以DANP、PLS-SEM與ANFIS探討產品模組化與製程模組化於被動元件大量客製化之影響
No Thumbnail Available
Date
2020
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
科技的進步提昇了生活水平，加上5G設備及人工智慧物聯網(Artificial Intelligence of Things,AIOT)的開發，造成被動元件需求增加。而被動元件廠商同時切入大量客製化產品及流程模組化的利基型元件市場，如近來電動車和綠能產業的應用領域興起，帶動特殊規格被動元件之需求。雖然大量客製化是廠商近年來努力的方向，但少有研究探討影響導入大量客製化及模組化之關鍵成功要素，因此透過決策實驗分析法(Decision Making Trial and Evaluation Laboratory, DEMATEL )來確認各構面的影響關係，再結合DANP (DEMATEL-based Analytic Network Process, DANP ) 來求各構面影響其他構面的程度之重要性及以偏最小平方結構方程模型(Partial Least Squares Structural Equation Modeling, PLS-SEM )檢定前述DEMATEL推論影響關係的假設顯著，以定義在大規模客製能力及產品模組化和流程模組化問題結構中，可做為關鍵要素的客製化知識利用及商業流程改善的重要性。並以此構面做為研究方向，建立自適應性網路模糊推論系統(Adaptive Network-Based Fuzzy Infer-ence System , ANFIS )預測模型，推衍影響製程化與大量客製化之關鍵要素與決策規則，並與 DANP 之結果比較。本研究將邀集位於台灣，全球被動元件領導廠商之專業經理人擔任專家，並由前述領導廠商之員工隨機取樣發放問卷。本實證研究之結果發現，實現大規模客製化能力的最重要因素是公司必須將這一關鍵問題放在首位，而客戶知識的利用，也是最關鍵的因素。
In recent years, living standards have improved thanks to technological advances, coupled with the development of 5G equipment and artificial intelligence of things (AIoT). This has led to an increased demand for passive components. Meanwhile, passive component manufacturers have entered the market of mass customization products and modular niche components. For example, the recent emergence of electric vehicles and green energy industries has increased the demand for passive components with the corresponding specifications. Although advanced manufacturers have already for a few years been focusing on mass customization and process modularization of passive components, there are few studies on how to best introduce these strategies. Therefore, we have used the decision-making trial and evaluation laboratory (DEMATEL)to study these processes in detail. Subsequently, DANP (DEMATEL-based analytic network process) was introduced to study the importance of each aspect of the process. After that, the partial least squares structural equation modeling (PLS-SEM) was used to verify the aforementioned assumptions of DEMATEL’s influence relation-ship. These aspects can be used to define the importance of factors and improve business processes in terms of mass customization capabilities, product modularization, and process modularization problem structures. Furthermore, an adaptive network-based fuzzy inference system (ANFIS) prediction model was used to identify the most important factors and decision rules that can affect process modularity and mass customization. The results were compared with the results obtained using DANP and PLS-SEM. Professional managers from globally leading passive component manufacturers in Taiwan were invited as experts, and the employees of the manufacturers were randomly sampled and surveyed using the PLS-SEM and ANFIS questionnaires. The study found that the most important factor for implementing mass customization is accumulating the mass customization and organization coordination. The use of customer knowledge is also a critical factor. The analysis results can be used as guidelines for passive component manufacturers. The analytical framework can also be adopted by other technology companies involved in mass customization or process modulation.
In recent years, living standards have improved thanks to technological advances, coupled with the development of 5G equipment and artificial intelligence of things (AIoT). This has led to an increased demand for passive components. Meanwhile, passive component manufacturers have entered the market of mass customization products and modular niche components. For example, the recent emergence of electric vehicles and green energy industries has increased the demand for passive components with the corresponding specifications. Although advanced manufacturers have already for a few years been focusing on mass customization and process modularization of passive components, there are few studies on how to best introduce these strategies. Therefore, we have used the decision-making trial and evaluation laboratory (DEMATEL)to study these processes in detail. Subsequently, DANP (DEMATEL-based analytic network process) was introduced to study the importance of each aspect of the process. After that, the partial least squares structural equation modeling (PLS-SEM) was used to verify the aforementioned assumptions of DEMATEL’s influence relation-ship. These aspects can be used to define the importance of factors and improve business processes in terms of mass customization capabilities, product modularization, and process modularization problem structures. Furthermore, an adaptive network-based fuzzy inference system (ANFIS) prediction model was used to identify the most important factors and decision rules that can affect process modularity and mass customization. The results were compared with the results obtained using DANP and PLS-SEM. Professional managers from globally leading passive component manufacturers in Taiwan were invited as experts, and the employees of the manufacturers were randomly sampled and surveyed using the PLS-SEM and ANFIS questionnaires. The study found that the most important factor for implementing mass customization is accumulating the mass customization and organization coordination. The use of customer knowledge is also a critical factor. The analysis results can be used as guidelines for passive component manufacturers. The analytical framework can also be adopted by other technology companies involved in mass customization or process modulation.
Description
Keywords
大規模客製, 產品模組化, 流程模組化, 客製化知識利用, 商業流程改善, Mass Customization Capability, Product Modularity, Process Modularity, Customized Knowledge Utilization, Business Process Improvement