During quality control of the aircraft manufacturing processes, Airbus generates large amounts of complex data detailing aircraft build characteristics together with the costs incurred and actions taken to ensure quality standards are met. To improve production efficiency by anticipating disruptions and accelerating problem solving, Airbus were looking to combine AI and data visualisation to offer more insight into their complex quality control processes.
The team first developed a data merging and aggregation pipeline to produce a consolidated dataset for AI and visualisation. In collaboration with the Virtual Engineering Centre, University of Liverpool they went on to create interactive dashboards and 3D data visualisations, enabling individual build characteristics and associated costs to be viewed in the context of similar products. This allowed cost variation to be interrogated over time. The team are now looking to apply machine learning techniques to understand the influence of aircraft customisation and modification on the cost of the build process.
Undertaken as part of the Innovation Return on Research (IROR) programme, a collaboration between STFC and IBM Research – this work is generating insights into the collection, organisation and processing of production data required for the application of AI methods.
The data analytics from the project will quantify the opportunities for efficiency improvements in the production process, enabling Airbus to target improvements more effectively. With the analytics tools and visualisations generated, Airbus will be able to interrogate their production data more efficiently, resulting in better data-driven decision making encompassing all available information.
"The work undertaken as part of the Innovation Return on Research programme could help to drive efficiency improvements in our production and quality control processes."
At a glance
- Using data analytics and artificial intelligenceto discover opportunities for production process improvements
- Interactive dashboards and 3D data visualisations allowing cost variation to be
interrogated over time
- Offering insight into quality control by anticipating disruption, accelerating problem
- Machine learning to understand the influence of aircraft customisation and continuous product improvement on manufacturing efficiency.