We have developed a model which integrates calculation of lnAlGaAs multiple quantum well (MQW) transition energies using the envelope function approximation with a statistical analysis of the PL emission wavelength, net strain and MQW period measured for a variety of MQW designs grown by MOCVD. The model relates the measured MQW parameters directly to MOCVD process parameters, allowing an accurate prediction of the process parameters required to grow a specified MQW design. This greatly reduces the need to grow and characterize individual calibration layers. The difference of the measured and predicted MQW parameters is recorded run-to-run over time, which allows process variability to be analyzed across a number of process parameters with intentional variations to grow different MQW designs. Crown Copyright (C) 2013 Published by Elsevier B.V. All rights reserved.