Architecture value mapping: using fuzzy cognitive maps as a reasoning mechanism for multi-criteria conceptual design evaluation

"The conceptual design phase is the most critical phase in the systems engineering life cycle. The design concept chosen during this phase determines the structure and behavior of the system, and consequently, its ability to fulfill its intended function. A good conceptual design is the first step in the development of a successful artifact. However, decision-making during conceptual design is inherently challenging and often unreliable. The conceptual design phase is marked by an ambiguous and imprecise set of requirements, and ill-defined system boundaries. A lack of usable data for design evaluation makes the problem worse. In order to assess a system accurately, it is necessary to capture the relationships between its physical attributes and the stakeholders' value objectives. This research presents a novel conceptual architecture evaluation approach that utilizes attribute-value networks, designated as 'Architecture Value Maps', to replicate the decision makers' cogitative processes. Ambiguity in the system's overall objectives is reduced hierarchically to reveal a network of criteria that range from the abstract value measures to the design-specific performance measures. A symbolic representation scheme, the 2-Tuple Linguistic Representation is used to integrate different types of information into a common computational format, and Fuzzy Cognitive Maps are utilized as the reasoning engine to quantitatively evaluate potential design concepts. A Linguistic Ordered Weighted Average aggregation operator is used to rank the final alternatives based on the decision makers' risk preferences. The proposed methodology provides systems architects with the capability to exploit the interrelationships between a system's design attributes and the value that stakeholders associate with these attributes, in order to design robust, flexible, and affordable systems"--Abstract, page iii.

Name thesis:degree_name

Ph. D. in Systems Engineering

Grantor

Missouri University of Science and Technology

Year dc:date.available

2016

Author dc:creator

• Singh, Atmika

• Systems Engineering

Repository record dc:identifier

https://scholarsmine.mst.edu/doctoral_dissertations/1955

OAI identifier oai:identifier

oai:scholarsmine.mst.edu:doctoral_dissertations-2957