Asset Failure Analysis – Assets break. Assets wear out.
Asset failure analysis challenges your team to investigate ways tangible assets fail at your organization and take steps to reduce asset failure risk. This five-step approach to asset failure analysis draws upon the contributions of all team members and puts you on track to get asset failure under control.
Asset Failure Analysis – Step One – Record Asset Failures
Record failures as they occur. Create Word docs, Excel worksheets, Notepad files – whatever works best for you. Keep it simple to start.
Record the date with each failure occurrence. When failures occur often, build occurrence tallies by day, or week, or month. Equip your team with data to report failure incidents by date (and time when time-of-day is relevant to the failure occurrence).
Make a visual representation of your failure data with a bar chart. Show failures in order of decreasing frequency.
Asset Failure Analysis – Step Two – Determine Failure Investigation Priorities
When your bar chart paints a vivid asset failure history picture, convene the team for a chat.
First Impressions – Bar charts direct your team’s attention to failures that occur most frequently.
It’s a common practice to use failure frequency to prioritize investigation efforts. This suggests you should work on what’s happening most frequently first.
You can do better.
Gain insight with Failure Mode Effects Analysis (FMEA). Assess failure likelihood and the impact of various failures you’ve observed. This will help you to be even smarter about choosing the failures that should get your attention.
To apply FMEA to failure data you’ve recorded, three factors are utilized.
- Probability of Occurrence
- Risk of Non-detection
These factors draw heavily on team judgement to determine. They are used to calculate a ‘risk priority number’ (RPN) for each failure you observe. Failures can then be prioritized for investigation on the basis of their RPNs.
Severity relates to failure impact on the organization. A disabled vital asset can shut down a factory or cause deliveries to come to a halt. Some kinds of failure may have the asset disabled for an extended time. Other failures might entail a quick fix. The weight assigned to severity could also reflect cost of preventive maintenance; the idea being that you might purposefully forego maintenance when a failure is regarded as inconsequential when compared to the cost of maintenance.
Using a scale of 1 – 10, a failure with an easy fix, a low downtime duration, and small consequential cost could be assigned a 1. A failure with a complicated fix, a long downtime duration, and high consequential cost, could be assigned 7 or 8.
Probability of Occurrence, for this example, is the relative failure frequency among all the recorded failures. Other probabilities may be explored. Your team might use an absolute frequency, say, number of expected failures per 100 uses, rather than a relative frequency. The purpose is to apply a factor that more heavily weighs failures that occur more frequently.
Risk of Non-Detection can be given a scale of 1 – 10. Failures that are likely to detected can have a low factor, say 1. Failures that may go undetected (until they have significantly impacted the organization) can have a high factor, like 7 or 8.
Reserve factor weights of 9 or 10 for failures that relate to catastrophe like loss of life, injury, or severe damage to the business.
RPN calculation can be Severity x Probability of Occurrence x Risk of Non-Detection.
As a result, when the failures are sorted by RPN, you can see that the failure sequence differs from when they are sorted by frequency alone. This sorted list establishes the team’s failure investigation priorities. Failures with the highest RPNs get the most attention.
Determining factor values to use will challenge the team. As a cross-functional group, varying business perspectives will impact opinions about factor magnitude.
Furthermore, factor scale can be a point of contention. If the “Probability” factors were all divided by 10, they would still reflect relative probability of occurrence among the different failures. Their overall impact on RPN values would be much less substantial. Team wisdom is required to assign reasonable factors, and factor scaling, to arrive at RPNs the team can rely on to influence the investigation priority assigned to each failure.
Asset Failure Analysis – Step Three – Determine Failure Root Causes
Problems get solved when their root causes are identified and treated. The web is replete with articles about ways to get to the root cause of any problem. This post introduces two widely used techniques that work well with problem solving teams.
- Cause and Effect Diagrams
With a whiteboard and a marker you can start a lively discussion to uncover failure root causes.Start the conversation with a statement of the “effect” – the asset failure you wish to avoid. Then, you propose cause categories like “Materials”, “People”, “Equipment”, and “Process”. For each of these cause categories, your team develops a list of potential causes for the undesired effect. Finally, each of the potential causes is investigated to assess its probable contribution to the effect.learn about cause and effect diagramsHere’s a link to a PowerPoint template for cause and effect diagrams.
cause and effect diagram template
- 5 Whys
Asking “Why?” can help you and your team drill down to uncover problem root causes.Why did the truck get the flat tire?
Because it drove over a nail in our vehicle service shop.
Why was there a nail on the floor in the vehicle service shop?
Because a box of loose nails was spilled from a work table.
Why was a box of loose nails sitting on a work table where it could be spilled?
Because a workshop table repair required nails. A box of nails was brought in from the wood shop to support the repair.
Your team will typically get to the root cause of most problems by asking “Why?” approximately five times.
In this example, the team implemented a “corrective” action. (Procedures were updated to require thorough sweeping of the work area after repairs are performed. This picks up stray tools and hardware, like nails, that are brought to the vehicle service shop to support repairs.)
Asset Failure Analysis – Step Four – Take “Corrective” and “Preventive” Action
Action must be taken to remedy the failure root cause(s) found by the team. The desired outcome is to eliminate, or at least reduce the frequency of, observed or anticipated failures.
What’s the difference between a “corrective” and a “preventive” action?
A corrective action is taken as a response to an event that has already occurred. The goal is to reduce or eliminate the likelihood that the event will recur.
A preventive action is taken in anticipation of an event that might occur. The goal is the reduce or eliminate the likelihood that the event will ever occur.
Most team actions will be corrective because they will be in response to observed occurrences. Discussion may occasionally reveal an opportunity to make a change to prevent a failure that has not yet occurred. Such actions would be regarded as preventive.
What kinds of action might your team pursue?
When an asset failure is connected to human error, enhancements can be made to training to reduce the occurrence of the error.
When a fault is found with the tangible asset, the issue may be traced to missed or improperly performed preventive maintenance. The team might also discover quality differences among asset suppliers. These findings could influence purchasing decisions to favor suppliers whose assets fail less often.
Asset Failure Analysis – Step Five – Chart Your Progress
Line charts provide objective evidence to show the impact of your team’s efforts on reducing failure rates over time.
Asset Failure Analysis – It’s An Ongoing Process
Build failure recording into daily work practices. Develop and streamline your day-to-day recording techniques. Periodically create bar charts and meet with the team to calculate RPNs, identify root causes, take corrective and preventive actions, and chart your progress.
Finally, you should see the benefits of having a cross-functional team in your failure analysis process. Identification of root causes and formulating corrective actions can draw upon all disciplines within your organization – purchasing, human resources, quality assurance, operations, etc.
Asset Failure Analysis – Take a Deeper Dive if You Wish
Your team can further refine factors used to calculate RPN by tweaking the factors to reflect asset age, supplier history, and seasonality.
We found good treatment of FMEA in a dissertation from University of Groningen (The Netherlands) FMEA for Asset Maintenance