Product design starts with the analysis of customer needs to result in a design specification that includes such information as: 1) customer needs, 2) metrics, 3) targets, 4) importance of each metric, and 5) dependence relations between any two metrics. We call the metrics design attributes. In generating a design specification, one important issue is how the design specification meets customer needs. Another important issue is the procedure for generating the targets that conform to the design attributes and optimizing them to achieve the maximum customer satisfaction. This issue requires us to represent the aggregation of a set of utilities of design attributes. Existing approaches in the literature generally do not consider the situation where design attributes may be dependent on each other. In this paper, we present an approach to address the dependent attribute problem leading to a function form with design attributes as independent variables. With this function form, we show the potential to optimize the design specification. We present an example to illustrate our approach.