Entity-Based Simulations—

Exploiting Their Benefits to Train Battle Command

by Major Stephen K. Iwicki

The U.S. Army must train to meet a gamut of potential missions ranging from major regional conflicts to security operations and stability operations. Our soldiers will also face a variety of threats from conventional battle to actions against criminal and terrorist organizations. In today’s fiscal environment, however, it is not possible to conduct enough live training exercises to meet all these needs. The Army is relying more on simulations to fill this training void. Today’s current suite of simulations falls short in providing the simulation environment necessary to properly train an information-dominant battle command process. We need reality simulated to the degree we will find on the Force XXI battlefield and displayed by the intelligence and information systems we will use.

Analysts of the future must manage and employ evolving suites of intelligence systems. Many of these systems receive intelligence data from a variety of sources not organic to the unit’s table of organization and equipment (TOE). As technology advances, fewer intelligence systems process increasing amounts of intelligence data. This magnifies our training challenges by forcing more operational requirements on our soldiers, while decreasing their ability to train with these systems. Units cannot afford to fund live Joint Surveillance Target Attack Radar System (Joint STARS) flights ($25,000 per flight hour) to maintain Common Ground Station (CGS) operator proficiency. Even if the funding were available, most theater and national intelli- gence assets are employed in real-world collection missions and are not available to support Army training.

Our doctrine calls on commanders to gain information dominance and to precisely shape their battlespace for decisive operations. They cannot do this without hands-on control of their supporting intelligence processes and a training venue allowing the development of integrated, crossfunctional area technical skills. Moreover, they cannot do this with the current suite of aggregate-level simulations. Fortu- nately, developmental programs will meet those needs shortly.

Aggregate or Entity Simulations?

Aggregate simulations represent a unit and all its personnel, equipment, and behaviors with one simulation icon. We typically represent entire battalions with one simulation icon. With entity simulations, we represent each major platform (truck, tank, or helicopter) with its own icon. The advantage with entity representation is that you can detect, target, and destroy a single object (e.g. tank) and portray the loss of that object to its parent unit. This approach creates the high-fidelity training environment we need to support Force XXI training.

The Problems with  Current Systems

For years, senior Army leaders have repeatedly commented that “Blue” forces always win in Bat- tle Command Training Program (BCTP) exercises where there is a notional enemy and lose to the live opposing forces at the Combat Training Centers (CTCs). One dramatic difference between these two training environments is in the representation of intelligence operations. The challenge is to develop a realistic, well-balanced portrayal of intelligence capabilities in both training environments.

Today, short of live exercises, we train primarily with a constructive simulation called the Corps Battle Simulation (CBS). This aggre- gate-level simulation drives a variety of our command post exercises, including the BCTP Warfighter Exercises. The BCTP Intelligence Collection Model  or a combination of the BICM and the Tactical Simulation (TACSIM) portrays intelligence. BICM unrealistically provides a more accurate portrayal of the battlefield situation than we can create with our real-world capabilities.TACSIM produces realistic products but only models Blue intelligence, surveillance, and reconnaissance (ISR) systems. Additionally, TACSIM is extremely resource-intensive to operate.


When we train with the CBS, we typically fight icons arrayed in echelons, each representing the combat power of 10, 30, or 90 tanks, tracked vehicles, or other high-end pieces of combat equipment. These icons comprise our target base and we feed their locations into the linear, almost assem- bly-line processes devised for annihilation warfare. We attack these targets, and typically view quality training as providing more targets than a linear system can possibly process for attack. We call this “stressing the system to determine weak links.” While there is some merit in this training method, its merit rests with contingencies involving Soviet-style massed forces. The CBS (and the warfare it supposes) train the processes designed to pummel or annihilate the threat. It does not complement our emerging Force XXI doctrine. Force XXI capabilities call for targeting the correct entity with the right lethal weapon or nonlethal force to achieve precision results.

Creating a Precision Training Environment

In this venue, precision results mean more than simply firing Army Tactical Missile Systems (ATACMS) rockets against a tank at 80 kilometers (km). They also mean placing psychological operations loudspeakers at the corner of 5th and Main. Precision operations imply the use of a “needle gun” type of targeting process, applying the force necessary to shape and decide tactical circumstances without the hub-to-hub application of force we have trained to apply.

