To achieve competitive design, it is essential to develop an optimization method that ensures both high customer satisfaction and robustness for products with multiple criteria. While several studies have proposed optimization methods that integrate TOPSIS with Taguchi method or desirability function, no single study has yet combined all three methods into a unified optimization framework. Therefore, this study proposes an integrated optimization method that combines TOPSIS, Taguchi method and desirability function. The overall process of proposed method is based on the TOPSIS framework. To incorporate Taguchi method and desirability function into TOPSIS, we propose using desirability function for normalization, replacing the traditional vector normalization used in standard TOPSIS. In addition, Signal-to-Noise(S/N) ratios are calculated to evaluate the degree of customer satisfaction. To demonstrate the effectiveness of the proposed method, a hypothetical example is generated under specific conditions, and the resulting rankings are compared with those derived using the original TOPSIS approach. The comparison revealed that the rankings of design alternatives differed between the original TOPSIS and the proposed method. This difference is attributed to the influence of the desirability function’s threshold points, the specific type of desirability function applied (from Kano’s perspective), and the Taguchi S/N ratio used to assess satisfaction levels. These factors enabled a more nuanced evaluation of customer satisfaction and robustness, thereby validating the effectiveness of the proposed optimization method.

1. 서 론
2. Proposed Method
    2.1 Step 1: Build Decision Matrix
    2.2 Step 2: Define Desirability Function forCalculating Customer Satisfaction andNormalizing Decision Matrix
    2.3 Step 3: Calculate Taguchi S/N Ratio forRobust Design
    2.4 Step 4: Set Best & Worst Point andCalculate Euclidean Distance
    2.5 Step 5: Calculate Relative Distance forRanking of Design Alternative
3. 설계 예제
    3.1 Step 1: Build Decision Matrix
    3.2 Step 2: Define Desirability Function forCalculating Customer Satisfaction andNormalizing Decision Matrix
    3.3 Step 3: Calculate Taguchi S/N Ratio forRobust Design
    3.4 Step 4: Set Best & Worst Point andCalculate Euclidean Distance
    3.5 Step 5: Calculate Relative Distance forRanking of Design Alternative
4. 결 론
References

• Jae-Hun Jo(Department of Mechanical Engineering, Hanbat National University) | 조재훈 (국립한밭대학교 기계공학과)
• Yoon-Eui Nahm(Department of Mechanical Engineering, Hanbat National University) | 남윤의 (국립한밭대학교 기계공학과) Corresponding author