TOPSIS 是多屬性決策中一個重要的方法，此法利用各方案與理想方案間的距離作為選 取最優方案的依據。然而當各準則所蒐集到的資料因為資訊不充足表現出語意上或人為 判斷上的不明確性而為區間資料，以及決策者的風險偏好未知下的決策態度，都會影響 TOPSIS 最優方案的選擇。因此，本研究針對區間值提出一個新的區間資料標準化方法， 並加入決策者風險偏好的考量以強化TOPSIS 方法最優方案的選擇能力。根據初步的研 究結果及敏感度分析顯示，本研究具有解決上述問題之能力。 特別是，在所舉的範例 中，歸類的三種的風險態度包含風險趨避，風險中立和風險偏好的決策者決策行為皆可 納入TOPSIS 模式中而選擇出最佳的方案。 顯然的，當資訊不清下，本研究所提出的方 法將是一個重要工具，因決策者可採用此方法避免任何政策制定的風險並作出最佳方案 的選擇。 TOPSIS is a multi-attribute decision making (MADM) technique for ranking and selection of a number of externally determined alternatives through distance measures. When the collected data for each criterion is interval which is not precisely determined since human judgments are often vague under insufficient information, and the risk attitude for a decision maker is unknown according to their different risk propensity, we propose a new TOPSIS method for analyzing, normalizing and ranking the alternatives interval values. In the propose method, we first derive a new normalized method and risk attitudes for analyzing the TOPSIS with interval data. The results indicated considerable decision makings for different decision maker with different risk attitude. Both the illustrated example and the in-depth analysis confirm the potential benefits of the new approach in terms of TOSIS analysis. Especially, the illustrated example is archived with three kind risk attitudes of risk-averse, risk-neutral, and risk-seeking to derive the best alternative. The results show that our approach will be an important tool when a decision maker would like to avoid any decision-making risk and choose the best alternative.
