Modeling and analysis of enhanced information and decision-making through introduction of combat information system

Abstract

Nowadays, cutting edge information system has been developed and adopted into a military organization to overcome an uncertain circumstance of warfare. We also indirectly experienced the effects of information system through the real warfare such as, Gulf War, Iraq War, and Afghanistan War. The concept of Network Centric Warfare(NCW) and Effect based Operation(EBO) are same line with an introducing of an advanced information system. Thus, it needs to quantitatively evaluate these effectiveness of information system adoption for various reasons. To introduce a more effective system, it needs to evaluate in perspective of cost-benefitt analysis. In addition, it needs to adjust parameters of system capability to maximize a combat effectiveness. Among the many analysis approaches, M&S(Modeling and Simulation) method has several advantages to analyzing complex and dynamic environment, especially for combat situation. However, we have faced some troubles when we evaluate the effectiveness of information by current simulation models. First, it doesn't have functions to simulate distortion and loss of information. Second, there is no function to simulate human-like decision-making. Therefore, this dissertation addresses a modeling method how to overcome these problems for evaluating the effectiveness of combat information.In chapter 2, to address the problem, I firstly define a combat information, related flow, and activities regarding combat situation. I categorize it into the three phases of information flow such as collection, transmit, and utilizing. In each phase, I investigate the related information and activities. Finally, I suggest key concepts of modeling requirement for evaluating the effect of information. The proposed modeling requirements consist of two levels. The first level is a simple decision process, we name it engagement level decision process. The second level is a contingency decision process, we name it tactical level decision process. The engagement level decision can be defined that some portion of increased combat effectiveness comes from the detecting and striking of targets at long range. The tactical level decision can be defined that some portion of increased combat effectiveness comes from the command and control of its own force's dispatch. Both decisions are strongly related to the enhanced information. To model the decision process of human, I introduce a concept of uncertainty which has three dimensions

location of uncertainty, sources of uncertainty, and level of uncertainty. Consequently, I suggest a speicific location, sources, and level of uncertainty to evaluate the combat effectiveness by enhanced information. In chapter 3, I perform two approaches to examine the relationship between the varied information and the quality of decision-making. The rest approach is model-based experiment. The model is constructed by Bayesian Networks which consist two decision levels

engagement level decision and tactical level decision. Building two Bayesian Network, I conduct two experiments how to vary the performance of decision-making by varied information. In the second approach, I conduct human subject experiments which consist of two phase experiments. In the 1st phase experiment, I screen the major factors to 2nd which criteria of information affects the quality of human decision-making. After the 1st phase experiment, I merge the severe quality criteria into levels of 'low', 'medium', and 'high' like a real world, then perform the main experiment to 2nd how quantity and quality of information could affect to the human decision-making. The experiments are conducted through computer-based simulated C2 in a laboratory environment. Throughout two approaches, I find coping strategies of human when receives imperfect information. Based on these results, I suggest a decision-making logic of human for evaluating combat effectiveness by enhanced information In chapter 4, I apply the results of experiments into the combat simulation. The objective of combat simulation is a finding the best candidate ISR system for a introducing the next generation infantry battalion. To reflect a real world situation, the real systems are abstracted. While the candidate A system is the current UAV system, the candidate B system is rotor-based micro UAV system. I set up a baseline candidate to compare the combat effectiveness of two policies based on introducing ISR system. The scenario in the combat simulation is built based on a typical defensive operation. In the scenario, there are four main decision points to simulate the effectiveness of decision-making. Consequently, the results of the combat experiments are evaluated in forms of combat effectiveness. Throughout this dissertation, we can understand the relationship between information and combat effectiveness. Comparing to the previous works, this dissertation tries to 2nd sources of enhanced combat effectiveness based on enhanced information. The key factors for understanding this relationship are simulating human-like decision-making. Although this dissertation focuses on the specific scenario and domain, it could be a very first step to understand for simulating of human-like decision-making in a military domain specific.