Turn industrial data into trustworthy performance insights with AI
Join TRUST_IRVA and explore how intelligent data validation, explainable AI and automated root cause analysis can help you improve KPI reliability, understand production losses and make better decisions on the shop floor.
Challenge
Manufacturing companies increasingly rely on real-time production data and KPI dashboards to monitor performance and improve operational efficiency. However, turning this data into reliable, actionable insights remains a challenge.
Sensor failures, communication interruptions, manual data corrections and inconsistent data sources can compromise the accuracy of performance metrics. As a result, operators and production managers may struggle to determine whether poor performance indicators reflect actual production issues or unreliable data.
Even when the data is correct, understanding the underlying causes of performance losses can be difficult. Traditional dashboards show what is happening but rarely explain why. Identifying issues such as microstops, speed losses or production bottlenecks often requires extensive manual analysis and expert knowledge.
This makes troubleshooting time-consuming and increases the risk of incorrect interpretations, delayed interventions and missed opportunities to improve productivity.
Project Goals
In this project (TRUST_IRVA), we aim to develop a trustworthy and explainable AI-driven solution that transforms shop floor data into reliable performance insights and actionable recommendations.
The project combines automated data quality monitoring, anomaly detection and causal reasoning to identify unreliable measurements, explain performance deviations and uncover the root causes of production losses. By integrating sensor data, production logs and operational context, the solution will provide a more complete and accurate understanding of manufacturing performance.
An advanced, user-friendly interface with an AI assistant will translate complex data into clear explanations and context-aware decision support, helping operators and managers identify problems and take appropriate action faster.
The project targets 75% less time spent on data validation and correction, 30% fewer false KPI interpretations and 50% faster root cause identification.
By making industrial data more reliable and easier to interpret, TRUST aims to help manufacturers reduce troubleshooting effort, improve operational decisions and increase productivity.
Join our research
We are looking for companies that want to participate in our research and work with us on the valorisation of the project.
Interested?
Then complete the form below.
