The phase of knowledge acquisition and formalization is being considered as the key one for the development of decision support system. The main problem at this stage is to find a knowledge representation (KR) and a supporting reasoning system that can make the inferences your application needs. The main criterion of choice is what kind of inference the developers prefer and is more appropriate to the problem domain. However, already at the stage of the user interface creation there arise tension between KR and end-user "needs" (preferences and habits). So, multiple representations of the acquired and structured knowledge and management techniques could provide a solution for decision support system inference and interface requirements satisfaction.