針對TTQS新舊兩種版本搜尋企業關鍵訓練項目組合最佳化之研究

Searching Optimal Training Items of Enterprise Based on Talent Quality-Management System

林佳真(Chia-Chen Lin)

指導教授 ： 陳民枝

共同指導教授 ： 陳同孝(Tung-Shou Chen)

國立臺中科技大學/資訊與流通學院/資訊工程系碩士班/碩士(2016年)

摘要

在這企業競爭激烈的年代，經營者為讓企業永續經營與獲利，隨著技術不斷的進步。如何透過企業內部職場教育訓練，提升人員專業知識與技能為一大挑戰。故行政院勞動力發展署為協助企業提升人力素質，輔導企業導入(Talent Quality-management System，TTQS)管理制度，藉由規劃完善的職場教育訓練，達到強化企業營運模式與提升競爭力的目的。基於此原因，建置人才發展品質管理系統TTQS，以計畫、設計、執行、查核、成果五大管理作業循環流程，作為企業規劃辦訓的依據及檢視辦訓品質之辨識工具，以達到培育人才及提升企業競爭力的組織目標。考量在人力與資源投入有限的情況下，要企業短時間內就落實TTQS各個評核項目的要求，在實際執行上實屬困難。因此，本論文提出以企業在2011年21項及2012年19項評核項目新舊版TTQS評核成績，結合基因演算法來找出關鍵訓練項目組合成效。本論文將評核成績依評核等級分5種類別，作為基因演算法分類的依據，設定參數並將LIBSVM設定為本研究分類器以找出適應值，經由運算過程找出關鍵訓練項目組合演化過程。本論文以2011年為例，從原本的21項評核項目中，粹取出其中的6個評核項目作為關鍵訓練評核項目，與原來21項評核項目相比其準確率達91.56%。在人力投入與時間有限的條件下，可快速找出關鍵訓練項目組合，協助企業在評估找出影響辦訓成效的關鍵訓練項目組合時之依據。將實驗結果運用在企業辦訓上，可快速提升辦訓績效與減少投資浪費，進而達到提升企業競爭力的目標。未來的研究可將評核機制應用在學術單位提升教育品質。

關鍵字

TTQS ； 基因演算法 ； LIBSVM ； 評核項目 ； 教育訓練

並列摘要

In this competitive era, entrepreneursare continuously making progress in order to stay in the competition and be profitable. It’s a big challenge for enterprise to try to upgrade professional knowledge and skill through in house training. Thereby, the minister of labor has developed a training protocol for counseling enterprise to adopt the Talent Quality-management System (TTQS) for upgrading human resource. The goal is to upgrade competitiveness and to strengthen management via in house training and educating. Based on these reasons, the five major management process cycle of TTQS, namely, planning, design, execution, evaluation and results, is used by the enterprise as identification tools for training and quality in order to meet the goals of nurturing talents and to increase competitiveness. For the enterpriseto fulfillthe required assessment elements of TTQS within a short period of time is simply difficult in lieu of the lack of human resources. Therefore, this thesis proposed a TTQS assessment based on the 21 assessment in 2011 and the 19 assessment elements in 2012, and combined with genetic algorithm to find critical effective training elements combinations. In this thesis the assessment result for assessment grade is divided into five types which are used in the genetic algorithm as classification basis. The parameters setting are used in LIBSVM as classification for finding the fitness value. Based on mathematical calculations, the critical combined training elements evolution are located. Experiments in this thesis is based on the 21 assessment elements of year 2011 where six assessment elements were used as critical training elements. Compared to the original 21 elements, its accuracy is significant at 91.56%. Under the limited human resources and time frame, the proposed assessmentcan quickly determine the critical training elements combination which can be used to assist the enterprise to find the basis under which the critical combined training elementsis effective. The results of the experiments maybe used by the enterprise to increase competitiveness throughin house training which can quickly increase effective training and to reduce resource wastages. Future research may apply the assessment program to the educational institutions to enhance academic quality.

並列關鍵字

TTQS ； Genetic algorithm ； LIBSVM ； assessment elements ； training

參考文獻

[28] 彭至賢，"以TTQS評核項目尋找影響不同產業關鍵績效指標之研究"，國立臺中科技大學資訊工程系碩士論文，2013。

[1] T. Alberga, S. Tyson and D. Parsons, “An Evaluation of the Investors in People Standard,” Human Resource Management Journal, 2007, vol. 7, no. 2, pp. 47-60.

[2] C. C. Chang and C. J. Lin, LIBSVM A Library for Support Vector Machines. http://www.csie.ntu.edu.tw/~cjlin/libsvm/, 2009.

[3] W. L. Chang and S. T. Chen, “The Performance of Taiwan's Training Quality Excellence System,” Total Quality Management & Business Excellence, 2013, vol. 24, no. 5-6, pp. 561-576.

[4] R. C. Chen, J. Chen, T. S. Chen, C. H. Hsieh and K. Y. Wu, “Building An Intrusion Detection System Based on Support Vector Machine and Genetic Algorithm,” Lecture Notes in Computer Science, 2005, vol.3498,pp. 409-414.