結合支援向量機與多準則決策方法評鑑綠色供應商績效

USING SUPPORT VECTOR MACHINE COMBINED WITH MULTI-CRITERIA DECISION MAKING METHODS TO EVALUATE GREEN SUPPLIERS PERFORMANCE

隨著環保意識的不斷提高和國際環保組織的要求，綠色供應鏈管理發揮了至關重要的作用。因此，如何選擇符合國際綠色法規的綠色供應商是企業最重要的戰略之一。在過去的研究中，大多數評估準則是透過文獻或專家訪談獲得的，但這方式往往過於主觀。很少有研究使用資料探勘方法來確定評估準則。因此，本研究提出一個新穎的混合模型，分三個階段對綠色供應商進行評估和選擇。首先，根據供應商的實際數據，使用支持向量機（SVM）模型提取有影響力的關鍵準則。其次，應用模糊最佳最差方法（F-BWM）來計算準則的權重。第三，本文利用改進的模糊TOPSIS方法應用於供應商選擇。最後，使用敏感度分析來探索偏好權重的影響。此外，本研究透過某電子業的綠色供應商案例分析，證明了所提出模型的有效性。

With the incessant rise of environmental awareness and requests from the international environmental protection organizations, the green supply chain management has played a vital role. Therefore, how to choose green suppliers that comply with international green regulations is one of most important strategies for the firms. In past studies, most evaluating criteria were obtained through literature or expert interviews, which tended be too subjective. There are few studies using data mining methods to decide the evaluating criteria. Judging from this, this study adopts three stages to evaluate and select the green suppliers. First, a Support Vector Machine (SVM) model was used to extract the influential key criteria according to the actual performance of the suppliers. Second, the Fuzzy Best Worst Method (F-BWM) was applied to calculate the weights of criteria. Third, this paper applies the modified Fuzzy TOPSIS for the supplier selection. Finally, a sensitivity analysis is used to explore the influence of preference weights. In addition, to demonstrate the usefulness of the proposed model, the proposed model was demonstrated through a green manufacturing in the electronics industry.