The purpose of this paper is to model the optimal pricing problem in the following business practices : one firm produces various brands which belong to one product category. We pose an optimal pricing model for brands be2 longing to one product category , characterizing the complex interactions among price , sales volume , and cost per u2 nit , and the resulting sophisticated influence on a firm’s profit . Then , we apply a newly introduced neural network modeling technique to establish a neural network model needed by the solution of the optimal pricing model . In ad2 dition , we evaluate the performance of the neural network algorithm , and give an experimental example to show the wonderful performance of it . Furthermore , we illustrate an example of pricing for three brands , build its optimal pricing model , and derive its optimal solution using the neural network algorithm.