For example, imagine the effects precision attacks would have against the threat commander’s command and control (C2) if we could simultaneously destroy the tanks of the battalion commander and company commanders surgically, without attacking the rest of their force. The entire unit would have to react to the instantaneous loss of their chain of command. Would they halt, disperse, retreat? Would the shock value make junior leaders hesitant to assume command? Almost certainly, the answer is “yes.” However, we will not determine this while training with the current aggregate-level simulations. The point is simple: we cannot develop precision Force XXI warfighting skills without a precision training venue.

The creation of the precision training environment described requires two concurrent efforts. First, we must ensure entity-level fidelity in the simulation system. The time to accomplish this is during the simulation system’s design and software development. Second, we must give the commander the same hands-on control of his assets that he enjoys in the real world by forcing the simulation to interface with our real-world command, control, communications, computer, and intelligence (C4I) systems.

The first effort calls for object-oriented programming and a high-fidelity synthetic environment. We need battlespace variables replicated at the entity level. We need weather and terrain, cities with traffic jams, threat tank formations, and enemy fire detec- tion-control-delivery processes. We need a virtual reality representation of the real world. This is the only way we can deal with the “fog of war” and properly train as we would fight. This technology is feasible, but it is expensive and time-intensive. We are approaching this caliber of programming with the Warfighter’s Simulation (WARSIM) and its companion simulation, the WARSIM Intelligence Module (WIM). The design of these simulations will meet the demanding training requirements of Force XXI, and will reach initial operating capability (IOC) in April 2001.  I will discuss these systems later in this article.

The second effort—interfaces to organic C4I systems—delivers the high-fidelity simulation control into the commander’s tactical operations center (TOC). We do this today with two training devices called FIRESTORM (the Federation of Intelligence, Reconnaissance, Surveillance, and Targeting, Operations, and Research Models) and CSTAR (the Combat Synthetic Training Assessment Range). These technological applications translate aggregate- level simulation feeds into templated semblances of entity simulations. The important word here is “semblances” of entity representations. FIRESTORM’s tran- slation of an aggregate-level tank battalion into its template 31 tanks may look and feel like entity operations, but it is not (see Figure 1.) Targeting the individual entity requires the commander to find and attack the single tank that represents the aggregate-level battalion icon location with 31 from which to chose. Hitting the right tank yields disproportionate effects by degrad- ing the entire battalion. Destroying any other templated entity has no effect on the tank battalion’s combat strength. While this training methodology is not perfect, it is valuable in developing and exercising the precision targeting processes we will employ on the Force XXI battlefield.

Combat STAR

CSTAR takes the FIRESTORM approach a step further to create realistic intelligence play at the National Training Center (NTC). Our forces training at the NTC do not receive the benefit of intelligence assets that would normally be available to support their operations. The brigade entering the maneuver “box” can bring its direct support intelligence systems, but receives no support from the division and higher-echelon intelligence systems and organizations that would doctrinally support their fight. From an intelligence standpoint, the unit is forced to fight with its feet bound and one hand tied behind its back.

CSTAR will resolve part of that dilemma by integrating the NTC’s live-instrumented ranges with a constructive simulation wrapped around the NTC maneuver area. In effect, it expands NTC to a seamless 300-km by 300-km virtual world (see Figure 2). The simulation “plays” the intelligence assets at division and above in the simulation world. The constructive simulation “tracks” enemy forces until they enter the live maneuver box. NTC integrates the live instrumentation data into the virtual exercise area; the tactical commander in the box “sees” it all as one seamless picture. The commander can see the enemy situation outside the maneuver box. Deep operations employed against the enemy force in the constructive simulation can reduce the size of the force entering the live maneuver box to match the situation in the virtual world. The Simulation, Training and Instrumentation Command (STRICOM) will field CSTAR to Fort Hood, Texas, and NTC in fis- cal year 1999, extending this hands-on capability to III Corps and units training at NTC.

