Simulation Approaches for System of Systems: Events-based versus Agent Based Modeling

Abstract

In order to understand how we may engineer system of systems (SoS), we will have to rely significantly on our abilities in modeling and simulation. While there are some models of non-specific SoS, few attempts have been made to demonstrate these models with simulation. Simulations of SoS are usually specific cases. Likewise, in the description of a SoS, most of the approaches have focused on their characterization, yet this characterization has not been greatly utilized as an underlying feature in the modeling and simulation of SoS. We review different modeling techniques and use two converse techniques, i.e. agentbased and event-based modeling, to run a simulation of hypothetical systems collaborating into a SoS. The results of the empirical comparison indicate an agent-based modeling approach would achieve a characterization model better with validation achieved through an event-based approach.

Stimuli-SoS: a model-based approach to derive stimuli generators for simulations of systems-of-systems software architectures

“…M&S enable visualization of SoS dynamics [17,[22][23][24][25]. Several application domains adopt M&S [17].…”

“…Baldwin et al summarize current techniques found in the literature to simulate SoS [24]. Event-based modeling is the most prominent approach, as researchers can program different states a system undergoes to comprehend the behavior of the SoS as a whole [24].…”

“…Event-based modeling is the most prominent approach, as researchers can program different states a system undergoes to comprehend the behavior of the SoS as a whole [24]. In particular, DEVS is the most popular event-based simulation formalism [30].…”

Collective understanding of system behavior : incorporating simulations in conceptual systems design

“…Interactions in lower domains are clarified [Erden et al, 2008] Mathematical systems theory Multidimensional feedback systems which can either be continuous, sampled (temporal discrete) and discrete event Understanding feedback operations in systems [Maier and Rechtin, 2000] [Maier and Rechtin, 2000]. Identify emergent behavior [Baldwin et al, 2015] Public choice and behavior models Models of human behavior and decision making Understanding the human system [Maier and Rechtin, 2000] Considering the threads and scenario type, especially scenarios are useful tools to envisage future usage scenarios and contexts for the system [Ionita, 2005]. More recently, [Curry and Ross, 2015] discuss extensions to the Epoch-Era Analysis framework.…”

“…This often allows modeling of emergent behavior. This is especially useful when modeling Systems of Systems [Baldwin et al, 2015].…”