Component level reuse enables retention of value from products recovered at the end of their first lifecycle. Reuse strategies determined at the beginning of the lifecycle are aimed at maximizing this recovered value. Decision based design can be employed, but there are several difficulties in large scale implementation. First, computational complexities arise. Even with a product with a relatively small number of components, it becomes difficult to find the optimal component level decisions. Second, if there is more than one stakeholder involved, each interested in different attributes, the problem becomes even more difficult, due both to complexity and Arrow's Impossibility Theorem. However, while the preferences of the stakeholders may not be known precisely, and aggregating those preferences poses difficulties, what is usually known is the partial ordering of alternatives. This paper presents a method for exploiting the features of a solution algorithm to address these difficulties in implementing decision based design. Heuristic methods including non-dominated sorting genetic algorithms (NSGA) can exploit this partial ordering and reject dominated alternatives, simplifying the problem. Including attributes of interest to various stakeholders ensures that the solutions found are practicable. One of the reasons product reuse has not achieved critical acceptance is because the three entities involved, the customers, the manufacturer and the government do not have a common ground. This results in inaccurate aggregating of attributes which the proposed method avoids. We illustrate our approach with a case study of component reuse of personal computers.