of animals in a variety of ways. In livestock, climatic conditions, in particular temperature and humidity, directly affect different yields, health and even the viability of animals. Until today, temperature-humidity index (THI) values have been calculated and used to determine the effects of temperature and humidity on cattle in the literature. However, different thresholds exist for ideal temperature, humidity, and THI levels. This study was conducted to determine the optimal temperature and humidity levels for dairy farms using the fuzzy logic model by adding expert judgment to the THI levels reported in the literature. In the study, the THI values were calculated with three different formulas from the literature, using different temperature (between -20 and +42 oC) and humidity (between 0% and 100%) values, which are probably under agricultural conditions appear. The Mamdani-type of fuzzy logic method was utilized to determine the linguistic expressions of temperature, humidity and THI values. Considering the results of the study, a significant correlation was found between the THI values obtained using the fuzzy logic method and the other three formulas (P<0.001). According to the THI thresholds, the areas below the Receiver Operating Characteristic (ROC) were found to be significant (P<0.01) in all fuzzy algorithms. Although it showed the same trends in sensitivity, specificity, and accuracy between the THI scores obtained with the fuzzy logic algorithm and the THI66 and THI72 thresholds reported in the literature, it showed 5.65% sensitivity, 99.85% specificity and 73.6% accuracy using the THI74 threshold. As a result, the results obtained with the fuzzy expert system are higher with increasing THI values and lower with decreasing THI values than in the literature. More frankly, the results obtained using fuzzy logic were found to be less risky (or safer) than the results of the studies reported in the literature. There is a need for future studies to determine the effect of THI threshold values determined by fuzzy logic method on various yield or welfare of livestock.