Quantitative data for design models

New product development has the objective of meeting customer requirements and being producible. In order to have a design meet these objectives, the concurrent engineering team must use metrics to manage the design process from concept to production. In some cases the chosen metrics use concepts that are not always intuitive for the engineers on the team. Also they are not sure where to get the data that is necessary to make a good estimate of the metric. In many cases they have been trained in the arithmetic for computing the mean, standard deviation, and variance, but when it comes to making predictions of these, there is sometimes confusion. This paper will discuss a design process that focuses on critical product requirements, for which statistical values are assigned to each parameter in order to be aware of the target and the variation to be expected. Sources of data will be discussed that can make this design modeling exercise meaningful. The purpose of this work was to better identify the methods to be used in the design concept stage for evaluating the capability of a product to meet its quantitative performance and producibility requirements.>