本文主要係提出模糊財務危機預警模型建構的方法，以對目標公司發生財務危機的可能性進行分析預測，提供財務風險的參考警訊。本模型選取財務報表中的流動比率、負債比率、稅後淨利率和存貨週轉率（次）、應收帳款週轉率（次）與總資產週轉率（次）的乘積作為財務性模糊輸入變數。另外，選取9種公司治理相關的積分制指標作為非財務性模糊輸入變數，並制定模糊推論規則庫，建立模糊預警模型的架構，透過模糊推論計算問題公司和正常公司的風險指標值以設定警訊規範，提供使用者投資參考。 風險指標是各公司財務狀況比較的相對性指標，指標值只代表相對可能性而非絕對必然性，但是指標的趨勢走向對於公司財務狀況具有值得重視的意義，而模糊預警模型的建構方式和選用為模糊變數的各項比率，只是應用模糊理論的一種方式，應用者可藉由調整各項模糊要素，使得系統的表現更符合要求。

This research constructs a financial distress prediction model based on fuzzy theory. This fuzzy model can be used to analyze and predict the possibility of the financial crisis, and can generate a financial crisis risk index to provide a warning signal to the banking institute or the individual investors. We select current percentage, liabilities percentage, net income percentage, and the product of inventory turnover, A/R and N/R turnover and total asset turnover as the financial fuzzy input variables as well as the nine non-financial input variables which are related to the company’s management. The fuzzy inference rule base is constructed based on the experts’ experiences for the calculations of the corresponding risk indices of the normal group and the risky group. We compare these risk indices to establish the alert criterion for investors’ reference. The risk index indicates only the relative possibility but not the absolute certainty for which a financial failure could occur. However, the trend of the indices over consecutive years is an important factor to predict the related company’s financial conditions. The construction process of the fuzzy model and the determinations of the corresponding membership functions of the fuzzy variables show flexibility when applying fuzzy theory to the financial distress prediction model. One can adjust these flexible fuzzy variables to enhance the performance of the model.