Design factors and their effect on PCB assembly yield - Statistical and neural network predictive models

Abstract

The authors relate circuit board design features to assembly yields. Design parameters that may affect the assembly yield are identified using knowledge of the assembly process. These parameters are then quantified for a set of board designs and related to the actual assembly yields by statistical regression models and artificial neural network models. These models are able to predict the assembly yield with a root-mean-square (RMS) error less than 5%. They can be used to predict the assembly yield for new board designs on the same line. Alternatively, they can be used to compare the performance of different lines by comparing the expected yields for a given design with the actual yields.<>