運用管理科學方法建構戰場管理資訊化決策支援系統模式之研究

Applying management science methods for modeling decision support systems used in digitized battlefield management

李豐池(Fong-Chih Lee)

指導教授 ： 陳文華

國立臺灣大學/管理學院/高階公共管理組/碩士(2007年)

https://doi.org/10.6342/NTU.2007.10076

摘要

資訊技術的快速發展及其在軍事領域的廣泛應用，有力地推動著各國軍隊資訊化建設的步伐。軍隊指揮自動化建置成果直接反映一個國家軍事資訊化的程度，同時也是現在戰爭中掌握資訊優勢、實施聯合作戰指揮的重要基礎和根本條件。指揮自動化理論及指揮自動化系統的應用，是各級指揮人員管理部隊和進行作戰指揮的必備知識。
本研究旨在應用管理科學方法以其研究人類各種管理活動的科學方法。對人們為實現預期目標而對各種實踐活動實施科學的計劃、組織、指導、控制和協調的一系列組織管理方法。雖然管理科學是ㄧ個範圍廣泛、對象複雜的科學門類，但在運用管理科學方法對各種行為進行管理的一般程序大致為：1.確定管理目標，管理目標往往追求以最短時間、最少的支出得到最大的效果；2.尋找影響管理目標實現的有關因素；3.建立被管理對象的系統模型(或數學模型)；4.運用現代管理手段或理論對系統模型(或數學模型)進行分析、評價(或求解)；5.根據分析、評價(或解出)的結果選擇出最優方案或直接做出最佳決策；6.通過計劃、控制協調等手段將選擇的方案或決策付諸實施；7.對實施情況及由決策所發生的各種行為加以檢查、監督，從而達到最初確定的管理目標。
決策是人類的一項基本活動，是人(們)行動的先導。決策學是為決策提供科學的理論和方法，以支持和方便人們作決策的科學，是自然科學與社會科學並涉及人類思維的新興交叉學科。在科學技術不斷進步，社會經濟迅速發展，資源和生態環境日漸惡化的情況下，人類正面臨著越來越多的重大、複雜而困難的問題有待決策。
因此，世界各國都在大量經驗的基礎上研究決策的理論方法，以保證決策的正確和有效性。從而逐漸形成了一門新興的科學-決策科學。決策科學研究的內容很廣泛，從廣義來說，包括決策思想、決策理論原理、決策方法和決策支持環境等。尤其在作戰指揮領域，決策既包括下定決心，也包括制定作戰方案。眾所周知，在作戰中一切指揮要素都必須以決心作為依據和出發點，決心錯了，將付出高昂的代價。此外，在具體實現決心的過程中，比如擬制作戰計劃、部署兵力武器、任務分工和協同動作等的過程中，也都離不開決策。七十年代以後，隨著決策問題的日益複雜化，人們逐步認識到單純依靠數學分析方法已經遠遠不夠了，必須把定量分析與定性分析結合起來。軍事運籌學(作業研究)是一門為軍事決策提供科學方法的軍事學科。
決策是客觀事物及其發展規律在決策者頭腦中的反映和邏輯表達，決策過程是一種認識過程。為了實現科學決策，就需要一個關於認識的正確理論原理做為指導。認識論把人類的社會實踐作為全部隊認識的基礎，從認識與實踐的相互作用中科學地說明了認識的形成、認識的本質及其發展規律，指出人的認識能力是隨者實踐的發展而變化而發展的，人的認識過程是一個由不知到知，由淺到深的充滿矛盾的辨證過程。
科學的決策程序是實現決策目標重要保證。決策程序可以分為六個步驟。這六個步驟形成一個決策環:第一步是調查研究，也就是全面了解和掌握與決策有關的實際情況和資料，特別是儘可能獲得完整可靠的統計數據。第二步是確定目標。ㄧ個好的決策目標應該具有各種約束(限制)條件下的實際可行性，同時又對所希望達到的結果具有關鍵意義。第三步是擬定可能採用的各種備選方案。能否擬定出有價值的、高質量的備選方案，是決策過程中很關鍵的一步。這些備選方案應是經過驗證證明是能夠實現目標的可行方案，同時，每個方案應有各自的特點，體現不同的利弊、得失，從而為下一步的選優決斷打下堅實的基礎。第四步是分析審定、選優決斷。即從各個備選方案中通過價值評價和可行性分析選擇出最優的(或滿意的)方案。上面所說四個步驟是一個認識過程。其中第一步和第二步，即調查研究、發現問題、建立目標，是決策的準備階段，大致相當於認識過程的感性階段；第三步和第四步，即資訊分析、設計備選方案、擇優決斷等，是決策的完成階段，大致相當於認識過程的理性階段。第五步：實施決策；第六步：追蹤決策。所以決策過程的最後兩個步驟可以叫做決策的執行與檢驗階段或實驗階段。
在現代條件下，不僅戰爭的情況十分錯綜複雜，而且往往瞬息萬變，若依靠人腦決策，由於受到生理功能的制約，很難適應客觀的需要。如戰術火力系統的運用決策，完全是在對抗的環境條件下進行的對策活動，由發現目標、敵我識別、選擇方案、對敵追蹤、進行發射、火力轉移等，必須在極短時間裡完成，真是分秒必爭，以致毫秒必奪，延遲不僅會失去戰機，而且還意味著被動挨打。因此在使用高技術化武器裝備的條件下，要求反應速度之快，而且所涉及的各方面又是如此之多，完全靠指參人員臨時收集和加工情報資料，擬制和選擇行動方案，必是十分困難的。然而隨著每秒千億次以上巨型電子計算機的出現和應用，尤其是智能電子計算機的問世，將會有真正的”電腦參謀”、”電腦助手”，這就為解決這些難題提供了充分有利的條件。所謂人-機共生決策方法即是戰場管理自動化指管系統C4ISR，就是指揮官、決策者運用人工智能技術實現決策自動化的方法。這種方法的特點，既不單純依靠人腦進行決策，也不單純靠電腦進行決策，而是由人和智能機器(電腦硬體與軟體應用系統)協調地、共同地、自動地完成決策過程。運用這種方法進行決策，不僅可以繼續發揮人的主動性、積極性和創造性，而且通過使用智能化資訊，可以克服人反應與計算速度慢、長時間工作容易疲勞、受心理波動影響、存儲資訊能力有限等弱點，以人與資訊化設備相互之截長補短，從而能夠應付任何複雜和變化速度極快的戰場上的軍事活動。
C4ISR系統的發展，主要是藉著全新的作戰架構、技術架構、系統架構和網狀化作戰概念，將指揮、管制、通信、資訊(電腦)、情報、監視、偵察等系統，同時作縱向與橫向的全面整合。作戰人員及決策著雖已發展完成許多架構，以律定資訊技術之需求、功能描述及提出解決方案。然許多單位均以建立特定目的之架構。此類架構常用類似術語說明不同事物。由於缺乏架構發展之標準步驟，更甚者，架構使用共同術語之結果，導致無法快速相互比較及整合。本研究即在運用現代化管理科學方法建構一發展模式以補充原架構作業方式之含混模糊所造成之發展與需求相背離之情況，期能發展與建構確符需求之戰場管理決策資訊化支援系統。

