Many research articles appeared in the literature on the shelf space allocation problem. The problem deals with how to optimally allocate shelf space to each brand of items so as to maximize the total sales volume. They reported that a large pile of goods displayed on shelves will lead the customers to buy more and so sales at the retail level tend to be proportional to the quantities on display. Additionally, some research papers found that in the case of multi-level shelf, the level of shelf also has substantial effects on the sales volume. Motivated by the above research findings, this paper addresses a problem of retailer who sells various brands of items through displaying on multi-level shelves. The decision variables are the level of shelves and shelf space allocated to each brand of items. We formulate the shelf space allocation problem in a mathematical model with the objective of maximizing the retailer``s profit. It is assumed that arriving orders are stored in the baclcroom and continuously restocked into the shelves. Then a solution procedure is developed based on the gradient search. The validity of the model is illustrated with an example problem and the solution is compared with those obtained from the total enumeration.