This work addresses the problem of optimizing the design of a family of products and simultaneously maximizing the commonality between these products. By commonality, we mean the proportion of components that are shared between the products. The set of all components common to all products is called platform [2, 3]. We consider here the multiplatform problem that allows the commonality to start as soon as one component is common to two products. The multiplatform formulation encompasses every possible case of commonality and leads to a highly complex optimization problem: for NP products, of each N design variables, there are (BNP )N multiplatform configurations (BNP is the Nth P number of Bell). In this paper, we propose an algorithm to tackle this multiplatform product family optimization. Similarly to [1, 3] , we will estimate the Pareto front between an aggregation of the products engineering performance and the family commonality. The proposed method is based on a quadratic approximation of the product performance functions and has a complexity of O(N ×N2 P ×log(NP )), which is much better than the complexity of any enumeration strategy ((BNP )N).