全球創新系統之動態網絡探討-以平面顯示器技術為例

本研究分別探討環境對產業網絡演化及技術擴散的影響，並且同時探索互動效果對平面顯示器產業創新能力的影響力。過去研究者探討產業發展時，因為侷限於資料的完整性，所以普遍進行大量的訪談來進行質性研究；而本研究不同於以往的產業研究，利用專利引證網絡分析對產業的發展提供一個不同角度的詮釋。另外，以往的網絡分析大多著眼於雙方關係的討論，本研究利用專利引證網絡進行全面性的分析，也給了傳統網絡分析不同的思維。本研究以Datastream資料庫收集到電子相關產業的平均股價來衡量環境的資源豐裕度(munificence)，以及美國專利商標局專利引證資料計算出網絡中心勢(centralization)指標來衡量環境的不確定性(uncertainty)。在網絡演化方面，本研究利用關鍵字收集了美國專利局(United States Patent and Trademark Office, USPTO)1976到2008年相關專利引證資料來探討產業網絡的形變，並且利用此一引證資料來探討社會網絡的兩種不同概念的互動效果-凝聚力(Cohesion)與結構同位(Structural Equivalence)模型對於產業創新能力的影響。研究結果發現網絡的連結增加(Tie Creation)與連結減少(Tie Deletion)的現象不如Koka, et al.(2006)所預期的發展方向，連結增加(Tie Creation)與連結減少(Tie Deletion)在三個產業環境區間全都是上升的傾向。另外，兩種擴散模式皆對於國家的產業創新能力有正向的影響，並且發現凝聚力的效果較結構同位來得明顯以及產業初期，政府的政策具有相當重要的影響力。本研究認為台灣在平面顯示器產業若要加強其創新能力，則要與南韓、日本相互合作，將彼此視為合作夥伴而非競爭對手，才能增加台灣在平面顯示器產業的研發能力。

關鍵字

專利分析(patent analysis) ；社會網絡分析(social network analysis) ；創新能力(Innovation capacity) ；平面顯示器(FPD) ；環境(environment)

並列摘要

This study discusses the effect which relate to network evolution and interactive effect are influenced by environment; in the meantime, we also discover how interactive effect influence the innovation ability of flat panel display industry. In general, scholar prefer adopt interview to explore industry development due to scarce industry data; in contrast with past way, we take advantage of patent citation network analysis to explain industry development which differ from past study. In addition, traditional network analysis always focus on bilateral relation, but this study uses patent citation network analysis to supply a complete discussion, and also provide traditional network analysis with a diverse sight. We evaluate munificence via average stock price concerning electricity industry from DataStream Database and uncertainty through centralization from USPTO’s patent citation data. In network evolution field, we take advantage of industry keyword to collect patent citation data from 1976 to 2008 which discuss how two interactive effects, Cohesion and Structural Equivalence Model, affect industry innovation capacity. We find phenomenon, tie creation and tie deletion’s direction, are not only like what Koka, et al.(2006) expected, but all raised in three environment distinct. Furthermore, both cohesion and structural equivalence interactive effect positively influence industry innovation capacity as well as cohesion model’s effect have more significance determination to industry innovation capability than structural equivalence. What’s more, we find that when industry is in initial stage, innovation capacity is up to government’s policy which intends to increase collaboration aroma. We believe that Taiwan should not see Korea or Japan as competitor, but partner before raise flat panel display industry’s innovation capacity.