Dynamic Optimization Of Systems Of Systems Using Value Measurement

Abstract

Many systems of systems (SoS) occur today through the collaboration of multiple stakeholders, without any clear authority or direction at the SoS level. Examples include a supply chain, an airport, and the Internet. Because of the uncontrolled interactions, such a SoS exhibits complexity behaviors such as reflexivity, emergence, self-organization, and adaptation. This paper presents underlying theory to explain a methodology for dynamic optimization of SoS, in which the stakeholders make changes based on local perceptions of the value of the SoS. Each stakeholder has a different local perception of value, measured against an objective function that may be unique to the stakeholder. As the stakeholder compares perceived value to desired value, the stakeholder attempts to optimize the SoS from its own point of view by making decisions that lead to SoS changes. These changes then affect other stakeholders, resulting in a dynamic architecture and implementation of the SoS. We present the optimizing value proposition of each stakeholder in terms of a quantifiable objective function based on the life-cycle value of the SoS to that stakeholder, and we present the dynamic interaction as a model of self-organization. An example based on supply chain management demonstrates the approach.