Algorithm for selecting adjustment checks based on the Savage criterion

This paper examines the problem of selecting the best strategy when quality indicators deviate from acceptable values during combine harvesting. An approach to solving the problem is developed based on criteria used in game theory. The mathematical apparatus of game theory is applied to decision-making under uncertainty, which is typical for combine harvesting of grain crops. The prospects for applying various criteria used in "games with nature" are analyzed. The application of the Savage criterion for selecting a rational sequence of checks to eliminate the detected violation is substantiated. A fuzzy knowledge base is created, reflecting a system of complex relationships between the external signs of process violations, their causes, and the controlled and technical parameters of the combine. The concept of an efficiency matrix is introduced as an analogue of a payoff matrix in game theory, which reflects the effectiveness of applying various strategies to solve the problem. The paper presents a matrix of the efficiency of applying various strategies for one of the common violations for classical combine harvesters. The developed approach is intended for the implementation of a logic block for developing a strategy for searching for the cause of the identified deviation for an intelligent information system for technological adjustment of a grain combine.