Research conducted so far on the process of food frying has been primarily concerned with the physical and chemical changes occurring in fats under the influence of high temperature and prolonged heating. Physicochemical changes occurring in food due to the effect of heating were also measured. From the consumer point of view nutritive, taste and aesthetic values of food are very important. From the standpoint of the producer providing consumer services the efficiency of the frying process is equally important. To meet the requirements a conception of frying process control has been developed using logical, programmable controllers based on fuzzy logic. Basic differences between bivalent and fuzzy logic were stated. A two-stage control over food frying process was suggested. The input values to the first stage of the controller are the values of thermal power necessary for heating raw product (Q(2)) and for heating frying oil (Q(3)). The input values to the second stage of the controller are frying oil temperature (T-3) in the fryer tub and travelling speed of conveyor belt v(p), Controlling heat streams in the first stage aims to obtain the oil temperature necessary to achieve desired properties and taste values in the fried food. Frying process productivity depends on the obtained oil temperature (T-3) and on travelling speed of the conveyor belt (v(p)) of the fryer. Necessary rules of the decision making process were construed to form a basis for the controller programming. It was found that there is a possibility of increasing frying process efficiency while maintaining required food quality features. (C) 2003 Elsevier Science Ltd. All rights reserved.