Managing unarticulated value : changeability in multi-attribute tradespace exploration

Thesis (Ph. D.)--Massachusetts Institute of Technology, Engineering Systems Division, 2006.Includes bibliographical references (p. 305-310).A framework for creating value robust systems in the face of changing value perceptions during the architecture and design of systems is proposed. Both unarticulated value, that which is not explicitly communicated to system designers, and dynamic value, that which changes over time, are used to motivate the dynamic Multi-Attribute Tradespace Exploration (MATE) process. Value can be represented as decision maker perceived attributes, which can be classified according to the ease by which the system can display them. The attribute class spectrum from least to most costly ranges from articulated, class 0 attributes, to inaccessible value, class 4 attributes. Supporting the value-adding approach, the system property concepts of flexibility, adaptability, rigidity, robustness, scalability, and modifiability are proposed to be different aspects of the same concept: changeability. A quantification of changeability is shown to be the Filtered Outdegree of a design within a networked tradespace formed through explicit consideration of transition paths between design instantiations. A focus on designing not only for value, but for changeability as well, leads to the concept of path enabling variables, whose purpose is to increase change paths or decrease cost for change.(cont.) Value robustness is shown to be achieved through either passive or active means. Passive value robustness can be quantified as the Pareto Trace number of a design, reflecting the number of contexts within which a particular design is determined to be best value at a given level of resource expenditure. Active value robustness is achieved through a strategy of pursuing designs with increased changeability and accessibility to likely high value regions of a tradespace. Supporting the process, the Design-Value Matrix and the Rule-Effects Matrix help system designers visualize the key factors for creating dynamic value-generating systems by capturing the important relationships between decision makers, design variables, attributes, path enablers, and resources. The dynamic MATE process is applied to two real system cases including the Joint Direct Attack Munition (JDAM) and the Terrestrial Planet Finder (TPF). The framework is shown to be applicable at both quantitative and qualitative levels, giving insight into assessing and designing for changeability and value robustness for systems.by Adam Michael Ross.Ph.D