|Ph.D Student||Yaniv Mordecai|
|Subject||Cyber-Physical Disruption Modeling, Analysis, and|
Management: An Evolutionary Object-Process
Model-Based Robust Systems Engineering
|Department||Department of Industrial Engineering and Management||Supervisor||Full Professor Dori Dov|
|Full Thesis text|
System disruption is any deviation from the nominal behavior, function, or structure applied to, in, or by a system. Disruptions have both advantageous and adverse impacts. System evolution, interoperability, and autonomy, for instance, are generally advantageous, while risk, uncertainty and complexity are generally adverse. Systems usually act in disrupted environment, and must be sufficiently ready for likely disruptions.
Conceptual models of phenomena and systems play a central role in science and engineering: they describe and analyze complex natural and artificial systems. Conceptual modeling is significant when variability and complexity are high. The primary value of models stems from their ability to express and clarify non-trivial information about the systems they relates to. However, conventional conceptual modeling frameworks fail to accommodate disruptive, unordinary, exceptional, anomalous, variant, or stochastic aspects within their nominal structure, behavior, and function specification. This deficiency is often due to the inability of the conceptual modeling framework to integrate disruptive factors into the nominal system model. Consequently, disruptions are sometimes described with dedicated conceptual models with ad-hoc syntax, semantics, and ontology. The result is not only loss of information, insight, and potential for understanding of the system in its entirety, but also inconsistency, contradiction, and troubling model maintenance and coordination issues. As a result of this discrepancy, nominal conceptual models fail to provide sufficient informativity to system and model stakeholders, while disruption-centric models, such as risk or failure models, are detached from the core system model. As the system evolves, extraneous models of this nature require constant maintenance and adjustments to the core nominal model and are therefore often neglected or abandoned.
This doctoral thesis proposes a Model-Based Robust Systems Engineering framework, MBROSE, which caters and applies to both the nominal and disruptive factors of complex systems, their models, and the system engineering process. MBROSE is founded on Object Process Methodology, OPM - a holistic conceptual modeling paradigm for multidisciplinary, complex, and dynamic systems and processes. OPM is ISO 19450 standard, and a state-of-the-art conceptual modeling and model-based systems engineering methodology.
MBROSE consists of two primary modules. The first module is a mechanism for model informativity analysis and quantification, which defines how information is generated by and gained from conceptual models. The second module is a catalogue of modeling and design patterns that cater to a host of disruptions, disruptive factors, and disruptive impacts. The introduction of a wide variety of disruption kinds into conventional models using the proposed patterns is shown to elevate the Informativity of conceptual models of systems in a variety of domains, including nuclear reactors, aerospace and defense, and information technology. The value that readers and users can gain from MBROSE is therefore double: (1) MBROSE enables the construction of rich, disruption-informed conceptual models of complex systems, and (2) MBROSE provides the analytical framework to evaluate the contribution of disruption-aware modeling to the Informativity of the model. These benefits of MBROSE potentially have a positive impact on the whole model-based systems engineering process, and eventually, on the quality, robustness, and stability of the system.