Process design and operation is concerned with the optimal selection and efficient utilization of resources along time. The operational efficiency of equipment units depends strongly on the maintenance policy employed. This work addresses critical operational issues in the sugar cane industry such as the problem of determining the optimal cyclic cleaning policy in the evaporation section and the corresponding optimum steam consumption profile of both evaporation and crystallization sections. A main feature of this problem is the performance decay with time of each evaporation unit which must be restored by appropriate cleaning operations. in this paper, a detailed mixed integer nonlinear programming (MINLP) performance model which includes the effect of fouling on the overall heat-transfer coefficient is considered. The problem formulation can also handle multiple-unit parallel evaporation lines. Problem solution provides for each production line the optimal cleaning schedule, mass flow to be processed, and vapor bleeds. (c) 2006 Elsevier B.V. All rights reserved.