中小製造業永續發展評估模式之建立與分析

近年來，各企業為了維持市場之競爭力，已漸漸將永續發展視為企業營運之重要目標，並對此推動諸多策略，以求在全球化之競爭下達到永續經營之目的。為求企業永續性之經營，須將經濟、社會、環境等構面作為永續發展之三大主流。而此三大主流已成為企業永續性發展之關注焦點，並同時作為提升企業競爭力之重要指標。然而過去有關永續發展之研究中，大多以理論性進行探討，甚少有學者使用數量方式進行相關數據分析之研究。基於以上之原因，本研究發展出整合多屬性決策分析模式，以評估台灣中小製造業在永續發展之研究與分析。本研究運用之方法，首先以模糊德爾菲(Fuzzy Delphi Method, FDM)篩選永續發展三大構面之準則，進而分別運用理想解類似順序偏好法(Technique for Order Preference by Similarity to Ideal Solution, TOPSIS)、模糊積分(Fuzzy Integral, FI)及灰關聯分析(Grey Relational Analysis, GRA)等三種方法，進行績效評估，並且透過其績效之排序值予與分類，再透過約略集合理論(Rough Set Theory, RST)探究其核心屬性集合，最後進行敏感度分析比較在三種模式之共同點與不同點。本研究採用上述三種評估模式之動機，為(1)理想解類似順序偏好法針對每一個評估指標皆具有單調遞增或遞減之特徵，以建立整合多屬性決策分析模式；(2)模糊積分不需要假設評估構面間相互獨立，在量測方面只要符合單調性就可使用；(3)灰關聯分析可針對不明確與不完整資訊系統模型，進行關聯分析與模型建構的工作，以上三種模式各有其特點，本研究希望藉由與約略集合理論之應用，探究其產生結果之異同。本研究以台灣中小製造業為探討對象，分析中小製造業致力於推動永續發展時，應優先採取何項關鍵策略。經本研究分析之結果顯示，社會指標之「增加當地社區就業機會」、經濟指標之「準時交貨」以及經濟指標之「利潤」為核心屬性；環境指標之「減少有害原物料的使用」、社會指標之「就業實踐」以及社會指標之「紀律管理的有效性」為非核心屬性，此結果可提供業界參考，以提升企業之競爭力，並作為推動永續發展之參考依據。

關鍵字

永續發展；中小製造業；模糊德爾菲；灰關聯分析；模糊積分；理想解類似度順序偏好法；約略集合理論

並列摘要

In recent years, enterprises have gradually shifted their core towards sustainable development in order to maintain market competitiveness and achieve the goal of sustainable operation and management. To obtain the sustainability, three aspects must be taken into account as mainstream for sustainable development, which are economy, society, and environment. These three mainstream aspects not only have already gathered everyone’s attention, but are deemed as important indicators in terms of improving competitiveness. However, studies regarding sustainable development in the past mostly focus on theoretical discussions instead of data analysis via quantitative approach. Based on the above reasons, this study offers an integrated multi-attributed decision analysis model to examine the sustainable development among small and medium enterprises in Taiwan. This study uses the fuzzy Delphi method (FDM) to select the criteria of the three facets of sustainable development, and the grey relational analysis (GRA), the fuzzy integral (FI), and the technique for order preference by similarity to ideal solution (TOPSIS) to conduct a performance evaluation. Through the sorting of performance, the performance evaluation is then classified. Next, in order to explore the core properties this research uses the rough set theory (RST), and lastly, to compare similarities and differences in the three models using sensitivity analysis. Motives of using the above-mentioned three evaluation models are (1) the grey relational analysis could be used to conduct correlation analysis and model construction in terms of inaccurate and incomplete information system; (2) the fuzzy integral method does not require a mutual independence among evaluation facets as long as it complies with monotony; (3) the technique for order preference by similarity to ideal solution is used to establish an integrated multi-attributed decision analysis model because it comes with increasing and decreasing features in terms of each evaluation indicator. Each of the above three models has its own features, and is adopted in this study together with rough set theory to investigate similarities and differences in the results derived from such. This study aims to analyze which key strategy should be given the first priority to be taken into account among small and medium enterprises in Taiwan in order to achieve the sustainable development. According to the results, “increasing employment opportunities in local regions” in the social indicator category, “on time delivery” and “profit” in the economic indicator category are the core properties; “reducing the use of hazardous raw materials” in the environmental indicator category, “employment practice” and “effectiveness of discipline management” in the social indicator category are non-core properties. The results can be adopted as a basis for enterprises to enhance their competitiveness in terms of sustainable development.