You are currently viewing Planning for Results

Planning for Results

There are two things you must do in a successful maintenance program: be good at doing your work, and only do the right work. Both are needed to deliver asset reliability – the cornerstone of sustainable, safe and quality production levels. In chasing reliability many turn to programs for defining the right work, yet many of those efforts will fail. Why? Poor or ineffective planning. The greatest benefits come from defining the right maintenance program using RCM and then implementing with quality work and on schedule.

Failure of reliability improvement programs can be from poor execution of RCM, but more often it is the result of something more basic. They are stuck in a culture of unreliability.

Maintenance and production have a “break-then-fix” mentality. Production runs hard until assets break, then maintenance, acting like a fire department, comes to the rescue. It’s exciting and some may even enjoy it. Maintenance gets to be the hero! Usually, production calls the shots. Maintenance only gets the time to work on assets when they break. Doing proactive maintenance, especially if it requires downtime, is rarely allowed. Production egos dominate and maintainers are treated like their servants. This culture always under performs, failing to achieve world-class levels of performance and desired results.

In a high performing, reliability culture, operations and maintenance share ownership of the asset reliability. Maintenance sustains productive capacity; it does not just restore it after failures. That demands preventing and predicting failures as well as repairing. Prevention and prediction are the dominant philosophies with repairs undertaken in a way that entails the least disruption to productivity. Production understands how failures can occur, operates the equipment within its limits and allows maintenance to do its job.

Operations trusts that short outages for preventative work will results in longer run times overall. It trusts that predictive techniques truly reveal problems in their infancy so that the consequences of pending failures can be mitigated cooperatively. Of course maintenance must have that proactive program that is truly effective.

If you are in a “break then fix” mode, then there is work to be done and trust to be earned. Production must trust that maintenance can deliver on its promises. If it takes equipment out of service for a half a day, then it must be only half a day. Achieving that requires good planning and work scheduling with sufficient material support to the work.

Minor shutdowns and repairs must be completed on time. Finishing late reveals a lack of planning. Any preventative repairs that are followed closely by start up failures (infant mortality) will undermine credibility. All too often, maintainers set themselves up for failure and struggle to gain the needed trust.

Also, there needs to be a realization that some maintenance work can actually make things worse. Too much of the wrong work (usually intrusive preventive work) can actually lead to early failures. Defining the “right work” is what we do with RCM. However, RCM alone, isn’t enough. RCM will fail without the ability to deliver work on schedule. It requires the same on-schedule performance that builds credibility and trust with production. Good planning and scheduling enables work to be completed efficiently and frees up maintainers’ time. In a break-then-fix environment, there is never enough maintenance capacity for all the work.

Planning defines “what” work will be done, the resources needed to do it, and a forecast of how long the work will take. Good planners are experienced trade persons who know what is required to execute the work. Scheduling is done using work priorities and the estimated duration of the planned jobs. Scheduling is all about “when” the work is done. Scheduling does not require trade skills, but it does require organizational skill to ensure that only reasonably achievable schedules are produced given the availability of trades’ resources on hand.

Job plans do not need to be perfect. Striving for perfection is a mistake. By trying to be “perfect” plans take a long time to be produced and inevitably there are flaws anyway. Accept that the first plan will be flawed. Errors will show up during work execution and will be corrected using feedback from the trades when the job is closed out. An 80 per cent correct plan, created in a relatively short time, is far more effective than having no plan at all. When it is corrected after it is first used, it will be much closer to 100%. Save those plans for re-use – you should never need to plan the same job more than once.

Schedules will also need flexibility. Estimates of job duration won’t be 100 per cent accurate. Any job plan flaws will result in delays. In the early days switching from break-then-fix, there will still be many breakdowns and “emergencies.” It will take time for good planning and scheduling practices to take hold. But it won’t take as long as you might think. I’ve seen dramatic changes in just weeks.

Reasonable scheduling (with a minimal amount of planning) demonstrating “on time” delivery, will help calm jittery production managers and reduce pressure for urgent action on everything. Credibility will grow and results will accumulate as you demonstrate “schedule success.” You’ll know you are on the right track when production stops asking for top priority on every job!

Once these practices are firmly in place and a sense of calm is restored, efforts such as RCM will stand a good chance of raising the bar and can deliver even higher levels of safe, productive capacity and profitable performance.

[maxbutton id=”3″ ] Facebooktwitterredditpinterestlinkedinmail