Fuzzy Logic based Soil Nutrient Categorization System for Decision Making in Soil Nutrient Mitigation

Soil health determines plant health, quantity and quality of the yield. By assessing soil nutrient status, farmers can address nutrient imbalance and promote long-term soil health. The soil nutrient status of the particular soil can be categorized as Low, Medium, or High. Conventional methods for evaluating soil nutrient usually give exact numbers for nutrient levels, but these numbers might not reflect the natural uncertainty or variability in soil properties and measurements. Fuzzy logic, on the other hand, uses linguistic variables and fuzzy sets to represent imprecise information in a way that’s easier to understand. Using fuzzy if-then rules, soil nutrient values can be translated into a more intuitive soil nutrient status within a a set range. The present study aims to develop a comprehensive framework that utilizes fuzzy logic-based categorization methods to assess soil nutrient status. Linguistic variables are defined to categorize soil nutrient levels as Low, Medium, or High. Membership functions are established to quantify the degree of belongingness to each category. Fuzzy rules are formulated based on expert knowledge, encompassing 12 parameters to assess soil nutrient status. Simplifying the system into four distinct categories mitigates complexity, focusing on macronutrients, physical attributes, secondary and micronutrients. A final fuzzy inference system with 36 rules was developed to categorize soil nutrient status that facilitates informed decision making for agricultural practices using Python's skuzzy tool. This paper outlines the advancement of soil fertility assessment and underscores the applicability of fuzzy logic-based systems in agricultural decision-making processes.