Inzu Group – Federated Artificial Intelligence for Predictive Maintenance on Bridges (FAI4CIM)
Inzu Group
Sector: Industry
Business Case
Currently, there is no cost-effective method for automatically detecting structural issues in bridges (10% in Europe, 30% in the USA). Inspection (procedure for assessing the condition of an infrastructure in use) was based on subjective criteria. The development of an ML-based KPI (health index), generated from continuous monitoring of the bridge is proposed to detect structural anomalies.
Objectives
Provide a single explainable KPI that provides information about the structural response of the bridge, without having to analyse multidimensional data or be an expert in structural analysis. Adjust the maintenance curve to the real behaviour.
Use case
The behavioural model is developed without requiring the exact theoretical model of the bridge and without considering the bridge’s original condition. The system can learn in conjunction with other similar bridges, and data can be captured from third-party SHM systems, installed sensors or existing datasets. It allows prioritising maintenance objectively according to the real dynamic degradation.
Infrastructure
Edge Computing
Technology
Machine learning and deep learning
Data
Private time series datasets
Resources
It was necessary to involve experts in HW, FW, SW development, mechanical and structural engineers, and experts in algorithms and artificial intelligence. Virtually all Aingura staff were involved at some stage of development.
Difficulties and learning
Among the many problems encountered, those that caused the most disruption were linked to installation and power supply.
KPIs (business impact and metrics of the model)
30% reduction in the cost of repairs. 80% reduction in the number of inspector visits. Payback period: 6 or 7 years (including periodic reporting).
Funding
The project was funded with a combination of 30% own investment, 50% customer projects, and 20% public funding.
Collaborators, Partners
Multiple: UPM, UCLM, BSC, Ardanuy, Oxys Corp, Hampshire University, Titanium.
