Bridging the gap

Abstract

In Operations Research and Management Science (OR/MS), Discrete Event Simulation (DES) models are typically created using commercial off-the-shelf simulation packages (CSPs) such as AnyLogic TM , Arena TM , Flexsim TM , Simul8 TM , SLX TM , Witness TM , and so on. A DES model represents the processes associated with a system of interest. Some models may be composed of submodels running in their own CSPs on different computers linked together over a communications network via distributed simulation software. The creation of a distributed simulation with CSPs is still complex and typically requires a partnership of problem owners, modelers, CSP vendors, and distributed simulation specialists. In an attempt to simplify this development and foster discussion between modelers and technologists, the SISO-STD-006-2010 Standard for COTS Simulation Package Interoperability Reference Models has been developed. The standard makes it possible to capture interoperability capabilities and requirements at a DES modeling level rather than a computing technical level. For example, it allows requirements for entity transfer between models to be clearly specified in DES terms (e.g. the relationship between departure and arrival simulation times and input element (queue, workstation, etc.), buffering rules, and entity priority, instead of using specialist technical terminology. This article explores the motivations for distributed simulation in this area, related work, and the rationale for the standard. The four Types of Interoperability Reference Model described in the standard are discussed and presented (A. Entity Transfer, B. Shared Resource, C. Shared Event, and D. Shared Data Structure). Case studies in healthcare and manufacturing are given to demonstrate how the standard is used in practice. ACM Reference Format:Taylor, S. J. E., Turner, S. J., Strassburger, S., and Mustafee, N. 2012. Bridging the gap: A standards-based approach to OR/MS distributed simulation. ACM Trans. Model. Comput.

Distributed simulation: state-of-the-art and potential for operational research

“…Figure 1 (b) shows the far simpler alternative presented by DS where simulations are kept separate but run together. On this basis several authors have identified arguments for DS from an OR perspective (Mustafee, et al 2012;Taylor, et al 2012;Boer, de Bruin, & Verbraeck, 2009;Lendermann et al, 2007). These include:…”

“…Long (2016) presents a comprehensive (non-HLA) DS for supply chain simulation. As noted earlier in this paper, some work has produced standards that have augmented the HLA for use in this area Taylor et al, 2012). Associated work has studied the transformation of Business Process Models into DS (P. Bocciarelli, Pieroni, Gianni, & D'Ambrogio, 2012).…”

Hybrid Simulation Development – Is It Just Analytics?

“…In general, it will be highly beneficial to stay within the boundaries of a single simulation system even when performing multi-method modeling. Although distributed simulation based on the High Level Architecture (HLA) (Strassburger 2001) in principle offers a solution for interoperability between heterogeneous simulation systems (Taylor et al 2012), and thus also enables hybrid systems modeling, the ease-of-use argument still weighs heavily when deciding which modeling tools to use. For some scenarios, especially when existing and complex model components shall be reused and brought together, distributed simulation may be the best choice for enabling hybrid simulation, for others, especially when building models from scratch, it may not.…”

A Web Based Architecture to Operationalize Geospatial Simulation Environments

“…• Modeling and simulation as a service (MSaaS) has been consistently identified as a necessary step towards increased accessibility to simulation in decision-making processes [29], [32], [33]. Cloud-based MSaaS has been a way to decrease the technological cost of simulation-based processes, to provide access to high performance, scalable simulation hardware and software.…”