Databold's approach was born in aerospace. We are developing right-first-time rigging for aircraft components, and bring our team's experience of turning plant and fleet data into decisions on reliability, maintenance and quality.
Two areas of expertise
Doors and other moving components rigged from a computed set of settings, instead of trial and error on the final assembly line.
See the approach → Industrial AIData platforms that connect telemetry, maintenance, operations and engineering data, from real-time health monitoring to reliability analysis.
See the use cases → Industrial AIAI across the plant: process control, defect prediction and detection, non-conformity analysis and a plant control tower.
See the use cases →Component rigging
An aircraft door, passenger or cargo, and the fuselage it closes are built and measured separately. Their geometry only meets at final assembly, where every step, gap and seal has to be right before the aircraft passes its pressurisation test.
The approach
Four steps, run in two phases, with a go / no-go decision in between based on evidence from real aircraft.
Door and fuselage upstream, then the same aircraft again at final assembly. One consistent protocol.
Separate real deformation from measurement noise; map what moves, where and why.
Model the geometry at final assembly from the upstream measurements, aircraft by aircraft.
One computed set of rigging settings and fittings, and the door / fuselage pairing, ready before rigging starts.
Poorly matched door / fuselage pairs are flagged from the upstream measurements, before any adjustment work starts on the line.
Phase 2
After door completion. Minimum protocol.
Door surround, measured upstream. Minimum protocol.
On the final assembly line: rigging values applied; steps, gaps and pressurisation verified first time.
Closed loop. Actual rigging and test results are fed back, so the model keeps learning with every aircraft.
What it targets
Industrial AI
Most of the value sits in data that already exists, spread across telemetry, maintenance systems, logbooks and engineering files. The approach that works: connect it in one data platform first, then build the use cases that act on it.
Intelligent asset management
How we see data and AI along the asset lifecycle, drawn from our team's past work: from the minute a fault message arrives to reliability decisions taken over months.
All your data connected to the same data lake:
Manufacturing & quality
Where AI creates value in a plant, from preparation to monitoring, drawn from our team's past work. Highlighted: the use cases most mature across the industry today.
Optimise the energy consumption of production assets.
Steer the raw material mix to balance production cost, material availability and quality.
Predict asset failures to plan (and assist) maintenance.
See production in real time and get alerted on any deviation.
Tune process parameters to raise efficiency and quality.
Estimate the probability of defects per batch or product.
Automate manual checks on product quality.
Improve inspection results and efficiency through recommendations.
Qualify and group similar non-conformities to support continuous improvement.
Identify the weak signals that lead to non-conformities.
Suggest relevant remediation plans for recurring non-conformities.
Steer production and industrial flows from one central environment.
Tell us about it in a 30-minute call.