可變空間下的決策、習慣領域及認同圈的量化

Abstract

可變空間下的挑戰性問題存在許多複雜決策參數的特性，這些參數（例如準則、方案…等）可能隨著時間和環境的變動而改變。這些重要的參數也許是得到有效解答的關鍵，但潛藏在潛在領域的深處。在這快速變動的世界（包括科技和看法…），如果沒有注意到在可變空間下的問題，我們可能由於決策陷阱、決策盲點及決策震驚而容易導致嚴重的錯誤決策。本論文首先簡要敘述多目標決策的發展朝向可變空間下的挑戰性問題；接著闡述人類動態決策流程、習慣領域學說及認同圈，及其對尋找相關參數和決策的重要影響；接著介紹能力集合且說明在可變空間下的決策盲點、決策陷阱及決策震驚，並提供找出重要決策參數的方法及檢查表；接著提出一個系統性尋找重要決策參數的動態循環架構－「創新動態學」，該架構著重於潛在領域裡參數的動態變化；接著舉例說明習慣領域三大工具箱可幫助決策者深入潛在領域，擴展及豐盛決策者的習慣領域，讓決策者有系統的方法避免決策盲點和陷阱，更能有效地處理挑戰性問題 本論文用習慣領域的認同圈的概念，定義及探討認同函數、認同程度、認同矩陣、認同圈和不認同圈之量化與估算。並用兩個應用範例解說，人對人之間，針對某事件所形成的認同圈及其影響。範例一（見7.1.）說明如何善用認同程度和認同圈的概念，制定新的選舉規則，這規則的最大特點之一，是選民對候選人表達認同程度，不是只有黑白或零（不認同）一（認同），選民也可以表達不認同程度，在此新規則下，眾多的選舉形式可展現出不同的選舉結果，被選上的人也較能反應民意。範例二（見7.2.）說明如何善用認同矩陣，設定一個認同門檻，選出社群內最少或成本最低的成員組合（核心人物、意見領袖…），含蓋（影響、關照）所有關係人的數學規劃。當企業要進行新理念或產品推行時，若能先與這些核心人物進行理念的溝通或產品的試用，善用他們對社群組織成員的影響力，則企業可更能有效行銷新理念或新產品。

Challenging decision problems in changeable spaces are characterized by existence of complex decision parameters that are changing with time and situations, including criteria and alternatives. Some of these parameters may be critical for their effective solutions, but hidden in the depth of potential domains. In this rapid changing world, including technology and attitude, without paying attention to the problems in changeable spaces, we could easily commit serious mistakes due to decision blinds, decision traps and/or decision shocks. The dissertation starts with a brief description of the evolution of MCDM (Multiple Criteria Decision Making) toward challenging problems in changeable spaces. Then it briefly sketches a dynamic human behavior mechanism, identification spheres, and habitual domain theory which provide an effective list for us to search relevant decision parameters and pave the way for latter discussion. Competence set analysis, derived from habitual domain, is then introduced to exemplify decision blinds, decision traps and decision shocks in challenging decision problems. Checking lists and methods for discovering blinds and traps and for dealing with shocks are also provided. Innovation dynamics, a systematic network of thoughts, is introduced to further look out relevant key parameters in dynamic challenging problems. The related academic subjects in each link of the innovation dynamics are also explained, which allow us to see the complexity and interconnectivities among different challenging problems in changeable spaces. Three habitual domain tool boxes to empower ourselves to expand and enrich our thoughts into the depth of the potential domains of the challenging problems are introduced, which allows us to more effectively identify hidden parameters, problems and competence sets to reduce decision blinds, avoid decision traps and solve the problems, or dissolve the problems before they occurs. A quantitative approach to study the identification conception is proposed. Specifically, based on the definition of identification sphere of Habitual Domains Theory, we define and explain the concepts of identification function, degree of identification, identification matrices, identification spheres and dis-identification spheres. The quantitative concepts are then applied to formulate the mathematical models of election of a social group leader. The models consider both degree of identification and that of dis-identification. Different models lead to different results, which enrich our thinking about election. We also introduce mathematical programming models, utilizing identification matrices, and thresholds of identification, to find an optimal combination of key members to influence all targets in a social group. The results are important for marketing promotion of a new product, service or that of a new concept.