Nagi Gebraeel's guest appearances

Nagi Gebraeel

Industrial and Systems Engineering

OwlTail

Nagi Gebraeel, associate professor in the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech, conducts research in the area of detecting and preventing failure in engineering systems as they degrade over time. The goal is to avoid both expensive downtime and unnecessary maintenance costs. "We could be talking about a fleet of airlines, trucks, trains, ships -- or a manufacturing system," Gebraeel said. "In any of these cases, it's extremely useful for a number of reasons to be able to accurately estimate the remaining useful lifetime of the system or its components." In one project, Gebraeel and his team worked with Rockwell Collins -- a Cedar Rapid, Iowa, maker of avionics and electronics -- to monitor and diagnose the performance of circuit boards that control vital aircraft communications systems. Since the exact time of component failure is unknown, airlines are forced to anticipate when replacements are needed. Scheduled maintenance can result in replacement of parts that still have usable life. Using circuit boards until parts actually fail will resulted in unplanned and expensive downtime. As Gebraeel methodically exposes an avionics component to heat and vibration, he employs a network of computers and sensors to record and analyze data on the degradation rate of the part he is testing. If he can reliably predict the failure rate of a component , he can help airlines replace parts at the most cost effective time. In another effort, Gebraeel has developed an adaptive prognostics system (APS), a custom research tool that allows him to investigate how quickly components degrade under vibration and other stresses. Gebraeel and his team can use the APS to can test a complex system -- such as a gearbox -- by using multiple sensors in a triangulated pattern to detect the frequency signals coming from individual components. Gebraeel is currently in talks with a major airline to use APS to analyze critical engine components. The aim to be able to predict engine wear rates in ways that will help optimize aircraft maintenance procedures. "There's a real need for information about the remaining life of components, so that users can find the economical middle ground between the cost of scheduled replacements and the cost of failure," he said. "Think of the everyday problem of whether we really need to replace vehicle engine oil at 3,000 miles. If we replace it early we sacrifice some useful time, but if we replace it later we risk engine damage. It's very useful to have detailed information about degradation in a system over time."