Читать книгу Simulation and Wargaming - Группа авторов - Страница 17

Modern wargaming centers provide at least three components to support the wargame, namely the operational components, analysis component, and the simulation component. The operations components prepare, conduct, and evaluate the wargame. Since wargames have been conducted, this group has been the important counterpart to the subject matter experts who participate in the wargame itself. To make a wargame successful, it needs to be defined, planned, designed, developed, rehearsed, and finally conducted. The operations group is responsible for all these tasks, from the first ideas to the detailed game plan. During and after the game, they must analyze the results and prepare evaluation reports, outbrief presentations, etc. Some of them may be given as interim reports to the subject matter experts, others are collected to provide the overarching insights captured in the final reports about the wargame. In the earlier days of wargaming, the experts analyzed the situation by themselves, very much like they would do in headquarters. With the increasing complexity of the situation on the battlefield and a more complex solution space, more professional support needed to be provided. Within the defense domain, this analysis group is referred to as Operations Research & System Analysis (ORSA). ORSA experts assist decision‐makers in solving complex problems by producing the analysis and logical reasoning necessary to inform and underpin those critical decisions. They are as much part of modern headquarters as they are part of wargaming support components. The simulation component provides numerical insight into the dynamic behavior of the complex battlefield. This is the youngest component, as only with the rise of computational capabilities was it possible to develop simulation systems that implement the theory of war, movement, attrition, and other relevant effects through the computational representation of entities, relations, activities, and effects. While traditionally rooted in the domain of physics‐based modeling of mostly kinetic phenomena, recent developments in human and organizational behavior modeling research address such elements of modern warfare as well. As such, simulations did not only replace the rulebooks and result tables of traditional wargames, but also support the ORSA group with the evaluation of decision spaces. Furthermore, modern simulation systems provide powerful interfaces that allow not only the immersive displays of combat simulations, they also provide analytic tools to capture and display wargame metrics interactively.

This section evaluates the role of simulation in more detail to show that simulation is a powerful tool in support of wargaming in many phases, from early design to the generation of after‐action reviews and reports. However, there are also several modeling challenges that must be addressed to ensure the best use of these powerful methods.

In order to better understand the potential as well as the pitfalls and dangers of simulation in wargaming, it is necessary to clearly understand modeling and simulation. Every simulation is based on a model, if this model has been documented explicitly, or if it is just implicitly captured in the form of the concepts, properties, relations, and processes implemented in the simulation. A model is a task‐driven, purposeful simplification and abstraction of a perception of reality. As so often, the modeling process starts with the problem of the sponsor, which can be a research question to be answered, or a training task to be conducted, or ideas for a new system or new tactics, techniques, and procedures – or even doctrine – to be evaluated. The task usually drives the required abstraction level: is the sponsor's problem on the level of system components, or the entity level of weapon systems, or are units and organizations the topic of concern? Once the abstraction level is clear, not everything on this level is important. Just as scientists plan their experiments with focus on the research question, so too modelers focus on only the concepts of interest, simplifying their perception to the essential components. This perception is shaped by physical, cognitive, and even moral constraints: It reflects the understanding of the modeler, and is shaped by knowledge, experience, and other factors. Therefore, two models from modelers with different background can be quite different, even if they start with the same problem and the same references.

A simulation implements such a model. Simulations are often understood as the execution of models over time, and in the scope of this chapter, the focus is on those using computers to execute a programmed version of the model to do so. The implementation is characterized by numerical challenges, computational complexity, and use of heuristics. Different programming languages, compilers, and platforms add more challenges. Even the same model can therefore result in various and quite different simulations. Even the change of the hardware can lead to surprising changes in predicted outcomes, in particular in complex, nonlinear systems with a high dependency on the initial conditions. When NATO upgraded their hardware, some of the important analysis results had to be revisited, as some battle outcomes changed significantly using the new hardware. This is not a mistake of any programmer, it is just the nature of highly complex, nonlinear systems that become discretized and solved numerically.

Despite such obstacles, modeling and simulation is a powerful tool that helps to reproduce well‐known effects, predominantly of physical and kinetic nature, under diverse circumstances and constraints. The next subsection will evaluate where within the process of wargaming simulation can be of help.

