This paper addresses an object-oriented methodology in designing decision support systems (DSS). An analysis of computational needs at various levels of an organization defines the data objects of the DSS. Using a two-tier approach, we first devise high-level object constructs in supporting decision analysis. With the message-passing mechanism, a higher-level object (solver) can be composed with certain problem-solving capability. Such composability provides further problem-solving powers. A hierarchical inventory management system and a company-wide order processing system are implemented using our framework.