SBIR-STTR Award

Product development remains an uncertain process especially for highly engineered and complex DoD programs. Development uncertainties and evolving external factors encumber the balancing of conflicting design criteria and constraints using the existing trade-space exploration (TSE) and optimization and affordability analysis capabilities for solution concept generation and assessment during the material solution analysis and technology maturation and risk reduction phases. This proposal develops a framework for an innovative and comprehensive multi-stage set-based design approach driven trade-space optimization and exploration capability to support DoD’s PD processes and achieve PD life-cycle resilience. Proposed framework innovatively integrates the state-of-the-art of sequential decision making under uncertainty with combinatorial trade space optimization and exploration based on set-based design approaches. To cope with computational complexity, framework will design and detail intelligent and scalable value-iteration algorithms that guarantee optimality of set-based solutions as well as state-of-the art multi-criteria optimization algorithms. By integrating variable fidelity physics- and simulation-based analysis with efficient set-based design driven TSE and sequential decision optimization, proposed framework increases design resiliency and confidence. Additional functionality includes extensive and practical sensitivity analysis module, real-time PD decision support capability, visualization and characterizations of PD decisions in design criteria space and in terms of system designs.

Product development (PD) remains an uncertain process, especially for highly engineered and complex DoD programs. Historically, PD systems relied on a process that quickly zeroed-in on a single promising “point-solution” concept and most of the effort went into realizing that solution. This is the predominant approach to PD around the globe and is the case with the DoD 5000 acquisition process as well as the recent DoDI 5000.02 Adaptive Acquisition Framework. In recent years, there is growing recognition that point-solution design (PSD) approach limits PD program management and development resilience with respect to changes in requirements and to uncertainties in the availability and maturation of subsystem/component alternatives during system development and integration. Therefore, an effective “set-based design” (SBD) approach is necessary for best addressing program resiliency and trading risks and uncertainties with cost efficiency and program success. While SBD approaches have been entertained in practice, the studies are mostly qualitative in nature. There exist no effective, scalable, and commercial SBD decision support platforms to guide the program management (PM) and development process with multi-stage and multi-criteria considerations, the development of which is the core focus of this Phase II proposal. This proposal seeks to develop an innovative and comprehensive multi-stage SBD tool platform for program management and development to support DoD’s acquisition processes. The resulting Resilient Program Management & Development (RPMD) Tool can fundamentally transform the highly complex and sophisticated DoD acquisition programs to promote on-time and on-budget development of product systems while accounting for the entire product life-cycle. RPMD tool innovatively integrates the state-of-the-art of sequential decision making under uncertainty, as Markov Decision Processes (MDP), with combinatorial trade space exploration and optimization (TSE&O) based on set-based design approaches. To cope with computational complexity in handling large-scale real-world DoD programs, the RPMD tool implements the intelligent and scalable MDP solution algorithms developed in Phase I. The developed novel solution set-based TSE&O algorithms employ state-of-the art multi-criteria optimization algorithms to guarantee optimality of SBD solutions. For highly complex programs, the tool platform also employs meta-heuristic optimization algorithms for scalability. By integrating variable fidelity physics- and simulation-based analysis with efficient SBD driven TSE&O and sequential decision optimization, proposed tool platform increases PM/PD resiliency and confidence. Additional functionality includes extensive and practical sensitivity analysis module, real-time PM/PD program and technical risk-management decision support capability, visualization and characterizations of program and technical decisions in design criteria space and in terms of system designs.