Continuous frying systems are very energy intensive. They also involve the interaction of many variables and complex heat and mass transfer processes. Better control of these processes can lead not only to improved product quality but also energy reduction. This paper presents a model that can simulate the dynamic behaviour of a continuous frying system in sufficient detail that enables the influence of controls on the important product attributes of moisture and oil content to be investigated. A conjugate 2-D transient model of the fryer was used to develop correlations for the oil and moisture content based on the important fryer control variables, namely supply oil temperature, potato mass throughput, fryer paddle velocity and crisp takeout velocity. The correlations were validated against data for an industrial continuous fryer system and used within the overall frying system model which was implemented in MATLAB. Proportional Integral (PI) controls implemented within the MATLAB/Simulink environment have shown that closer control of the important parameters can lead to 10% energy savings compared to the actual industrial frying system. The modelling approaches developed can also be used to investigate further improvements in the design and control of continuous frying systems. (C) 2013 Elsevier Ltd. All rights reserved.