At what stage would you expect the aviation industry to ensure 100% safety and dependability of every airplane flying out? Not at the tarmac! The assurance must start at the manufacturing level itself. Undoubtedly, it is a mission-critical industry, heavily relying on accurate Quality Gate checks before the aircraft roll out.
The “2022 Aerospace Industry Outlook” published by Deloitte reported the current macroeconomic trends, indicating the demand for small and medium-sized aircraft is growing again. Consequently, aircraft manufacturers need to shift toward digital and operational efficiencies that could accelerate time to market and reduce cycle times. [1]
The Importance of a Scalable and Robust AI/ML Strategy
What company would want to risk losing its competitive thrust? A global leader in the aerospace manufacturing sector was struggling with a terrible lag in its aircraft delivery rate.
Multiple factories were involved in the production of the vehicle structure. The aircraft assembly utilized 100+ workstations to handle 3,000,000+ parts. The final assembly line was also a complex system of a series of workstations designated to complete crucial tasks.
Quality Gates that were established after every few stations to assess the critical parameters detected several internal time and quality perturbations such as the late start of assembly, quality deviations need for reworks, etc. Also, there were external perturbations such as delays in supplier deliverables, crucial deliverables with open work items, etc. An early warning alerts system was also completely missing, which worsened the situation further.
Aside from the lack of visibility on the upstream processes and the pending work in progress (WIP), the failure to pass through quality gate checks affected their output performance.
Consider the enormous impact on business due to such anomalies in the production process!
Yes, AI/ML can give aerospace manufacturing wings!
Our Bodhee® Predictive Quality AI App facilitated integrated data analytics, providing foresight on bottlenecks, predicted success rate at quality gates, and identified failure patterns in critical downstream parameters. Predictive alerts and accurate recommendations for process correction were given by our AI app when the aircraft were only in the initial stages of assembly.
Our revolutionary and much-acclaimed Digital Twin technology enabled the accurate simulation of the entire complex flight engineering and intricate manufacturing processes. The virtual representation of the physical world provided the necessary levels of visibility for precision manufacturing.
The user-friendly and interactive real-time data visualization made the generated production models easily interpretable by production planners. End-to-end visibility of the manufacturing of flight vehicles allows AI-powered data analytics to provide actionable insights, which can minimize the scope for errors in the production line to slip through the Quality Gates.
The advanced ML algorithms utilized the training data and the LIVE production data to provide actionable insights for the operators to make informed timely decisions, which improved process control.
The value derived within just 8-12 weeks of having incorporated our Bodhee® Predictive Quality AI App and ML in their production process proved to be transformative, to say the least. The work closure rate improved by 20%, and there was a commendable quality improvement of 15%. [2]
That was just a high-level retelling of our data-intensive, scalable, and robust AI/ ML technique, which catalyzed the streamlining of their manufacturing, lifting it out of the doldrums!
References
[1] Deloitte. (2021). 2022 Aerospace and Defense Industry Outlook. Deloitte United States. https://www2.deloitte.com/us/en/pages/manufacturing/articles/aerospace-and-defense-industry-outlook.html
[2] Neewee. (n.d.). Case studies Archive – Neewee. Neewee. https://neewee.ai/case-studies/
Brunton, S. L., Nathan Kutz, J., Manohar, K., Aravkin, A. Y., Morgansen, K., Klemisch, J., Goebel, N., Buttrick, J., Poskin, J., Blom-Schieber, A. W., Hogan, T., & McDonald, D. (2021). Data-Driven Aerospace Engineering: Reframing the Industry with Machine Learning. AIAA Journal, 1–26. https://doi.org/10.2514/1.j060131