Decision Aids for Adversarial Planning in Military Operations: Algorithms, Tools, and Turing-test-like Experimental Validation

📝 Abstract

Use of intelligent decision aids can help alleviate the challenges of planning complex operations. We describe integrated algorithms, and a tool capable of translating a high-level concept for a tactical military operation into a fully detailed, actionable plan, producing automatically (or with human guidance) plans with realistic degree of detail and of human-like quality. Tight interleaving of several algorithms – planning, adversary estimates, scheduling, routing, attrition and consumption estimates – comprise the computational approach of this tool. Although originally developed for Army large-unit operations, the technology is generic and also applies to a number of other domains, particularly in critical situations requiring detailed planning within a constrained period of time. In this paper, we focus particularly on the engineering tradeoffs in the design of the tool. In an experimental evaluation, reminiscent of the Turing test, the tool’s performance compared favorably with human planners.

💡 Analysis

Use of intelligent decision aids can help alleviate the challenges of planning complex operations. We describe integrated algorithms, and a tool capable of translating a high-level concept for a tactical military operation into a fully detailed, actionable plan, producing automatically (or with human guidance) plans with realistic degree of detail and of human-like quality. Tight interleaving of several algorithms – planning, adversary estimates, scheduling, routing, attrition and consumption estimates – comprise the computational approach of this tool. Although originally developed for Army large-unit operations, the technology is generic and also applies to a number of other domains, particularly in critical situations requiring detailed planning within a constrained period of time. In this paper, we focus particularly on the engineering tradeoffs in the design of the tool. In an experimental evaluation, reminiscent of the Turing test, the tool’s performance compared favorably with human planners.

📄 Content

Decision Aids for Adversarial Planning in Military Operations: Algorithms, Tools, and Turing-test-like Experimental Validation

Alexander Kott Ray Budd Larry Ground Lakshmi Rebbapragada John Langston

2 Abstract Use of intelligent decision aids can help alleviate the challenges of planning complex operations. We describe integrated algorithms, and a tool capable of translating a high-level concept for a tactical military operation into a fully detailed, actionable plan, producing automatically (or with human guidance) plans with realistic degree of detail and of human-like quality. Tight interleaving of several algorithms – planning, adversary estimates, scheduling, routing, attrition and consumption estimates – comprise the computational approach of this tool. Although originally developed for Army large-unit operations, the technology is generic and also applies to a number of other domains, particularly in critical situations requiring detailed planning within a constrained period of time. In this paper, we focus particularly on the engineering tradeoffs in the design of the tool. In an experimental evaluation, reminiscent of the Turing test, the tool’s performance compared favorably with human planners.

3 The Quest for New Processes and Tools The US Army1 is exploring a significant computerization of the military planning process [1]: “…the Army must create fast new planning processes that establish a new division of labor between man and machine. … Decision aids will quickly offer suggestions and test alternative courses of actions.”

The reasons for exploring potential benefits of such decision aids are multifaceted. The process of planning an Army operation remains relatively cumbersome, inflexible and slow. Life and death decisions are made by a relatively meager staff of perhaps 3-4 individuals, working with imperfect information under significant time constraints. Success typically depends on one’s ability to synchronize the movement of units and the concentration of firepower and other effects at a precise moment in time.

The tools available to the planners are generally limited to rudimentary decision aids for analyzing the terrain of the battlefield and some office automation tools to help record decisions. The planners bring varying degrees of knowledge, experience and prejudices to the process. The planning process frequently involves disagreements on estimation of outcomes, enemy reactions, attrition and consumption of supplies. There is a fundamental complexity of synchronization and effective utilization of multiple heterogeneous assets performing numerous, inter-dependent, heterogeneous tasks.

The Course of Action Development and Evaluation Tool (CADET), is a tool for producing automatically (or with human guidance) the detailed tasks required to translate a basic concept into a fully formed, actionable plan, which is a key step in the military’s standard decision making process.
This step involves taking the proposed courses of action for the friendly forces, developed in a previous step and initially expressed as high-level concepts, and expanding them into the hundreds of supporting tasks required to accomplish the intended objective. Concurrently, the friendly course of action is tested

1 Views expressed in this paper are those of the authors and do not necessarily reflect those of the U. S. Army or any agency of the U.S. government.

4 against the most likely and/or the most dangerous courses of action available to the enemy. The intent of this analysis, called wargaming in the military, is to produce an analytical baseline from which the commander can choose the best course of action [2]. The process is particularly challenging for a relatively large and complex unit such as a Division, where the actions of 12,000 soldiers and over 3,000 weapons must be coordinated to produce a desired effect. Wargaming a potential tactical course of action for such a large unit typically involves a staff of 3-4 persons with in-depth knowledge of both friendly and enemy tactics.2

The input for their effort comes usually from the unit Commander in the form of two doctrinally defined products: a Course of Action (COA) sketch (e.g., Fig. 1) and a Course of Action statement – a high-level specification of the operation. In effect, such a sketch and statement comprise a set of high- level actions, goals, and sequencing, referring largely to movements and objectives of the friendly forces, e.g., “Task Force Arrow attacks along axis Bull to complete the destruction of the 2nd Red Battalion.” Typically, the unit Commander will develop a minimum of three courses of action for consideration, distinguishable from one another in force composition and application, designation of main and supporting efforts or in utilization of terrain and resources. The human plann

This content is AI-processed based on ArXiv data.