CSTAR made its debut in support of the Task Force XXI Advanced Warfighting Experiment (AWE) in March 1997. FIRESTORM made a strong showing at the Division XXI AWE (DAWE) the following November. Both enabled commanders to integrate entity-level simulations into their battle command processes, but neither have taken us to the level of simulation we need to develop Force XXI battle command and warfighting skills. The aggregate-level simulation employed during the DAWE, for example, matched U.S. Army heavy forces against a similarly equipped Soviet-style conven- tional force. Operational and tac- tical actions focused more on the force-on-force destruction of the threat’s armored and mechanized capabilities, and less toward precisely shaping the battlespace to achieve the precision objectives real-world deployments entail. Soldiers nominated entity-level targets, but these entities were representative of aggregate-level icons. The entity-level fight was illusory, but still had great training value. The commanders and staffs relied on intelligence to visualize the battlefield with systems such as Joint STARS, the unmanned aerial vehicles (UAVs), and the All-Source Analysis System (ASAS). Intelligence systems fed three of the seven large-screen displays in the Division Tactical Command Post (DTAC). Commanders paid significant attention to the information displayed on these screens. They recognized the criticality of these systems, and commensurately extended their battlespace reach.



The Next Generation

The Department of Defense is building a new generation of entity-based simulations for use by all the Services.  The overall project is the Joint Simulation System (JSIMS).  Each of the Services has its respective component piece of the overall project (see Figure 3).  

The Army is building the Warfighter’s Simulation (WARSIM) and the WARSIM Intelligence Module (WIM).  These two simulations will model the objects and processes necessary to support the training of ground combat arms, combat support, and combat service support units.  The Army is also the executive agent responsible for building a synthetic natural environment or virtual operating world for the various simulation models to share.  This will provide the “maneuver space” on land or sea in which our simulations forces operate.

The Air Force is building the National Air and Space Model (NASM).  NASM will model military operations in the atmosphere and in space.  The Navy’s JSIMS Maritime will model naval and amphibious operations.   The Marine Corps is not developing a specific model, but will use portions of WARSIM and JSIMS Maritime to meet its training requirements.


In addition to WIM, two other intelligence simulations are under development within the JSIMS. The developers are creating the National Simulation (NATSIM) and the Joint SIGINT (signals intelligence) Simulation (J-SIGSIM) to model our national intelligence collectors and processes.

These simulations will collectively model the necessary platforms, weapons and sensors to replicate the joint battlefield properly at the entity level.  Together, the JSIMS components will more closely represent reality in the 21st century battle.




In the near-term, we must satisfy the commanders’ present requirements while ensuring continued development toward high-fidelity simulation demands. We must in- corporate the functionality of FIRESTORM, CSTAR, and battlefield visualization capabilities in our future systems. Combat com- manders soon will receive Force XXI equipment such as the CGS, the Outrider Tactical UAV, and ASAS-Remote Workstation, and we can expect a surge in simula- tions training requirements. Training with this equipment using real sensors and systems will prove very expensive (e.g., divisions can ill afford actual Joint STARS support) and physical space will prove constraining (no single post offers a 300-km by 300-km training area). Entity- based simulations will be our only means of developing and maintaining the Force XXI warfighting skills we need on the battlefield.

The author wishes to acknowledge and thank a good friend and outstanding intelligence officer, Major Forrest L. Davis (USA, Retired) for his outstanding contributions to this article. Over the last six years, Forrest and I have worked closely together as fellow Brigade S2s at the 82d Airborne Division and in the modeling and simulations world.

Major Steve Iwicki is currently serving at the 302d MI Battalion in Heidelberg, Germany. He completed a 30-month Pentagon tour as the Army Deputy Chief of Staff for Intelligence Modeling and Simulations Officer. He has also served in intelligence and staff positions at all echelons of the Army, except division staff. Significant positions he has held include Chief of Current Intelligence, Army Central Command during Operations DESERT SHIELD and STORM; XVIII Airborne (ABN) Corps G2 Planner for Opera- tion RESTORE HOPE; Chief, Joint Intelligence Center Rear for Operation UPHOLD DEMOCRACY, Commander of the Army’s first Joint STARS Ground Station Module Company; and S2, 82d Aviation Brigade, 82d ABN Division. MAJ Iwicki holds a Bachelor of Science degree in Decision Sciences and Computers and a Master of Science in Strategic Intelligence. He is a graduate of the Command and General Staff College, and the Post-Graduate Intelligence Pro- gram, and the G2/ACE Chief Course. Readers can reach him via E-mail at g2depace@hq.c5.army.mil and tele- phonically at 011-49-6221-175256 or DST (314) 370-5256.