During the execution of the wargame, the focus will be on the human players. They provide the creativity allowing for operational agility in planning and decision‐making. They have the insights to support new ideas, such as multidomain operations planning in the national and international context. They understand how to explore human decision‐making and how to react to unanticipated decisions in the operation. In summary, they are the main players in the wargame, providing creativity and the vision for the big picture. However, the role of simulation is similarly important. It falls to the simulation to compute the mission thread by unbiased execution of decisions in the virtual battle space. This includes computing the mission thread effects as well as the effects of the wargamer’s human decisions. Simulations compute all orders of effects, and some higher order effects in complex, nonlinear environments can be surprising. It is not only possible but highly likely that some of these effects will lead to emergent behavior in the scenario, properties that the complex systems expose, but that cannot be exposed by the individual systems themselves. The immersive visualization of results, including detection and visualization of new emergent behaviors, is a pivotal role for the simulation. In other words, simulation can take over the role of secondary players, opposing forces, and supporting roles in planning and preparation, while also supporting computational and visualization requirements in execution and evaluation.

However, the use of modeling and simulation is not limited to the execution phase but can be applied by the operations group in support of many of their tasks. During the design and the development of the wargame, tools can help to visualize ideas and support the composition and reuse of services and rules developed in earlier wargames. Quick consistency checks can make sure that all entities necessary to evaluate a new idea are represented, and all of them have rules assigned that can help with the necessary scenario generation process. For rehearsing, artificial intelligence can be used to calibrate software agents and rules to play the role of human gamers to look for inconsistencies, opportunities to cheat, and other optimization of actions. It may be premature to think about artificial gamers as subject matter experts, but for the pure testing of the limits of the game, current technologies are sufficient. Communicating the results of the wargame is also a task well known by the simulation community, as in the domains of analysis and training providing after action reviews (AAR) has been required for many years. If done correctly, results and lessons learned can inform the development of new rules for the next events. A picture says more than a thousand words, but an executable simulation can say more than a thousand pictures. Scenario generation tools and other support software should also be utilized in this context.

There are several challenges that simulations for defense operations must overcome to be fully supportive of today’s requirements. Nonetheless, the increasing complexity of the highly nonlinear operational environment needs such computational support. The recently published primer on complexity for systems engineers, developed and published by the International Council on Systems Engineering (INCOSE), explicitly mentions simulation and artificial intelligence methods as necessary tools for decision‐makers in such environments, as complexity requires a new operational agility from the decision‐makers, which means to rapidly compose high‐performance teams out of the available systems to react quickly and precisely to often unforeseen, maybe even emergent challenges.

Simulation solutions provided for defense are reasonably effective in the modeling of physical and kinetic effects, such as needed for attrition‐focused force‐on‐force modeling. However, the structure of the opposing forces is changing rapidly, driven by increased use of robotic systems and other autonomous systems that lead to new tactics and procedures. New weapon systems, such as the 5th‐and 6th‐generation systems, provide a new set of capabilities. Opposing systems and future systems are hard to capture, as the parameters – or even the underlying architecture – are unknown or uncertain. Many lessons learned are no longer applicable.

How and where we will have to solve future conflicts is not only challenged by more than such technical uncertainties. With more and more people moving toward mega‐cities, many of them in coastal regions, the likelihood of an armed conflict in these regions increases. This will require high‐resolution modeling of this environment with high fidelity on a big scale. This will require sensor and weapon system models with equal resolution and fidelity, and the adaptation of rule sets on how to apply these systems, for all participating organizations. International multidomain operations require a new level of coordination between the systems of various services and nations as well as the local commanders utilizing their capabilities. These new kinds of operations are more than joint and combined activities, they are the creative mix of several mission threads optimally creating mutually supportive effects in all domains.

Furthermore, human, cultural, and social behavior modeling will be needed, which implies the use of computational social science models for both opposing and friendly forces. With the advancement of combat medicine saving more soldiers, new challenges emerged, like having to deal with post‐traumatic stress syndrome (PTSD). The use of information, including social media, to influence opponents, disseminate information in support of the objective of the organization, and other nontraditional intelligence operations may influence future warfare as well.

The traditional use of predictive simulations used for point optimizations in a well‐defined context, possibly supported by some sensitivity analysis, does not meet these emerging requirements. Composable simulation services provided as smart components are needed. The resulting compositions need to be applied to conduct exploratory modeling and analysis addressing the deep uncertainties of these complex environments by allowing a broad evaluation of the solution space. By combining the power of computer simulation‐based generation of data with technology of big data allows for a new application of simulation.

In summary, modeling, simulation, analysis, and visualization methods can and should enrich wargaming activities. No other methods allow the exploitation of options within a complex, nonlinear environment, such as the modern battlefield presents. Several chapters in this book provide examples of how these methods and derived tools help in the decision‐making process. Not utilizing these methods and tools to the full extent possible would be a mistake.