In predictive or condition-based maintenance, maintenance is performed on base of the actual condition of the machine or system. The condition is determined by periodic or continuous measurements (condition monitoring). This allows maintenance intervals to be extended, or the other way around, maintenance can be performed on a shorter than planned term to prevent early breakdown of a machine.
DB Services Asset knows the switch to predictive maintenance can be complicated and helps in proper implementation of predictive maintenance throughout the organization.
Advantages
With predictive maintenance, the unnecessary exchange of parts is reduced to a minimum. This saves a lot of direct costs. In addition, because the machines are monitored, the chance an unexpected failure occurs is also reduced till a minimum. Using various measuring techniques, the predictability of the lifetime of parts is brought to a maximum. These techniques can be used while the machine is in normal operation, so no down time is created for taking the measurements.
Implementation
The switch from preventive maintenance to predictive requires an investment in both equipment and personnel. The maintenance principle will have to be borne from operator to end-responsible. A good and proper training is the basis for this. DB Asset Services has extensive experience in providing training on everyone's level.
Strategies
There are various motivations to switch to predictive maintenance. These motivations can be combined, but also worth the investment separately:
- Postpone maintenance;
- Prevent failures / consequential damages;
- Prevent (long) repair periods;
- More time to buy spare parts;
- Product improvement;
- Repairs only during docking or planned stop.
Measurement methods
In predictive maintenance we use the following methods:
- Measuring standard machine parameters: power, current, speed;
- Analyse van vloeistoffen
- Analysis of liquids;
- Vibration analysis;
- Thermography;
- Performance measurements;
- Ultrasound measurements.
Trending
In the ideal case, a condition monitoring program starts after commissioning of installations. However, this is not a requirement. By taking measurements periodically a trend is created. This trend can be compared to many similar installations of which we have information. Thus, it is also possible to start measuring an installation which is already in use.
