分析層級程序法（Analytic Hierarchy Process, AHP）及其一般化的分析網路程序法（Analytic Network Process, ANP）可協助我們了解真實世界並用來解決問題，受到廣泛的重視，並為多準則決策（Multiple Criteria Decision Making, MCDM）的重要一支。本研究提出兩種模型建構流程改善方式，並分別應用於綠色供應鏈策略選擇與穿戴式科技接受程度之分析。 首先，係以產品生命週期管理（Product Lifecycle Management, PLM）概念建構綠色供應鏈策略選擇網路模型，同時強調控制準則之設定，使成對比較之影響評估結果更為精確，由模型推導而得之綠色供應鏈管理策略能更符合組織之環保觀點。接著再以某家台灣知名電子公司為例，解釋該模型之分析應用。因PLM管理概念與組織企業功能相近，使得該模型之分析結果更容易應用於各部門之政策、計畫訂定。 此外，本研究為了改善ANP模型建構之客觀性，嘗試使用科技接受與使用整合模型（Unified Theory of Adoption and Use of Technology, UTAUT）作為分析模型建構之基礎，該模式廣泛應用於新興科技接受程度之評估且解釋能力較高。接著，再以最近受到各界關注的穿戴式科技為探討對象，評估消費者對於此一新興科技之可能使用意圖及真正使用情況。過去研究多運用統計方法進行UTAUT模型分析，但改用ANP方法能避免一些使用限制，如需符合使用統計方法之基本假設及條件，需取得一定適量之樣本數等。又模型中各變數之顯著性差異分析會因每次蒐集樣本不同而有不同結果，進而無法瞭解所有變數之重要性。 本研究提出兩種之建構方式皆利用敏感度分析確認模型之穩定性，但以管理概念為基礎之建構方式，未來可思考結合其他數量方法，持續改善因素之間關係之判斷；而以理論模型為基礎之建構方式，可持續探討其他理論模型之轉換可能性，進而提出一套更為客觀且合理之模型轉換流程。 Abstract: The Analytic Hierarchy Process (AHP) and its general form, Analytic Network Process (ANP), help decision makers understand real world problems and solve them via a hierarchical or network structure. It has been popular for many business applications and has become an important technique in multiple criteria decision making. This study proposes two kinds of improvement processes of model-structuring and applies them to the selection of the appropriate green supply chain management (GSCM) strategy and the evaluation of consumers’ acceptance of wearable technology. First, the researcher organized the GSCM strategy-selection network structure according to the product lifecycle management (PLM) concept. Then the importance of control criteria which could make the results of pair-comparisons more precise was stressed and the GSCM strategy which conformed to the environmental perspective kept by the organization was chosen. Then the study selected a leading Taiwanese electronics company to demonstrate how to choose an appropriate GSCM strategy via the proposed model. The analytical results derived from the model can provide suggestions for departments’ policies or programs since the concept of PLM corresponds with their business functions/departments types. In addition, the Unified Theory of Adoption and Use of Technology (UTAUT) model which is a more robust theory model and used commonly to predict the acceptance of a new technology was utilized to be a basis for an analytical network-framing. Then, wearable technology was taken an example to evaluate customers’ intention to use the products and the level-low, medium and high of actual use of wearable technology in Taiwan. This approach not only improved the model-framing objectivity; but also avoided some limitations which existed in statistical methods including some assumptions, problems with sample sizes and so on. The method used in this study allowed for issues related to inaccurate statistical significance magnitude resulted from different samples which keep users from understanding the importance of each variable to be avoided. The stability of the above tow model-structuring approaches was confirmed via sensitivity analysis; however, the first approach which is based on managerial concepts should combine other mathematical techniques to show better judgments in relationship–linking between each factor. The other approach which is based on UTAUT model also continuously examined the feasibility of other theoretical models, and then advanced to a more objective and feasible transferring process.
