The EPSRC-funded SIMP project
(http://www.dcs.qmul.ac.uk/~simp/index.html) is working with BAE
SYSTEMS to model complex trade-offs between designs of complex
socio-technical systems of systems such as submarines, as part of the
UK's EPSRC Systems Integration Initiative. We are investigating
whether we can inform decision-making about the design of new systems
based previous design decisions that we represent in complex systems
engineering patterns. We use the I* modelling formalism to represent
the problem, solutions, rationale and trade-offs elicited from BAE
SYSTEMS systems engineers for each such decision. As a result we have
produced a first-draft pattern language for a submarine's manoeuvring
system that I will present at the workshop. However, the development
of this language highlighted the limitations of I* for modelling
different candidate solutions within a design space. I will also
present proposed extensions to the I* formalism for handling complex
non-discrete design spaces, and discuss ongoing work to integrate
multi-criteria decision-making techniques with I* models to provide a
mathematical basis for complex decision-making.