Towards a Distributed Simulation Interoperability (DSI) framework to address interoperability issues for simulation practitioners

Abstract

Simulations are now used more frequently to conduct experiments on real world or proposed systems, to understand the system behaviour, or for evaluating improvement strategies. Over time, the need has grown for big enterprise businesses to developed sophisticated and complex systems to compete with industry. Also, these businesses are now more connected with each other like a networked enterprise. This has further raised the requirement for developing more and more complex simulations that can interconnect with other businesses. Distributed simulation has been widely used in this context in military applications, but such popularity has not grown in other sectors. The reason behind this is the technical expertise required to establish communication protocols between distributed simulations. There have been efforts by research industry to bridge this gap and the most important work has been development of the High Level Architecture (HLA) standard for providing common communication protocols between distributed simulation models. The Modelling and Simulation (M&S) industry also provides a lot of literature for developers on modelling standalone simulation. The focus of conceptual modelling in this case has been on model accuracy and efficiency instead of interoperability. This is also discussed in detail in this research. Practitioners have also struggled to find support for underlying technologies until most recently. But with the introduction of standard Runtime Infrastructure (RTI) and simulation development platform support this gap has narrowed. The HLA standard promised to resolve interoperability issues between distributed simulation models, but only managed to provide standard guidelines up to syntactic level. Therefore, the Simulation Interoperability Standards Organisation (SISO) carried the research forward and identified the interoperability problems faced by practitioners at the semantic level and drew up a list of interoperability issues. However, the published standard SISO-STD-006-2010 only identified the problems but did not provide the semantic solution. The main contribution of this research has been the Distributed Simulation Interoperability (DSI) Framework that identifies semantic solutions for the interoperability problems listed in the Commercial-off-the-shelf Simulation Package Interoperability Reference Models (SISO-STD-006-2010). This research recommends including these interoperability semantic solutions in HLA Object Modelling Template specifications. By doing so, this will help industry practitioners achieve the interoperability promise made by HLA and make distributed simulation models more re-usable and composable.