Continuous monitoring is particularly relevant when a failure would result in high follow-up costs, the machine is difficult to access, or significant changes can occur between manual measurements. Fluctuating loads and a critical impact on the overall process also support this. We assess criticality, failure consequences, measurability, and existing maintenance processes before installing sensors.

Alarm limits should not be applied across the board to all machines. We take into account the design, speed, load conditions, manufacturer specifications, and the normal operating behavior of the specific machine. A learning or reference phase helps to distinguish operational fluctuations from relevant changes. Limit values ​​and trend rules are then reviewed based on the collected data and adjusted as needed.

A notification is initially an indication and not yet a complete diagnosis. We check which measured parameter has changed, whether the operating condition explains the deviation, and whether further data or an on-site measurement is required. We then assess the urgency and potential damage patterns. This leads to monitoring, targeted diagnosis, a planned measure, or, in cases of high risk, a rapid response.

The measurement interval depends on how quickly a relevant fault can develop, how critical the machine is, and how much the speed or load fluctuates. Data volume, power supply, and communication also play a role. Initially, we record a reference profile and check whether important operating states are adequately represented. Subsequently, intervals and event measurements are adjusted as needed.

Sensors are placed where relevant vibrations or temperatures can be detected with minimal interference. Bearing positions, force flow, housing stiffness, accessibility, and wireless connectivity all play a role in this selection. The positions are documented to ensure comparability of subsequent measurements.

For each alert level, responsible parties, response times, and the next step in the investigation are defined. A notification can initially trigger a plausibility check, the collection of additional data, or an on-site diagnosis. Only then, depending on the risk, does monitoring, planned repair, or immediate action follow.