Optimizing Multi-Domain System-of-Systems Using Model-Based Systems Engineering
Abstract
System-of-systems (SoS) that exist across multiple domains can offer new solutions to complex problems. Model based systems engineering (MBSE) is a popular approach used in Systems Engineering (SE) for architecting systems or SoS, but it is often difficult or time consuming to assess many different MBSE architectures. This research looks at applying the Object-Oriented System Engineering Method (OOSEM) to multi-domain SoS problems. The Systems Modeling Language (SysML) is used to create a multi-domain SoS composed of satellite sand aircraft for surveillance missions such as search and rescue. An executable architecture is developed to assess and find a near optimal SoS. The executable architecture consists of five components; architecture, optimization, simulation, performance, and value. The architecture component contains attributes describing the SoS and bounds which are passed to the optimization component. The optimization component evaluates candidate solutions using simulations, performance metric calculations, and a value model. Once the optimizer has finished, the optimal parameters are passed back to the architecture component and updated in the SysM L model. This research found, for a permissive environment, the optimal multi-domain SoS had a 7.4 and 28.7 increase in average value over the aircraft only and satellite only systems respectively.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2016
- Accession Number
- AD1054113
Entities
People
- Cody G Palmer
Organizations
- Air Force Institute of Technology