Capability-based Modeling, Evaluation, And Portfolio Decision Analysis Approaches For System-of-Srmament-Systems Architecture

At new stage in this new century, with the aim of effectively fulfilling new historical missions and enhancing capabilities of accomplishing diversified military tasks, the development of system of armament systems(So AS) is the material base for speeding up the transformation of the generating mode of combat effectiveness, diversifying the ways of employing armed forces, and pushing forward the building of a strong national defense and powerful armed forces. As wide-ranging challenges increase in complexity and uncertainty, the Army and its decision makers face a difficult tradeoff during system of systems(So S) acquisition process. On the one hand, they cannot afford to develop an indiscriminately large number of armament systems subject to limited budget constraints. On the other hand, they need to develop the right types of systems that can be acquired and fielded according to a broader range of changing capability requirements and that can be readily applicable to a wide variety of future scenarios or be quickly adaptable to deal with new challenges. In addition, they largely continues to base their investment decisions on service-driven analyses and individual programs, resulting significant cost and schedule overruns as well as gaps in warfighting capabilities. Such stovepiped decision-making processes also result in the wasteful duplication in weapon systems among military services does not interoperate effectively. Therefore, relative to the traditional selection of a single optimal system out of many, the successful So S development must places more emphasis on system portfolio decision and management, which facilitate the understanding and prioritization of overall capability requirements at the So S level, and then conduct integrated trade-off analyses and selection decisions about many alternative systems as a whole. Moreover, decisions made at the early concept analysis and design stages from the architectural level directly determine the structure of component systems as well as their functionality and behavior, and have a disproportionately large influence on total life-cycle cost and scheduling.Consequently, this thesis focuses on addressing the capability-based modeling, evaluation, and portfolio decision analysis for So AS architecture, with the aim of making an exhaustive study on the contents and processes of the So AS acquisition decisions from both social and technical aspects. As a result, the scientific decisions can be advanced to promote the coordinative development within different capability areas and finally reach a flexible, adaptive, and robust system portfolio of various types for So AS architecture that can enhance the overall capabilities as much as possible. The main contents and contributions of this thesis are listed as follows.(1) A governance structure and a socio-technical process are studied for capability-based So AS architecture development. For addressing the social and technical challenges emerged during So AS development, a governance structure with committed authority, clearly aligned roles and responsibilities, and effective accountability is first established according to the characteristics of capability-based So AS architecture development and its system portfolios. This structure, which aligns the individual interests with the So AS objectives and stipulates the involved scope for each stakeholder, will promote more creativity and initiative with a sense of cooperation and competition to agree the way forward. Then, a social process and a technical process are discussed to involve the full participation of all stakeholders in the So AS architecture development and decision making to obtain the most likely mutually acceptable system portfolios for So AS architecture.(2) A novel architecture modeling and analysis approach is proposed to facilitate capability-based architecture development for the challenging So AS. In order to address the problem of inconsistent representation for the same semantic content in architectural models, a high-level data meta-model, depicting the semantic relationships of constituent architectural data elements, is first presented as the common and consistent data dictionary for guiding the architectural data modeling. Then, the mapping rules at the meta-model level are presented as the common transformation specification to facilitate the development of architectural descriptions and executable models directly from architectural data. Last, static, dynamic, and experimental architecture analyses using static and executable models are discussed to allow the early understanding and exploration of the structure, behavioral, and performance characteristics for providing more flexibility and adaptability to meet the So AS challenges.(3) A preference programming-based approach for system portfolio evaluation and selection is presented. For dealing with the different uncertainties of So AS architecture development, a preference programming model is first used to formulate the system portfolio selection with incomplete information on capability criterion weights, capability value, and cost as well as variable budget levels and in the presence of the interdependencies, interactions and synergies among systems. Then, according to the various preference orientations of different decision makers, all non-dominated system portfolios of different types are computed by a tailored algorithm. Next, robust decision recommendations of the most promising system options and portfolios are provided according to the system-level core index and portfolio-level decision rules. In addition, the further effects of eliciting additional preference information to reduce the initial uncertainties are also discussed.(4) An interactive So S portfolio decision analysis approach using the graph model for conflict resolution is presented. To address the problem of multistakeholder system portfolio decision analysis across capability areas encountered in So S architecture development with desired capabilities, the systematic modeling of system portfolio selection decisions is first studied by making best use of little available qualitative information. Then, the system portfolio selections and trade-off analyses of different key stakeholders are discussed according to individuals’ disparate preferences and different conflict behavior patterns. Finally, the extensive analysis of possible strategic interactions among all stakeholders is studied to achieve potential compromises and predict possible mutually agreeable system portfolios for So S architecture development without aggregating individual preferences and values in advance.