由於機架式伺服器會因結構問題，衍生系統變形下陷而機殼互相干涉，可能會致短路當機。本文以一套創新改善流程，對此伺服器機殼進行創新結構改善。此流程係結合並導入群組化概念的QFD理論及改良後的TRIZ判讀法。相較於傳統TRIZ，改良後的TRIZ理論，可更快找出適用的發明法則，較不會讓設計構思及矛盾而造成時間的浪費。
首先，將欲改善之伺服器利用已群組化的QFD，配合TRIZ的參數分群法進行品質機能展開後，得到權重最高的品質工程參數為物理群組內的-「強度」。再以改良的 TRIZ 理論，進行欲改善參數「強度」之判讀。判讀後得到之發明法則為「事先緩衝」；和不欲惡化參數為「製造性」，將此兩項導入伺服器機殼結構以進行實際改善。之後，建立3D繪圖模型，將「事先緩衝」法則轉換為設計改善參數，對於局部結構下陷之處，找出可利用之位置，並於此處增加補強肋補強；同時考慮不欲惡化參數-「製造性」，所以在設計固定補強零件時將鎖附螺絲改為鉚合方式，可節省組裝時間及加工費用。第一次改善完成後，利用有限元素軟體進行實際模擬，根據數據結果顯示局部下陷已減少但尚未達到安全值。此時，再利用上述相同判讀法則針對結構較弱之處，再增加補強零件，由於考慮裝置複雜性，所以選擇能輕易組裝且易製造的支架補強。並再次以軟體模擬，根據數據結果顯示變形下陷值已在安全範圍1.35mm內，驗證由此方式改善下陷性不會產生機殼互相干涉及減少短路當機的機率。
最後利用專利檢索，搜尋業界相關產品之結構，比較本文以創新技術所改善之結構。篩選彙整相同領域之專利後，並無發現相同結構。所以由此可證實本文之創新結構具有新穎性；又因改善下陷結構使機殼不會產生互相干涉及減少短路當機的機率，故也較其他工具更具進步性及產業利用性。
最後，希望能藉此創新程序，讓研發之人員在設計時能正確地定義問題，和掌握改善技術；且能在短時間內能獲得創新之設計構想。也期望藉此系統化創新程序所提出的各階段設計工作流程，提供設計新手一個可遵循的流程及發展專利迴避的新方向。

Because of its structure, the unit of the rack-mount server might deform and sag, making the chassis interferes with each other and causing the system down or short circuit. This paper uses a set of innovative improvement process to improve the structure of the chassis of such server innovatively. Such process combines and introduces QFD theory having clustered concept with the improved TRIZ diagnostic method. As compared with traditional TRIZ, the improved TRIZ theory can find applicable invention principle more quickly and not wasting time in solving the contradiction and discrepancy of the design idea.
First of all, upon using the clustered QFD with TRIZ parameter clustering method to undergo quality function deployment towards the server for improvement, the quality engineering parameters having the heaviest weight is regarded as the “strength” of the physical cluster.
Secondly, use the improved TRIZ theory to undergo diagnosis towards the “strength” of the parameters for improvement. The acquired invention principle is the “Cushion in advance” and the parameter don’t want to worsen is the “design for manufacture.” These two items are then introduced to the structure of the server chassis to undergo practical improvement. Afterwards, 3D illustration model is built, making the “Cushion in advance” principle converts into design improvement parameters, so as to find and strengthen the usable location on the sagged area of the structure. At the same time, the parameter don’t want to worsen, that is, the “design for manufacture,” is considered, so that when designing the strengthening of fixed area, the lock with screw method is improved into that of riveting, saving the time and cost for assembly. Upon the first time improvement, finite element software is used to undergo practical simulation. Based on the result of the data, phenomenon of partial sag is improved but still not reaching the safety level. Then, the above diagnostic method is used upon the area having a weaker structure to strengthen the components. Considering the complexity of the device, a more simple assembly is used with the strengthening of racks that are easier to manufacture. Software simulation is again executed and based on the result of the data, the deform and sag values stay within the safety range of 1.35mm, verifying that this method can improve the phenomenon of sag, would not cause interference between unit chassis and reduce the possibility of short circuit and system down.
Finally, patent search is used to search for the structure of the related products in the industry, so as to compare them with the improved structure proposed herein by innovative technology. Upon filtering and compiling the patents of the related scopes, no similar structure has been found, proving that the innovative structure herein carries creativity. Also, since the unit chassis would not interfere with each other because of the improvement of the sagged structure, it is more advance and usable than any other tool in the industry.
Lastly, it is expected that by means of this innovative procedure, the research and development team can accurately define problems and possess improvement technology during their design process, and acquire innovative design idea within short period of time. It is also expected that the design workflow of every stage provided by this systematic innovative procedure can give new designer a process to follow, as well as a new direction for the development of design around.

