A frequent question: “How do we get a reliable plant?”
After years of working in various industries, I believe that there is no one correct answer. I often read articles that mention the only way to reliability is to have rigid maintenance programs, condition based monitoring and pro-active maintenance. I think, there are better suited answers, but not only one and definitely not only one way.
How do you define reliability?
Even the gurus argue about how you define reliability. Open your favourite search engine and just see how many definitions Mr Google will come up with.
Reliability should be defined as a function of both your ideal as well as your reality. If you have a plant that suffers 3 breakdowns a day, to define and place a target of 95% reliability on your plant will be a pie in the sky. You will continuously fail. Your team will fail, morale will suffer and you will remain in the same position or even end up worse than where you are now.
“Reliability is a function of Effort vs Risk vs Reward” – D. Martins
Define YOUR plant reliability, measure where it is right now and add 5% or 10%. Then start by aiming for that.
How do you improve reliability?
The best way to improve your reliability, is to change (improve) the things you control and control the things you can not change. What does that look like in practice?
If you do not manage and can not change the capital program, then control it. When you have allocated capital to spend, set very clear expectations of what is to be delivered and have very tight control on whoever is delivering it.
As a Maintenance Leader, you can change the plant’s scheduled downtime. If your plant suffers a major failure every two weeks, yet you only have a shutdown every 4 weeks, change your strategy. Have a major shutdown every two weeks.
When a maintenance contractor, does not deliver enough people on your shutdown, change your supplier. Or bring in a second supplier as competition.
However, the biggest change you can make, is to change your Maintenance and Engineering teams’ focus. Start by leading your team’s focus to the three failures that stopped the plant yesterday, and change something. Let’s say a flow indicator on your furnace cooling system have stopped you six times yesterday. Something needs to be changed. But to change all flow indicators with a new type, all 124 of them, will in all probability not be fully within your control and may even need capital approval.
“Change the things you control and control the things you can not change.” – D.Martins
More importantly, if this flow indicator stopped you yesterday six times, but ran flawlessly the last six months AND there is another 123 still working, why would you change all of them? Rather rule that if a flow indicator trips the third time (and physical flow was confirmed) it just needs to be replaced immediately. Then ensure enough spares are kept to support this strategy.
At a weekly review meeting, discuss the top 3 causes for stoppages, whether based on time or frequency (or both) and review the actions taken by your team. Also review the effectiveness of the previous week’s actions – did the Top 3 show up again? If not, you should be on your way to a reliable plant. If it did, you may want to consider doing a more detailed RCA on some of these failures.
The key is: the action needs to be quick, smart and simple. The change/impact needs to be almost immediately (at least within a week). If the action takes longer than a week – question the simplicity.
If you continue to do this, at some point you will run out of items to address, and your plant reliability will improve. The Maintenance and Engineering teams will have small successes to celebrate on a weekly basis. Small wins.
How do you improve your reliability? Well, how do you eat an elephant? Bite by bite.