Skip to main content
Infrastructure Resilience Conference 2018

Full Program »

Exploring Resilience Within an Ontology for Systems Qualities

This talk draws on research that we have been doing within the US Department of Defense (DoD) Systems Engineering Research Center. In 2012, one of the DoD’s seven Key Technology Areas was called Engineered Resilient Systems, and our SERC sponsor, the DoD Deputy Undersecretary for Systems Engineering, arranged for two workshops to review ongoing ERS research and to explore what the SERC could do to complement the existing ERS research. The workshops concluded that the main ongoing ERS research was doing an excellent job in research to improve the resilience of physical vehicles via field testing, supercomputer modelling, and improved vehicle design and experimentation. Our sponsor concluded that it would be good for the SERC to complement this research by addressing the tradespace between resilience and other system qualities for cyber-physical-human systems.

Our initial exploration found that types of resilience and other system qualities (SQs) varied across DoD stakeholder classes. Vehicle operators’ tradespace primarily involved Robustness, Maneuverability, Speed, Range, Capacity, Usability, Reliability, Availability, and Maintainability. Cyber-Physical-Human system commanders’ tradespace primarily involved Interoperability, Understandability, Timeliness, Accessibility, Simplicity, Completeness, Agility, Accuracy, Relevance, Robustness, Operational Trust, and Security, along with Maintainability attributes of Modifiability, Repairability, User-Adaptability, Economy, Speed, Survivability, and Understandability. In looking up Resilience on Wikipedia, we found over 20 different definitions, with 10 different desired end states. In evaluating the leading SQ standard, ISO/IEC 25010, System and Software Quality Requirements and Evaluation (SQuaRE), we found it to be primarily focused on computing and software systems, and particularly weak in comprehensiveness, consistency, and completeness in identifying the relationships among the SQs.

To improve the situation, we proposed and formed an 8-university team led by USC with myself as PI, and including AFIT, Georgia Tech, MIT, NPS, Penn State, U. Virginia, and Wayne State, to execute a 5-year plan (now in Year 4) to strengthen the foundations of SQ Tradespace and Affordability Analysis; to develop, exercise, evaluate, evolve, and transition SQ analysis methods, processes, and tools; and to research and develop next-generation systems engineering cost estimation and systems and software Total Cost of Ownership estimation. Drawing on earlier SQ ontology work at USC, MIT, and AFIT, we developed an IDEF-5 based SQ ontology framework for cyber-physical-human systems. For Resilience, we converged on the definition in the INCOSE Systems Engineering Handbook: Resilience is the ability to prepare and plan for, absorb or mitigate, recover from, or more successfully adapt to potential or actual adverse events. With respect to the SQ ontology, this identifies five classes of Resilience: Robustness, Self-Adaptability, User-Adaptability, Modifiability, and Repairability. We agree with most of the current research on Infrastructure Resilience and Safety that USER-Adaptability is the most important of these for cyber-physical-human systems, but even more effective when combined with the others.

The keynote presentation will include a short summary of the background above; a summary of the SERC System Quality Ontology, Tradespace, and Affordability (SQOTA) project as above and its relation to infrastructure resilience; and short summaries of two of our infrastructure-resilience-related research activities: infrastructure for coordinating swarms of autonomous drones, and massive-data analytics tools for continuously assessing software infrastructure maintainability and technical debt.

Prof Barry Boehm
USC and US DoD Systems Engineering Research Center

 

Powered by OpenConf®
Copyright ©2002-2016 Zakon Group LLC