Many industrial processes are difficult to control because product quality cannot be measured economically on line. One solution to this problem is to use secondary measurements in conjunction with a mathematical model of the process to estimate product quality. This paper presents a method for designing a static estimator which predicts product quality from a linear combination of process input and output measurements. The design method includes procedures for selecting a subset of the availble output measurements so as to obtain an estimator which is relatively insensitive to modeling errors and measurement noise. Application of the estimator to a simulated multicomponent distillation column shows that the composition control achieved with an estimator based on temperature, reflux, and steam flow measurements is comparable to that achieved with instantaneous composition measurements. The composition control using the estimator is far superior to the composition control achieved by attempting to maintain a constant temperature on any single stage of the column.