Abstract: When targets are integrated into selection of new product development alternatives, degree of target achievement of different alternatives can be considered, which is useful to select the most competitive one. Considering hybrid information representations and interacting attributes, an approach based on target-oriented decision analysis (TODA) and k-order additive fuzzy measures is proposed to select new product development alternatives. First, in the case of hybrid information representations of attributes and their targets, such as the forms of interval value, fuzzy number, linguistic value, these forms are transformed into probability density functions. By taking three kinds of preferences on attributes into consideration, probability of target achievement is then calculated based on TODA. Second, with initial preferences of the decision maker, such as sign and strength of interactions among attributes, k-order additive fuzzy measures are identified based on the minimum variance approach. Each attribute's probability of target achievement is then aggregated through Choquet integral, which is used to rank and select alternatives. Finally, the approach is validated through selecting competitive test equipment for large scale integrated circuits.

Key words: target, hybrid information, interacting attributes, target-oriented decision analysis, k-order additive fuzzy measures

摘要： 将目标值融入到新产品开发方案选择中,考虑方案属性值达成目标值的情况,有助于企业选择更具竞争力的产品开发方案。针对属性值和目标值的混合信息表征以及属性交互的问题,提出基于目标导向决策分析和k-可加模糊测度的新产品开发方案选择方法。首先,考虑目标值和属性值表示为区间值、模糊数、语言值等混合信息的情形,将其转化为概率密度;结合属性的三种偏好,利用目标导向决策分析计算属性值达成目标值的概率。其次,基于属性交互方向和强度等信息,利用最小方差法识别k-可加模糊测度,进而利用Choquet积分算子集结各属性的目标达成概率作为产品开发方案选择的依据。最后,将方法应用于大型集成电路测试仪的开发方案选择,验证了有效性。

关键词: 目标值, 混合信息, 属性交互, 目标导向决策分析, k-可加模糊测度