關鍵字

管理科學方法 ； 系統分析 ； 程序化決策方法 ； 系統功能圖法 ； 科學決策方法 ； 決策支援系統 ； 結構化程序設計

並列摘要

Due to the fast growth in IT (Information Technology) and its broad utilization inside military community, the pace in each nation’s forces IT development and modernization has been promoted with high emphasis. The results in military command automation directly reflect the degree of a certain nation’s military IT development and modernization, also, these results are significant and fundamental prerequisite for achieving information superiority in modern warfare when conducting joint command. Theories upon command automation and applications on commanding automation systems should be general knowledge for commanders at each echelon to manage military forces and to command military operations.
This research aims to use of scientific methods, i.e., management scientific methods, to study human various management activities. For human beings to carry out anticipated objectives, a series of organizational management methods including scientific planning, organizing, leadership, controlling and collaboration are used to complete each implementation task. Although, management science is broad and complex, the utilization of management scientific methods for conducting each task, in general, has the following procedures. 1. Confirmation of management objectives and completion of these objectives with the shortest schedule, the highest effectiveness, and the least expenditure. 2. Identification related factors that will influence achieving objectives. 3. Establishment of system models (or mathematic models) for persons that are managed. 4. Analysis and evaluation of the models with the use of modern management means or theories. 5. Selection of the promising operation or direct conduct of the best decision making based on the results of analysis and evaluation. 6. Realization of the option and decision made by means of planning, controlling, collaboration, etc. 7. Inspection and supervision of various behaviors derived from implementing the option or decision to ensure the initial management objectives.
Decision making is a fundamental activity for human beings that lead human beings’ actions. Decision sciences are to provide scientific theories and methods so as to support and facilitate human beings to make decision. Decision sciences involve natural science and social science that are new and mixed subjects related to human beings’ thinking. Over the constant progress in science and technology, as well as the fast growth in social economy with the impacts that resources and ecological environment are getting worse, human beings are facing with more and more severe and complicate problems waiting for resolution.
Therefore, each nation has paid close attentions to the research on theoretic methods of decision sciences with the basis of a great amount of existing experiences, so as to ensure the accuracy and effectiveness when making decisions. Gradually, a new science is developed, i.e., decision science. The research on decision sciences has its broad content. Broadly speaking, decision sciences include decision thought, decision theories or principles, decision methods, decision support environment, etc. Particularly, in the domain of operational command, decision includes not only decision making, but also operational course of actions. Considering military operations, all commanding factors depend upon decision making. If wrong decision is made, a high value or cost will be wasteful. Additionally, the process for realization of the decision, such as proposing operational plans, deployment of forces and weapons, mission allocation and collaboration are closely related to decision making. After 1980, the level of complexity in decision problems has been increased, people recognize that merely mathematic analytic methods are not enough, a combination of quantitative and qualitative analysis is thus considered necessary. Military operation research is a military subject providing scientific methods for military decision.
Decision is an objective manner and its developing regulations reflect and logically express inside a decider’s mind. Decision is a process of recognition. To realize scientific decision, a theoretic principle regarding recognition for guidance is required. Epistemology is based on human beings social realization, mutual activity from recognition to realization scientifically expressing the formation of recognition. The essence of recognition and its developing regulation identify people’s recognition capability that develops from realization and thus changes for further development. The recognition process is derived from unknown to known, from shallow to deep, and is full of contradictory recognition process.
