Lesson Learned Explained: how proper data collection and storage proved crucial in predictive maintenance
Lesson 1: take small, careful steps and keep context in mind
Need: predictive maintenance strategy in the aerospace industry
In the aerospace and defense industry, which is characterized by particularly high requirements for precision and reliability, key performance indicators in maintenance, failure prediction or machine condition monitoring, are crucial. The company, which has more than 70 CNC machining centers, sought ways to monitor the production process and optimize metrics by implementing Predictive Maintenance
Solution: predictive maintenance technologies
To achieve the desired goal, the following steps were taken:
- Historicization of data analysis: the collection and storage of readings and calculations has made it possible to analyze trends, monitor the status of equipment and draw conclusions for the future.
- Machine connectivity and integration with production systems: thanks to the use of IIoT (Industrial Internet of Things) technology, machine performance analysis has made it possible to continuously monitor machine status and production processes in real time.
- Monitoring machine performance through the installation of sensors: the sensors monitored additional parameters of equipment operation, providing necessary, and previously completely unavailable, data for analysis, allowing, for example, better monitoring of the production process.
- Measurement visualization: visualization of statuses and KPIs enabled ongoing monitoring and rapid response to anomalies and alarm conditions, minimizing the occurrence of failures.
Effect
Initial results based on artificial intelligence and Machine Learning (machine learning) did not yield the expected results. The AI/ML engine was unable to build adequate predictive models because it turned out that 99% of the downtime occurred when the machines were in manual mode – these situations therefore could not be predicted by the AI/ML algorithms.
The main challenge
A major problem was identified: downtime due to human error, which the AI/ML engines had trouble predicting. In response, additional staff training was provided, which significantly reduced downtime and improved overall KPIs.
Summary
The client, interested in the capabilities of Predictive Maintenance, wanted first and foremost to predict machine downtime, which up to that point had significantly affected the smoothness of production movement.
The real problem, however, turned out to be the lack of prior data work, verification and archiving, making it impossible to know whether the PdM solution would work. A system was developed that not only monitored production data, but also provided analytics to help identify the main problem – human error, which AI/ML could not predict. Solving the problem through staff training proved crucial. Skillful data collection and storage is the foundation of effective production management and a key element in implementing advanced systems such as Predictive Maintenance. Without solid data, any analysis and predictive models become worthless.
Conclusions
- It is impossible to achieve all ambitious goals at once: a realistic approach and phased implementation of new technologies are key.
- Data collection and awareness of production status: it is important to start with data collection to get a complete picture of the actual production status.
- Data transformed into information: context and historical data analysis transform raw data into valuable information.
- The role of humans and algorithms: humans should perform creative tasks, while algorithms can deal with precise and repetitive operations, thus avoiding „human” errors.
- The implementation of Predictive Maintenance at an aerospace and defense company showed that the technology, while powerful, requires the right context and collaboration with people to deliver the intended results.