Setting Targets for Oil Cleanliness: A Simplified Approach
Introduction
One of the key aspects of implementing an aggressive proactive maintenance program is the consideration of contamination control. Specifically, the initial emphasis will be on solid particulate. The primary focus in developing a cleanliness strategy is to ensure feasibility of the exercise. The senior management have to be convinced that the strategy is sound.

The first step in cleaning up is setting a target cleanliness level. There are a number of excellent approaches to determining the optimum target cleanliness level. Not least is the Noria approach that considers the Reliability Penalty Factor and the Contaminant Severity Factor. This approach is to be encouraged for accurate target setting. However, there are instances where the approach requires a detailed review of both the business and machine aspects. This can lead to differences of opinion as to the exact numbers used. Experience has shown that this is a particular issue where the Reliability Penalty Factor is concerned.

To overcome this, the approach outlined below is a suggested alternative. It is best suited for a simplistic approach. This makes it ideal for a reliability engineer to quickly ascertain a working target. The target can be later refined using more advanced methodologies. In addition, consultant engineers may find it of value in setting working targets. This is of benefit where information is not forthcoming or there is no consensus of opinion on detailed values normally used.

Determining the Optimum Target
In order to simplify the approach, it is necessary to understand the more complex methodologies. However, the inputs required in these strategies can be summarized as shown in Figure 1 below. Figure 1 illustrates the typical concerns in reliability engineering with regard to the optimized operation of the system.

The next stage is to take these points and apply an indexing value. The indexing value is designed to aid simplicity and therefore is based on a 3 point appraisal: Low/Average/High, or, Good/Acceptable/Poor. Figure 2 below shows how the indexing is set out for the key reliability engineering values.

The simplicity of this approach works in that it encompasses many of the specific value inputs with one key value. Guidance is given below on how the values work.

Capital Cost of Machinery
Whilst this is a self-explanatory value, it offers a simplistic approach to considering the machine as a whole. A low capital cost machine is unlikely to encompass complex components susceptible to contamination. However, a high capital cost machine is more to offer value in terms of improvement through a cleanliness improvement strategy. Small gearboxes and transmissions would be considered a Low value. A more complex engine might be considered an Average value. Turbines and sophisticated hydraulic systems would be rated as a High value.

Repair Cost of Machinery
As with the Capital Cost of the Machinery, the same applies to the Repair Costs. In working with the Low/Average/High values it allows for a cost-optimized selection of the appropriate index. However, where a Low Capital Cost Machine has some complexity it adjusts upwards for specialist repair considerations. Alternatively, where a High Capital Cost Machine is considered a simple enough repair focus, it adjusts downward for lower cost of failure rectification.

Downtime Costs
Again, the Downtime Costs allow for adjustment of values based on possible long lead time issues on parts or rectification issues.

Production Criticality
Production Criticality allows for adjustment where standby units exist or where there is low demand for the machine time. This again makes adjustments for the above values.

Health & Safety Risks
This factor allows for safety risk on factors such as oil changes and repairs and the event of catastrophic failure. It also allows for health risks for personnel close to the machine.

Environmental Impact
This factor is a two-fold consideration. First, it considers the severity of the operating environment and, for humid and dusty applications, would be rated higher than a dry and dust-free environment. Secondly, it also assesses the severity of the machine based on its environment. A machine in a sensitive location would be rated High by comparison to a machine in a controlled area.

Energy Saving Criteria
A machine benefiting from high levels of cleanliness would be more efficient with reduced replacement of lubricants and other components. This factor allows for the desire to reduce the carbon footprint of the machine. Where the machine is a large consumer of power in operation this value should be rated higher than a low power unit.

Current Level of Reliability
If the current level of Reliability is deemed to be good this factor adjusts downward for the need for improved cleanliness. Where other condition monitoring technologies are employed to ensure safe operation, the machine may also be deemed to have acceptable reliability.

Desire to be a Market Leader
This factor is included to allow for further adjustment on numbers. In reality, every business seeks to be a market leader, but to allow for a greater improvement this factor can be included to adjust upwards.

Using the Indexing Table
An example of a completed table is shown in Figure 3. Each of the criteria is circled to indicate the rating. The values are then added to score a Total.

The Total score indicates the improvement in terms of the Range Numbers from the existing level of cleanliness. If the existing level of cleanliness for the gearbox in the example above was ISO 22/20/17, then the target cleanliness level becomes ISO 18/16/13.

Cost Justification
As with all the approaches to setting the targets, the existing level of cleanliness should be known. With this information the engineer can set the new target cleanliness level.

However, it is important to justify the cost benefits of the exercise. This requires knowledge of the existing cost of maintenance and the failure and capital costs of the unit.

In order to calculate the savings, a tangible benefit is required. In this case, as with the Noria system, the predicted life extension resulting from the improved cleanliness is used as the factor.

The Life Extension Factor tables are well known to many Reliability Engineers and have been presented in many Noria training courses and conference papers previously. However, for the purposes of the simplistic approach, the tables have been re-formatted to allow a quick reference.

Consequently, in our example of the gearbox, the required improvement was a 4 ISO code shift. Therefore, the anticipated Life Extension Factor is 3 times the existing life expectancy.

Conclusion
Whilst the method is somewhat limited in detail, it does simplify the target setting approach. Unlike more complex systems that require a consensus of opinion with a wider range of indices on scoring, this is deliberately designed to eliminate difference of opinion. It is also a good check against the target setting of critical systems, whilst offering an easy basis for less critical machines.