Academic literature on the topic 'The hierarchical model'

As software is becoming larger and more complex, it is increasingly important to use the hierarchical modeling approach. Unfortunately, however, UML does not specify each metamodel with hierarchy for model by modeling phase. Thus, most UML-based methodologies do not address the hierarchical modeling for model. As a method for supporting hierarchical modeling on UML, this paper proposes a layered metamodel which defines hierarchically modeling elements of model according to the modeling phase. We describe each metamodel with hierarchy for models in UML, then present the hierarchical integrated metamodel combined with each metamodel by three modeling phases (conceptual phase, specific phase, and concrete phase). Therefore, designers are able to construct the hierarchical model by applying the metamodel with hierarchy. Using the hierarchical metamodel enables designers to improve the usability of UML and reusability of application model.

Business, economic, and agricultural YES-or-NO decision making problems often require multiple, different and specific expertises. This is due to the nature of such problems in which decisions may be influenced by multiple different, relevant aspects, and accordingly multiple corresponding expertises are required. Fuzzy expert systems (FESs) are widely used to model expertises due to its capability to model real world values, which are not always exact, but frequently vague or uncertain. In this research, different expertises, relevant to the decision solution, are modeled using several corresponding FESs. Every FES produces a crisp numerical output expressing the degree of bias toward “Yes” or “No“ decision. A unified scale is standardized for numerical outputs of all FESs. This scale ranges from 0 to 10, where the value 0 represents a complete bias ”No“ decision and the value 10 represents a complete bias to ”Yes“ decision. Intermediate values reflect the degree of bias either to ”Yes“ or ”No“ decision. These systems are then integrated to comprehensibly judge the binary decision problem, which requires all such expertises. Practically, the main reasons for independency among the multiple FESs can be related to maintainability, decision responsibility, analyzability, knowledge cohesion and modularity, context flexibility, sensitivity of aggregate knowledge, decision consistency, etc. The proposed mechanism for realizing integration is a hierarchical fuzzy system (HFS) based model, which allows the utilization of the existing If-then knowledge about how to combine/aggregate the outputs of FESs.