Autonomous Engineering to Streamline Aerospace Castings Manufacturing Process

Castings are an integral part of the aerospace industry being present in critical components such as engines, landing gear and gearbox housings. Years of development and knowledge must be applied and a number of trials performed to obtain a manufacturing process that yields consistent results for each individual casting component. To produce such high grade castings is an extremely complex process with over 100 different variables that can influence the component’s quality. Type of alloy, design of the part, design of the tooling in which the casting will be made and the method used in the casting process are just a few of the variables that affect the outcome. With the advent of simulation technology, this process has improved as defects can be predicted and prevented, reducing the cycles of trial and error and mitigating development time and investment. The limitation of the traditional process engineering approach is that the impact of process control variables could never be truly analyzed as each simulation would have to be setup and analyzed manually. This process could take months due to the size of the design space to trial, thus reducing the quality and quantity of information that could be analyzed and shared with the end customer in an efficient and timely manner. Modern casting facilities are utilizing Autonomous Engineering technology to develop high quality castings on shorter timelines, while gathering valuable process information and understanding that can be shared with the customer. Resulting data can then be further utilized by the casting designer to improve future part designs. Using the example of an aluminum casting this paper will demonstrate how the new methodology of Autonomous Engineering technology contributes to the demands of the aerospace industry.

Key Words: Autonomous Engineering, Casting, Genetic Algorithm, Statistical Analysis, Process Control