隨著全球科技快速發展，IC載板作為半導體產業鏈的重要環節，而智慧製造技術的引入，為IC載板產業提供了一條有效的升級轉型路徑。透過數據驅動的生產控制、AI製程優化和設備連線，可顯著提升生產效率和產品品質，並應對技術密度提升、成本壓力增加與激烈市場競爭的挑戰。本研究以個案公司為例，利用SWOT分析與產業五力模型，探討該公司在技術優勢與資源劣勢下的策略選擇。全面導入工業4.0雖然可實現智慧製造目標，但其高昂的資金投入與技術門檻對資源有限的中小企業構成挑戰。為此，該公司採用工業3.5作為過渡方案，通過分階段技術升級，在短期內有效益，同時為全面智慧化轉型奠定基礎。
研究結果顯示，工業3.5策略以漸進方式實現技術升級，重點涵蓋設備控制改善、機器學習導入製程優化及數據驅動的決策模式建構。針對關鍵製程的改進，工業3.5顯著降低不良率、提升良率並縮短試產周期。同時，與工業4.0相比，工業3.5策略在轉型成本、實施速度與風險管理方面更具優勢，適合當前資源有限且面臨市場壓力的企業需求。在實施層面，本研究提出三大改善方向：設備控制優化、機器學習應用及智慧製造整體改善，並透過前後比較，驗證其對生產效率與產品品質的顯著提升。此外，工業3.5的靈活性與低風險特性，不僅為企業節省資本支出，還建立了向工業4.0邁進的技術基礎。資源有限的企業不必追求全面的工業4.0轉型。工業3.5策略提供了一種低風險、高回報的技術升級方式，可快速應對市場需求，滿足客戶對品質與效率的要求，同時預留全面智慧化升級的發展空間，這對於資源有限但追求長期競爭優勢的企業具有重要參考價值。

This study focuses on the IC substrate industry, addressing the challenges and strategies of implementing smart manufacturing, particularly the application and outcomes of the Industry 3.5 model in early-established factories. Based on the case company’s implementation experience, a step-by-step technical upgrade strategy was adopted, starting with Golden Line pilot testing. Through equipment upgrades, data integration, and basic data analysis, the initial benefits of smart manufacturing were validated. The results showed significant improvements in equipment compatibility and health management, with downtime reduced by 30%, pilot yield rates increasing from 30% to 84%, and substantial reductions in pilot production costs and lead times. Additionally, the low-cost gradual upgrade approach effectively mitigated implementation risks and capital pressure while enhancing capital utilization efficiency. The study concludes that the Industry 3.5 strategy provides an efficient and low-risk pathway for resource-constrained enterprises to transition to smart manufacturing, enabling them to quickly meet market demands and enhance competitiveness.

一、中文文獻
翁胤竣(2022). 新產品上市之商業模式策略研究–以高壓氧艙為例. 國立成功大學 高階管理碩士在職專班碩士論文。
邱緯賓(2018). 基於SWOT分析建立考量群體需求之執法和服務品質管理網路選擇方法. 國立成功大學資訊管理碩士論文。
陳鎮銘(2020). 在開放式運算盛行之環境下伺服器代工廠的品牌發展策略-以廣達為例. 國立中央大學高階主管企管碩士論文。
賽盈盈(2011). 台灣晶片封裝測試產業之五力分析-個案研究. 國立台灣科技大學管理學院MBA碩士論文。
辛世安(2009). 以五力分析探討──台灣軟性印刷電路板產業競爭力. 國立台灣科技大學企業管理碩士論文。
吳威毅(2006). 台灣半導體設備產業之全球競爭策略研究- 以X公司為例. 國立中山大學管理學院高階經營碩士學程在職專班碩士論文。
SEMI 技術白皮書. (2023). Industrial 4.0 technologies in high-end PCB production. SEMI.
SEMI 技術報告. (2023). Predictive maintenance and digital supply chain in high-tech manufacturing.
二、英文文獻
Chen, H., Liu, X., & Wu, Z. (2020). Application of smart manufacturing in semiconductor industry. Journal of Advanced Manufacturing Technology, 35(4), 245-260.
Electronics Industry Insights. (2024). Big data and AI in smart factories: Applications in high-end IC substrate market. Electronics Industry Insights.
Electronics Market Analysis. (2023). Industrial and automotive electronics: Challenges for PCB manufacturers. Electronics Market Analysis.
Electronics Market Analysis. (2023). Smart manufacturing for consumer electronics and communication devices. Electronics Market Analysis.
Electronics Market Analysis. (2023). Smart manufacturing for green products: Competitive strategies in the global market. Electronics Market Analysis.
Environmental Science & Technology. (2023). Sustainable development in PCB manufacturing: Water treatment and resource optimization. Environmental Science & Technology.
Li, Y., & Zhang, Q. (2019). Cost reduction and quality enhancement through smart manufacturing. Journal of Industrial Engineering and Management, 12(3), 315-330.
Liu, M., Tang, Y., & Chen, F. (2021). Innovation and development in semiconductor manufacturing through smart technology. Technology and Innovation Management, 28(6), 311-326.
SEMI Technical Report. (2023). High-Density Interconnect (HDI) and its role in IC substrate advancements. SEMI Technical Report.
SEMI Technical Report. (2023). Green manufacturing practices in PCB industry. SEMI Technical Report.
Wang, L. (2021). Flexibility and efficiency in semiconductor manufacturing. International Journal of Manufacturing Systems, 42(2), 105-120.
Zhou, J. (2022). Global supply chain management in the semiconductor industry: The role of smart manufacturing. Supply Chain Management Review, 18(1), 45-60.
三、網路文獻
Adams, H. S. (2023). Embracing factories of the future at Hyundai. Manufacturing Digital. https://manufacturingdigital.com/production-operations/embracing-factories-of-the-future-at-hyundai
DIGITIMES Asia. https://www.digitimes.com
Electronics Weekly. https://www.electronicsweekly.com
Friesner, T. (2008). History of SWOT analysis. MarketingTeacher. https://www.marketingteacher.com/history-of-swot-analysis/#google_vignette
Kashyapa, R. (2020). Integrating machine vision AI with Toyota Production System. Qualitas Technologies. https://qualitastech.com/quality-control-insights/integrating-machine-vision-ai-with-toyota-production-system/
Morrison, M. (2012). History of the SWOT analysis. RapidBI. https://rapidbi.com/history-of-the-swot-analysis/
Morrison, M. (2012). History of the SWOT analysis. RapidBI.https://rapidbi.com/history-of-the-swot-analysis-3/
Samsung. (2017). Driving manufacturing productivity through the Industrial Internet of Things (IIoT). Samsung Newsroom. https://news.samsung.com/global/driving-manufacturing-productivity-through-the-industrial-internet-of-things-iiot
Samsung. (2023). Samsung Electronics unveils foundry vision in the AI era at Samsung Foundry Forum 2023. Samsung Electronics.https://semiconductor.samsung.com/news-events/news/samsung-electronics-unveils-foundry-vision-in-the-ai-era-at-samsung-foundry-forum-2023/
Wilson, G. (2021). Hyundai & Singtel advance smart mobility and Industry 4.0. Manufacturing Digital. https://manufacturingdigital.com/smart-manufacturing/hyundaisingtel-advance-smart-mobility-and-industry-40