Scientific decision making process is a significant assurance to embody decision objectives. There are six procedures comprising of a decision making process. These procedures altogether build up a decision ring. The first procedure is to investigate situations, i.e. to fully understand and handle real situations and data, particularly, acquiring as complete and reliable statistical data as possible. The second is to ensure objectives. An ideal decision objective should be practical and feasible under all constraints (limitations). Also, the objectives have critical meanings for desired goals. The third is to propose potential alternatives. It is critical for a decision making process to come up with valuable and high quality alternatives. All these alternatives should be validated to show they are feasible. Also, each alternative has its advantages and disadvantages. This procedure is the foundation of the next one. The fourth is to analyze, evaluate, and select the promising alternative. This procedure selects the optimized (promising) alternative among the ones by means of value analysis and feasibility study. The aforementioned four procedures is a process of recognition. The first and second procedures are to investigate, identify problems, and to establish objectives. This is the preparation phase for decision. This is the perceptual phase within the recognition process. The third and fourth procedures are to analyze information, design alternatives, and evaluate to select the promising one. This is the completion phase of decision. This is the rational phase within the recognition process. The fifth procedure is to execute the decision, and the final procedure is to track decision. The last two procedures can be treated as the decision execution and inspection phases or experimental phases.
Under the contemporary conditions, the situations for operation warfare are complicate and change from minute to minute. It will be hard to adapt objective need, if only human brain is utilized. For example, tactical fire system develops decision based on the situations, including discovery of targets, identification of enemies and friends, selection of course of actions, tracking of targets, fire engagement, transfer of fires, etc. All these actions need to be completed within the shortest period. Any delay might lose the opportunity to dominate the battlefield. Therefore, upon the condition of using high technological weapon equipments, rapid response speed is expected, only relying on man power is very difficult to collect and process intelligence data in time, and to propose and select course of actions. With the development of intelligent electronic computer, real “computer staff”, “computer assistants” will be created, and this provides a sufficient and beneficial condition to deal with all these difficulties.
The term “methods of decision based upon co-existence of human beings as well as machines” means the use of battlefield management automatic command systems that commanders and decision makers utilize artificial intelligence to reach decision automation. The characteristics of using this method are that neither human brains, nor computers are used separately for decision making. Both human being and intellectual machines (computer hardware and software application systems) collaboratively, mutually, and automatically complete the decision making process. Utilizing this method both continues to benefit from human being’s initiative, enthusiasm, and creativity, and suppresses human beings disadvantages including low response and calculation capability, experience of fatigue after long working hours, and influence from psychological movement, and limited information storage capabilities. By means of the integration of human beings and information equipments, military activities related to any complicated and fast changed situations in the battlefield are tackled.
The development of C4ISR systems mainly depends upon operation architecture, technology architecture, and system architecture, as well as the concept of network centric warfare that integrates systems from command, control, communications, computers, intelligence, surveillance, and reconnaissance in both horizontal and vertical directions. War fighters and decision makers although have completed many architectures to regulate the need for information technology, system functional descriptions and problem solving approaches, the architectures are described with different definitions on terminologies and formats, and thus used with different procedures which result in the production of products that are unable to integrate and compare with. This research is to utilize modern management science method to establish a developing model to supplement original architectural operational method that has the problem that systems developed do not link with their mapping operation requirements. With the support of this research, it is aimed to develop a battlefield management decision support system that meets the desired operational requirements.

並列關鍵字

Management Science Methods ； System Analysis ； Programmed Decision Method ； Function System Chart ； Scientific Decision Method ； DDS decision support system ； Structured Programming

參考文獻

21. 許士軍，管理辭典，2003年3月初版，華泰文化事業股份有限公司，pp228~231

4. William E Odom，Fixing Intelligence-For a More Secure America，2003，Yale University Press.

5. Sarkesian S.C、Williams J.A ,and Cimbala S.J，U.S. National Security-Policymakers, and Politics，2002，Lynne Rienner Publishers.

6. Christopher Coker，Waging War Without Warriors，2002，Lynne Rienner Publishers.

8. Steven Metz，AMERICAN STRATEGY：Issues And Alternatives for the Quadrennial Defense Review，2000 September，U.S. Army War College ,Strategic Studies Institute.