This paper presents a study on selecting electricity contracts for a large-scale chemical production plant, which requires electricity importation, under demand uncertainty. Two common types of electricity contracts are considered, time zone (TZ) contract and loading curve (LC) contract. A multi-period linear probabilistic programming model is adopted for the contract selection and optimization. Hence, by using the probabilistic programming, a solution procedure is proposed that allow users to determine the best electricity contract according to their desired confident level of the uncertainties. In addition, due to the fact that the demand of product is uncertain, if one considers the overage and shortage of the products in the market as well, an interesting result can be obtained. The methodology is explained in the paper. (C) 2005 Elsevier Ltd. All rights reserved.