Risky business – spare parts evaluation and asset criticality Print this article

Very often ‘spare parts’ are identified, rightly or wrongly, as one area where capital is expended for no declarable value. Bernard Longmore provides an introduction to the concept of spare part evaluation and the role that asset criticality plays in the process.

Due to the large number of parts typically involved, each with their own myriad of logistical complexities, spare parts evaluation is an exercise that is frequently started and rarely completed. What is required is a standardised methodology that can be used for large numbers of parts being analysed in parallel. Why are parts stocked? A common answer is that they are stocked to mitigate unacceptable risk to the business posed by not having the part if required. This answer creates more difficult questions: How do you quantify risk to the business? What is unacceptable risk? How many parts are enough to make the risk acceptable?

Defining risk
Clearly there is a need to quantify risk for all parts with a high level of repeatability and a tolerable level of error.

We need first to define risk:
Probability or threat of a damage, injury, liability, loss, or other negative occurrence, caused by external or internal vulnerabilities, and which may be neutralised through pre-mediated action.
www.BusinessDictionary.com

Probability is associated with reliability for machinery. We know that some machines are more reliable than others. We also innately know that the longer we are exposed to an ‘un-reliable’ machine the more likely we are to see it fail. This is important in that with enough experience we can determine the probability of failure for any time period. In particular, if we have to make decisions to buy a part today we can determine how likely it is to be required in the time it takes to arrive. If that probability is greater than the acceptable probability of failure we must place the order. Therefore we must know the lead-time for each part in order to know if we need to order that part. Taking a simple example; if a part has an acceptable probability of requirement of 0.5 and everyday there is a 0.1 probability it will be required, then, after five days it will be equal to the acceptable probability. If on day one we know the lead-time is six days we must commence the acquisition of this part.

Asset criticality
How can acceptable risk be determined? Here we must introduce the concept of asset criticality. An asset for our purposes is a physical object, which has value to the business in terms of its function.

Its criticality is determined by the importance of that function to the business or the risk posed to the continuity of the business when that function fails. In short, we define criticality as ‘importance to the business’.If we consider the definition of risk we must consider business criteria such as environment, safety, production (sales loss), and quality (sales loss) in terms of risk to the business. Some of these criteria may be more important than others in terms of risk to the business.

Clearly we need to understand the criticality of a spare part in order to determine how important it is to the business. Conventional methods of determining criticality are labour intensive and lack repeatability and a solution is required. Consider then that the part belongs to a piece of equipment that is also critical which belongs to a system, which is in turn also critical.

It seems fair to suggest that these criticality ratings are related. Following the relationship theme if we invert the model and call the system the ‘parent’, the parent asset’s equipment, i.e. the ‘child’ can inherit the parent criticality and so on down to the spare part level. It should also be possible to introduce a rule that the child cannot be more critical than the parent but can be less critical. A number of factors can affect this:

Parallel redundancy in the system meaning only part of the function is disabled at the child level;
Early detection such as condition monitoring technologies; and,
An ancillary function on the child not effecting the system.

Let us say we accept:

That asset criticality determines part criticality;
That criticality is determined by the relationship the part has with the key business criteria; and,
That the criteria have a relationship to each other in terms of risk.

Now, we have constructed a matrix where criteria occupy one axis with their own ‘weighting’ and systems occupy another with their own relationship to each of the criteria.

Resolving this matrix and following the inheritance rule will generate a criticality value for each spare part. In the examples used the numbers adopted are always from 0.0 to 1.0. This will make acceptable probability evaluations more straightforward later on.

One neat enabler for this method is software based analytical hierarchical process (AHP) such as Expert Choice shown in Figure 1. In the example shown criticality is being resolved for a number of pieces of equipment.

However, it may be more effective to only use such a tool at the parent asset level (Functional Location level in SAP PM). This example was carried out at Wellman International, a polyester staple fibre manufacturing plant in Mullagh, Co. Cavan.

Having determined a solution for determining criticality it is possible to move forward and:

Compare this value to acceptable risk;
Compare acceptable risk to actual reliability;
Understand how using this with lead-time and part cost will determine the eventual stocking requirement

The Author: