Modern flight test tends to be a complex, expensive undertaking, so any increases in efficiency would result in considerable savings. Lean thinking is an effective method to identify value and eliminate waste and is particularly effective when applied to the entire flight test enterprise. Design of experiments is a statistical methodology that results in highly efficient flight test where only the samples needed are collected and analyzed. The application of design of experiments to flight test's value stream can result in a significant increase in lean value to the stakeholders and a reduction of waste. Design of experiments uses objective, verifiable statistical analysis to plan, execute, and analyze test. With lean stakeholder input, design of experiment's planning process ensures that the variables of interest are measured to provide a specified confidence in the analysis. Statistical confidence provides critical information on both the magnitude and significance of system performance. Further, design of experiments provides a statistical power that tells the stakeholders what magnitude of the effect could have been detected. Another benefit is that system response models are continually generated as test matrices are executed. These models enable the stakeholders to understand and evaluate system performance. This increases test safety and flexibility, which also creates value and eliminates waste in a resource-limited experiment. A conceptual Regular Paper * This is a US Government work and, as such, is in the public domain in the United States of America. † Author to whom all correspondence should be addressed (e-mail: aatucker@usa.net; cdagli@umr.edu). Vol 12, No. 3, 2009© 2008 201 model is developed which shows how the design of experiments aligns the flight test process along the lean value delivery stream and identifies the stakeholders at each step.
Systems Engineering
