Predictive Maintenance

¿What is PdM?

Predictive maintenance (PdM) is the set of actions that are executed on assets to detect or correct failures early, in order to prevent these failures from causing losses in processes or severe damage to monitored machines.

To perform predictive maintenance, it is necessary to perform a constant monitoring of assets and to implement a system capable of interpreting the enormous amount of generated data. Based on its artificial intelligence (AI) algorithms, and enrolling in the paradigm of Industria 4.0, Terative is able to report the variations that occur in the operating patterns of different machines, and establish the possible causes of these variations.

Differences with other maintenance strategies

Predictive maintenance (or condition-based maintenance) tasks involve verifying if something is failing or is about to fail based on operating conditions.  Unlike this strategy, preventive maintenance tasks typically involve restoring assets or replacing components at fixed intervals. This means that many times items that could have been used for a longer period are replaced, thus increasing the costs associated with consumables and labor.  Corrective maintenance means repairing assets or their components when a severe failure has occurred and this failure is detected from its high cost consequences (equipment breakdown, process stoppage, or accidents).

Predictive maintenance advantages

Among the advantages of predictive maintenance and condition-based maintenance, we find the following:
  • Reduces the cost of asset failures
  • Decreases the time spent on maintenance.
  • Minimizes overtime costs by scheduling activities based on asset status information.
  • Reduces the replacement of functional parts.
  • Minimizes the requirement of emergency spare parts.
  • Optimizes maintenance intervals
  • Reduces the chances of collateral damage to the system.
The advantages of predictive maintenance over corrective maintenance can be also observed in the D-I-P-F curve. During this stage, the evidence of failures can be indicated by factors such as increased vibrations and temperature of the asset, degradation of the state of the oil, increased current consumption, increased sound pressure level or modification, not perceptible by the human ear, of its frequency spectrum. Regular or real-time routines allow for early detection of hidden faults. If the asset is not monitored at this stage, we enter the CORRECTIVE maintenance area. Hidden faults can impair the performance in which a Potential Failure is detected, indicated by point “P”. Evidence of deterioration is audible noise, noticeable increase in temperature, and appearance of odor, leading to subsystem failure or failure of the entire system, at point “F”.