Exploring DSM to Support Systems Engineering of ComposableSimulation Environments
DSM 2010: Proceedings of the 12th International DSM Conference, Cambridge, UK, 22.-23.07.2010
Year: 2010
Editor: Wynn, D.C.; Kreimeyer, M.; Eben, K.; Maurer, M.; Lindemann, U.; Clarkson, P.J.
Author: Roca, R.
Section: Industry Applications
Page(s): 419-431
Abstract
A critical distinction between distributed computing in general and distributed simulation specifically is rooted in that simulation systems are meta-systems such that their underlying models are essentially complex, multi-disciplinary systems-within-systems. A simulation environment refers to the software and hardware facilities that are integrated and configured in order to generate a set of simulated conditions for analysis, testing, and/or training. As such, systems engineering (SE) directed at modeling & simulation (M&S) must deal with complexity that extends beyond the more objective operational infrastructure of simulation environments to the reconciliation of more subjective multidisciplinary syntactic and semantic concerns associated with their underlying constituent models. Simulation composability has proven to be one of the most difficult challenges in the M&S frontier. Even the most experienced practitioners in this area of simulation research dismiss the illusion of pure plug-and-play composability – calling instead for the elaboration of more realistic M&S SE methods that address simulation complexity so as to minimize the composition time and level of effort in simulation-environment design and development. Intuitively, an effective M&S SE approach should consider a framework that facilitates the reconciliation of such fundamental concepts as fidelity, composability, and validation; these and other familiar concepts have not been standardized in the simulation community leading to some level of syntactic and semantic disconnect among M&S practitioners and other stakeholders that exacerbates the complexity of simulation projects. Simulation fidelity, in particular, has been the source of much debate and even controversy yet it holds the potential to resolve complexity as it is a key multidimensional concept that characterizes and influences the conceptualization, specification, design, and development of simulation environments. The Multiple Domain Matrix (MDM) derivative of the Design Structure Matrix (DSM) method has been employed successfully to analyze and manage crossdomain complexities in product development, manufacturing, construction, process control, software design, and various other engineering settings. This paper presents an initial approach that attempts to leverage the MDM as a framework for the management of complex simulation fidelity dimensions as an enabler of composable simulation environments.
Keywords: Design Structure Matrix, Multiple Domain Matrix, Complexity Management, Composability, Interoperability, Modeling & Simulation, Systems Engineering, Fidelity, Verification & Validation