A Software Prototype for Predicting Ground Test Costs on the A400M Aircraft

Abstract

Ground testing is an essential, but also expensive, process performed by the Airbus Defence and Space Company (Airbus DS) in the manufacturing cycle of an aircraft. Thus, this process involves a large number of company staff and a significant amount of resources. Consequently, both staff work hours and resource usage imply significant costs, and it is very important for the company to have a prior estimate of these. A research group from the Electronic Technology Department, University of Seville (Spain) worked with Airbus DS to develop a software prototype for predicting costs in the ground testing process for an aircraft. Specifically, the predicted costs were of three types: 1) the number of necessary executions of the tests required for the ground testing of the aircraft, 2) the test failures arising in the test process, and 3) the total time spent for the complete testing of the aircraft. The prediction models were generated through a data mining and modeling process with the simulated data registered by the company in its testing process of the aircraft delivered to date. Once the prototype was designed, a battery of tests on real cases was performed to validate the different models. The success rate of the system for predicting the global cost in tests of an aircraft was 90.48%, which is considered an excellent result. The software prototype makes it possible for Airbus DS to estimate and optimize the necessary resources for the ground testing process of an aircraft.