- 學位論文
從數字到決策 - 數位年代的反向思考
How to Make a Data-Driven Decision Thinking Backward
摘要
我們正進入人工智能（AI）主導的時代，數據是AI的最基本的材料，如何善用數據將成為成功的關鍵因素。 本論文從英特爾和亞馬遜的營運中，深入探討公司營運決策，數據驅動決策(Data-Driven Decision-Macking, DDDM)在兩間公司皆扮演營運中變革的力量。英特爾透過高階的數據處理和分析方法識別並修正製程中缺陷，利用DDDM大幅提高其半導體製造的產量。亞馬遜在其物流作業中從歷史數據中擬定策略，在顧客滿意度與營運成本中取得巧妙平衡，使物流產業中最難解的最後一哩路也可以被解決。 本論文在研究中倡導的「逆向思維」方法，是一種從目標出發，反向推演過程的決策方法。 這種方法與傳統的、線性的逐步推進的解決問題方式有顯著不同，它要求從最終目標開始思考，反向規劃出實現這一目標所需的步驟和資源。 在數據驅動決策（DDDM）的背景下，我們認為逆向思維更有價值，因為它能幫助企業更有效定位目標、資料收集和使用數據，從而優化決策過程和業務結果，更在動態商業環境中讓公司提高適應能力。從英特爾和亞馬遜的營運中，我們也可以看到，他們善用逆向思維方法，更可以解決複雜挑戰，不僅提出最佳的策略決策，還不斷地進一步改善策略，從而持續保持競爭優勢。 最後，本論文全面概述了提倡的DDDM，它提供了系統分析和策略數據的框架，對於動態決策和營運效率皆有顯著改進，更完整DDDM架構，幫助公司利用數據在數位年代中開創新的契機。
並列摘要
In an era dominated by digital advancements, the ability to leverage data has become a crucial element for organizational success. This thesis explores the power of Data-Driven Decision-Making (DDDM) in the operations of two leading industry giants, Intel and Amazon. We present practical practices of how systematic analysis and strategic data utilization can lead to insightful decision-making and substantial improvements in operational efficiencies. Both companies integrate extensive data analysis to refine operational processes and enhance strategic decisions, facilitating significant business growth and maintaining competitive advantage. Through meticulous examination of detailed case studies at Intel and Amazon, this research investigates how these corporations employ data analytics to tackle complex challenges, optimize processes, and navigate the competitive landscapes of the semiconductor and e-commerce sectors, respectively. Intel harnesses DDDM to significantly improve its semiconductor manufacturing yields by identifying and rectifying defect patterns through sophisticated data handling and analysis methodologies. Amazon integrates advanced data-driven strategies within its logistical operations to optimize last-mile delivery, one of the crucial elements to balance customer satisfaction and operational cost. The structured "Thinking Backward" approach advocated in these studies outlines the strategic alignment of actions based on empirical data to achieve specific business goals, thereby reversing the traditional, forward-thinking decision-making methods. This shift enhances the clarity and efficiency of operational processes and boosts adaptability in dynamic business environments. We propose an improved framework for organizations that capitalize on the digital revolution to bolster their strategic and operational frameworks.