Integration of Predictive Technologies into Mainsaver

Challenge:

There are many predictive technologies that may be used to identify problem areas before catastrophic failures occur. These technologies might focus on rotating components such as bearings and couplings and electrical components such as a motor control center. Five of the most common technologies are listed below.

1. Thermography
2. Ultrasound
3. Oil Analysis
4. Monitor Amperage Draw
5. Vibration Analysis

The technologies may be applied with in-house personnel or specialized contractors. In using predictive technologies, it is important to establish baseline values and then to trend periodic readings in order to identify a condition heading out of tolerance indicating a potential future failure. The identified issue may be remedied as a scheduled work order rather than incurring a failure and possible downtime.

Methodology:

1. Establish asset hierarchy

The pump skid below is made up of 5 components, each of which might have some level of predictive maintenance (PDM). The motor, gearbox, pump and 2 couplings might also have their own set of documents, specifications and work order history. For these reasons, it is important that the asset hierarchy go down to the component level.

Pump Skid

Pump Skid setup in Mainsaver as a parent and 5 child assets

2. Determine High Risk Components

Limitations on budget and manpower dictate that we apply predictive maintenance to the components that will give us the biggest bang for our maintenance dollar. Although there are many methods to prioritize these components, three are discussed below. • Asset Criticality Factor – a dropdown list in Mainsaver on each asset record which indicates the relative importance of the equipment. User definable in System administration, the asset criticality factor is a required field on the asset record and may also include a number factor known as RIME (Ranked Index for Maintenance Efficiency) which may be used to prioritize work orders. Using this method, the criticality ‘A’ assets would be priority with respect to predictive maintenance.

• Risk Matrix – Mainsaver allows users to perform a risk assessment on each asset and identify the assets with the highest risk factor. Within Mainsaver, risk is computed by multiplying 3 risk components which are setup in system administration with a related numeric value for the computation; o Condition Code o Consequence of Failure (COF) o Probability of Failure (POF)

The higher the risk score, the higher the risk

• FMEA (Failure Mode and Effects Analysis) Dating back to the 1950’s, FMEA is a systematic approach to determining the higher risk areas in a plant, system or subsystem looking at several contributors to risk.

3. Setup PM Master Records or Routes

It is common to use Mainsaver to setup PM tasks to perform the predictive work on a calendar interval however the completion of a simple PM work order does not allow the trending of data values. Mainsaver Routes and Job Plans allowing setting up predictive maintenance rounds and recording the readings in the database.

PM record for a quarterly PDM task

Mainsaver Route with a combination of data feedback types

4. Trend Data

A single data point outside of the context of historical values typically does not provide enough information to determine a potential failure condition. The user needs access to historical data values in order to identify a trend.
Contractors and vendors who specialize in predictive maintenance tools may have their own software for recording and trending values. Oil analysis and other reports often contain baseline data and past test history. The documents may be scanned and/or linked to the Mainsaver asset, PM or work order record to maintain an electronic bridge to this history through the references tab in Mainsaver.
If the data values are recorded in a Mainsaver Route, the historical data may be accessed in graph format on a tablet or laptop while recording new values.

5. Send failed components to ‘Quarantine Area’ for forensic analysis

When a component fails, there might be some lessons learned. What was the cause of the failure? Were we over-greasing the bearing? Do we have a misaligned coupling? Rather than just trashing failed components, they should be placed in an area for forensic investigation. Learn what cause the failure and what countermeasures may be employed to prevent such failures. Countermeasures might include re-engineering the component, changing lubrication processes, adding additional PM steps of increasing the frequency of predictive techniques.

Benefit of Predictive Maintenance

Heading off a catastrophic failure by finding failing components has obvious benefits with respect to downtime and productivity. Additionally, many of the PDM tools provide the ability to diagnose issues with the machine running and all safety guards in place.