Application of Fuzzy Logic in Sensory Analysis of Coconut Water Kefir and Implementation of Microbiology Education

Microbiology education often faces challenges in conveying complex fermentation processes and interpreting uncertain biological data, such as the organoleptic properties of fermented products. This study examines the application of fuzzy logic in analyzing the organoleptic data of coconut water kefir and its integration into microbiology learning through virtual simulations. Organoleptic data, including taste, flavor, carbonation, color, and overall acceptance, were collected from 20 semi-trained panelists across different storage times and sugar addition treatments. Fuzzy logic modeling involved fuzzification using trapezoidal membership functions, formulation of 243 expert-based fuzzy rules, inference via the Mamdani method, and defuzzification using the Center of Area approach. The model accurately predicted the overall acceptance level (Fuzzy Reasoning Grade, FRG) with a Pearson correlation coefficient of r = 0.92 compared to panelist scores. Kefir with sugar added before incubation achieved higher FRG values (3.8-4.2, high category) in weeks 0-2, while both treatments declined after week 3. A fuzzy logic-based virtual simulation was developed and tested on 30 high school students, resulting in a significant improvement in understanding fermentation microbiology concepts compared to conventional learning (mean score increase: 62.5 to 85.3 vs. 61.8 to 70.4; p = 0.01). Interviews indicated positive perceptions of the simulation, though further introduction to fuzzy logic and improved technological infrastructure are needed. This study demonstrates that fuzzy logic effectively manages uncertainty in sensory data and supports interdisciplinary, technology-enhanced microbiology education.