The way products are placed in a supermarket can be effective in increasing sales and profit. A reasonable approach is to group together items that are likely to be purchased together. Thus, managers with the support of mining methods and techniques, can assist customers locating the products they want to buy in an easy and quick way. Many product placement strategies have been proposed over the years to leverage an effective and efficient way to achieve this goal. In this paper, association rules for product arrangement in supermarkets are studied and an algorithm based on such rules is proposed. The algorithm considers several factors, such as the number of units sold of each product, a hierarchical structure for product classification developed by the United Nations Department of Economic and Social Affairs, a set of association rules generated from sales, and a set of constraints that restrict some products to be placed physically close to each other in a supermarket, even if they are usually purchase together. Real public sales data of a supermarket were used for the experiments, where the proposed algorithm is applied for the generation of supermarket layouts. The results show that some supermarket departments may share the same products or product categories.