[1] Altshuller, G., “To Find An Idea”, http://www.altshuller.ru/world/eng
[2] Lerner, L., “Genrich Altshuller: Father of TRIZ” Russian Magazine Ogonek, http://www.aitriz.org (1991)
[3] D. Mann and E. Domb, 40 Inventive (Business) Principles with Examples,The TRIZ Journal, Sep. (1999)
[4] Saliminamin, M. and Nezafati, N. “A New Method for Creating the Non-Technologic Principles of TRIZ” The TRIZ Journal October. (2003)
[5] Clarke, D. W., TRIZ: Through the Eyes of an American TRIZ Specialist, Ideation International Inc., Southfield, MI.(1997)
[6] J. Terninko, The QFD, TRIZ and Taguchi Connection : Customer– Driven Robust Innovation,” The Ninth Symposium on Quality Funcion Deployment,(1997)
[7] N. León-Rovira and H. Aguayo, “A new Model of the Conceptual Design Process using QFD/FA/TRIZ,” The TRIZ Journal, http://www.triz-journal.com/, July (1998).
[8] P. R. Apte and D. L. Mann, “Taguchi and TRIZ: Comparisons and Opportunities,” The TRIZ Journal, http://www.triz-journal.com/, November 2001.
[9] Ruchti, B. & Livotov, P..“ TRIZ0 based Innovation Peincipes and Process for Problem Solving In bussiess and Management”, The TRIZ Journal, http://www.triz-journal.com.,2001.
[10] 赤尾洋二，新產品開發－品質機能展開之實際應用，中國生產力中心QFD 研發小組編譯，中國生產力中心，台北，1991。
[11] SSI“ Thin Electronics Bay Specification-A Server System Infrastructure (SSI) Specification for Rack-Optimized Servers ” http://ssiforum.oaktree.com/ViewUserDocuments.aspx
[12] SSI“ Rack Server Specification- A Server System Infrastructure (SSI) Specification for Computer Server Cabinet Enclosures & Racks ”
http://ssiforum.oaktree.com/ViewUserDocuments.aspx
[13] 水野滋；赤尾洋二，「品質機能展開」，前程企業管理公司，中華民國76年6月。
[14] 赤尾洋二著，「新產品開發-品質機能展開之實際應用」，中國生產力中心，1991年.
[15] A Jebb and HP Wynn, Desgin function deployment—a design system for the future，No.16，Desgin Studies，pp.447-470(1995).
[16] 水野滋，「品質機能展開法-如何有效掌握顧客需求」，先鋒企業管理發展中心，中華民中壢市，1987年。
[17] Hauser, J. R. and Clausing, D., The House of Quality, Harvard Business Review, May-June, pp.63-73(1988).
[18] Chakraborty S. and Dey S.,QFD-based expert system for non-traditional machining processes selection. Expert Systems with Applications, vol. 32, pp.1208-1217 (2007).
[19] 冷耀世，專利實務論，全華科技圖書， 9月，西元2005年
[20] 專利通行網，http://www.patent104.idv.tw/。
[21] 清華大學圖書館專利資料庫服務，網站http://www.lib.nthu.edu.tw/services/ref/online/pate
[22] 賴榮哲，專利分析總論，翰蘆圖書出版有限公司，中華民國九十一年八月。
[23] 雲林科技大學管理學程，http://www.team.yuntech.edu.tw。
[23] 陳瑞田，「創新性之專利迴避設計」，經濟部智慧財產局，西元2006年4月。
[24] Shulyak, L., Introduction to TRIZ. 40 Principles: TRIZ Keys to Technical Innovation, By Genrich Altshuller; With new material by Lev Shulyak; Drawings by Uri Fedoseev, Vol. 1, P15-21(2002).
[25]莊錦賜，「TRIZ」創意技法，http://www.ccda.org.tw/。
[26]馮展華，「機械工程之創意工程設計方法」， http://sparc.nhit.edu.tw/~lochsieh/。
[27] Lerner, L.,“Genrich Altshuller: Father of TRIZ,” Russian MagazineOgonek, http://www.aitriz.org(1991).
[28] Clarke, D.W., TRIZ: Through the Eyes of an American TRIZ Specialist,Ideation International Inc., Southfield, MI(1997).
[29] Domb, E.,“The 39 Features of Altshuller Contradiction Matrix,” TheTRIZ Journal(1998).
[30] Terninko, J., A. Zusman and B. Zlotin, Systematic Innovation－AnIntroduction to TRIZ, St. Lucie Press, New York(1998).
[31] Altshuller, G.,“40 Principles: TRIZ Keys to Tecchnical Innvation,”Technical Innovation Center(1998).