Predictive Maintenance at the Mosser Sawmill

Success in a Nutshell

Unplanned downtime and fluctuating loads caused by varying wood qualities pose a massive challenge for sawmill operations. To ensure delivery reliability and quality standards, the transition from reactive maintenance to data-driven planability is essential.

▶ The central challenge for Mosser Holzindustrie GmbH was to increase transparency regarding the condition of their equipment. Since factors such as frozen or dry wood significantly affect machine load, a mere “gut feeling” was no longer sufficient to prevent costly production downtime in the harsh environment of a sawmill.

▶ As a solution, wirecube implemented the tailored solution wirecubeONE Predictive Maintenance. By deploying modern sensors (vibration, temperature, ultrasound) and integrating them with the machine control system, measured values are analysed in real time and made accessible to the maintenance team via an intuitive traffic light system on a dashboard.

▶ The result is a significant increase in equipment availability: impending bearing damage is detected at an early stage and repairs are carried out in a planned manner before any shutdown occurs. In addition to the technical safeguarding, Mosser has now gained a clear, data-based view of the influence of materials on machine load, thereby laying the foundation for future AI-supported optimizations.

1000

Data points per second

50

Sensors

200

Thresholds & Alerts

24/7

Machine Monitoring

2

Serious Damage Prevented (within 12 Weeks of PoC Operation)

The Starting Point

In the reality of sawmill operations, the load on machines varies considerably – depending on whether the wood being processed is frozen, heavily dried, or at normal moisture content, and depending on the volume of orders. These characteristics have a direct impact on motors, bearings, and rotating components. Without continuous measurement, however, the relationship between material properties and wear largely remains in the dark. The result: reactive maintenance. Failures occur unexpectedly, downtime becomes costly, and maintenance is difficult to plan. It was therefore particularly important to create transparency and improve planability in order to meet Mosser’s high standards.

What was needed was a robust solution capable of detecting wear and anomalies at an early stage, making measured values visible in real time, and triggering automatic alerts when threshold values are exceeded – usable by the maintenance team without any IT barriers or lengthy training, and reliable in a harsh environment characterised by dust, moisture, and temperature fluctuations. A market study also made clear that the Mosser sawmill required more than a standard turnkey solution – rather than focusing on individual machines, the entire plant needed to be considered, and sensor data had to be combined with data from the machine control system/PLC.

No Standstill, No Surprises

Even during the first phase of data recording, impending bearing damage was identifiable through vibration signatures. Specifically, an imminent failure in the bearing of a shaft in one of the band saws was detected early and averted. The replacement was carried out in a planned manner, and unforeseen downtime was avoided. At the same time, the data made it visible for the first time just how significantly wood properties affect machine load: frozen or particularly dry material produces a different stress pattern than wood with normal moisture content. This transparency fundamentally changed the approach to maintenance work.

„A project like this only works when both sides are willing to venture into new territory together. wirecube brings the data expertise, we know our operations – what emerges from that is more than just a solution. It is genuine mutual learning, and that is exactly the foundation for what we still have planned.” – Plant Management, Mosser Holz Industrie 

Outcome & Outlook

With wirecubeONE Predictive Maintenance, Mosser has successfully made the transition from reactive to predictive maintenance.
The result: fewer instances of downtime, plannable interventions, and a clear, data-driven view of the influence of materials on plant equipment. This creates resilience in production and lays the groundwork for the next optimisation steps – such as linking sensor data with order and production data to evaluate load patterns by wood type, order type, and process parameters, or training AI-supported models for automatic anomaly detection and more precise predictions.

On the sensor side, a specially developed ultrasound sensor from wirecube electronics was also deployed. In upcoming analyses, this will enable the identification and utilisation of further damage-specific patterns, particularly in relation to the saw blades.