The quality of raw milk needs to be accurately regulated, as the quality of raw milk is not only important for consumers of milk but also for the quality of associated dairy products. Every milk delivery is inspected to determine certain qualitative properties. Microbiological and physicochemical analyses are important tools for monitoring the quality of raw milk and its products. Microbiological contamination has a major effect on the quality of raw milk. Consumer requirements for high quality milk and dairy products have placed a significant responsibility on dairy producers so dairy factories need a good system to evaluate and classify the quality of the raw milk they receive. This paper demonstrates the application of fuzzy logic to classify raw milk based on microbiological and physicochemical qualities. Thus, a fuzzy logic system was developed using a fuzzy inference engine. Input parameters of the system were as follows; total aerobic microorganism count, somatic cell count, coliform count, fat and solid non-fat percentages. This input data was used to classify raw milk according to the following scale; Excellent, Very Good, good, Bad and Very Bad. Triangular and trapezoidal membership functions were applied for input and output fuzzy linguistic variables of the fuzzy inference system. In total, this study employed 675 rules with the logical AND operator, mamdani implication, and centroid method for defuzzification. Results showed that fuzzy logic provided the means for more stringent classification of the quality of raw milk than the presently employed system. Results obtained from the developed fuzzy inference system scheme were 82.5% in agreement with the results of other researchers.

Keywords