Motorblog

Why Our Maintenance Is Ineffective Part 1

Issues such as wrench-time and key performance indicators actually push our industry backwards.

Even with things ‘slowing down,’ activity within business sure seems to be going full steam ahead.  Companies are working hard at controlling their costs, which puts our industry in a bit of a predicament.  Worse is that a growing number of consultants, both within the reliability or business arena, are approaching companies with ways of ‘reducing maintenance costs’ as the lead.  However, when the activities brought to bear by these consultants are implemented, after ‘reducing staff’ and other cost and budget cutting measures, the workload brought about by their new philosophies tend to increase.  In effect, a lot of the activities are ineffective and are usually implemented full-scale without a full understanding of the impact to the bottom line.

A great many philosophies are built around the idea that we must implement something ‘new,’ that by paying a large bill and tying up personnel we are going to somehow improve efficiency of single points within the maintenance organization.  For instance, we must improve wrench-time, select and use key performance indicators, utilize new techniques and disregard the recommended procedures, have vendors develop our specifications and dictate the quality of what we are to expect, allow our purchasing organizations to focus everything on initial purchase cost, and more.  We are so focused on the small picture that I have gone into companies where there are motion sensors and tags next to light switches talking about the cost of electricity related to that light, yet they have air and steam leaks, no lagging on steam pipes and other energy related issues that measure into the $Millions.

We spend our time working on improving the local efficiencies related to energy and reliability without a full understanding of the impact on the rest of the system.  We are performing cost-centric maintenance instead of throughput-centric maintenance, which is required.  This means getting away from focusing on the individual cost centers and looking at the system as a whole.  Basically, determining what points in the maintenance system are the constraints that limit your maintenance throughput.  These are areas where maintenance backlogs are the greatest, and there may be more than one due to the systems involved in maintenance, just as I have mentioned in the past that there are multiple ‘production scheduling’ methods built into the maintenance scheduling system, depending on the type of maintenance (reference ‘Physical Asset Management for the Executive’).

In our present environment we tend to work against these constraints, building up unreliability in our system.  In effect, utilizing our resources, time, energy, and budget to drive poor reliability of a system.  Instead, a step back from the individual systems to a view of the whole will allow you to see where the issues are.  DON’T SWEAT THE SMALL STUFF!  (Reference ‘A Novel Approach to Industrial Assessments for Improved Energy, Waste Stream, Process, and Reliability’ a book by Howard W Penrose, Ph.D., 2000) If you see a significant build up of backlogged maintenance this will identify the location of your constraint.  Once it is identified, you can subordinate everything to the constraint and try to raise it.  Once the constraint is broken you move on to the next.

For example, if you find that you have a significant backlog of Motor Circuit Analysis (MCA) tests in a particular area, you need to determine why the backlog exists.  Could it be that the analysts have too many tasks?  Production will not turn off the equipment?  Analysts are pulled to do other jobs or tasks not related to the work?  Is the MCA being performed on critical equipment?  Questions such as these can be used to help us identify how to address the constraint.  Perhaps the equipment isn’t critical and the other tasks the analysts are assigned to are on critical equipment?  Or, the planner/scheduler or supervisor does not understand that predictive maintenance inspections are necessary on this equipment.  The constraint should trigger an investigation that identifies the reason for the constraint and then can be used to determine how to raise the constraint through a number of potential options, including:

1. Determining the frequency required and performing at that frequency;
2. Is the test appropriate for that equipment, or would a different test be more effective;
3. What changes to the process of the inspection would eliminate the backlog; or,
4. Can some type of automated data collection system be installed.

For instance, it turns out that the general feeling is that the MCA tests are ineffective.  A review of past records shows that failures were occurring that were not detected but no other problems were found with MCA.  At this point, the evaluation should identify whether it is effective, or if the frequency is not close enough.  The evaluation determines that the MCA analysis is ineffective for this application and the inspection is removed as a task.  No other inspection is appropriate and the equipment is determined not to be critical, so this type of inspection would not be necessary.  The history is rechecked 6 and then 12 months out from the change to ensure that it is effective.  The backlog is removed from the system and the next constraint is located.

The process requires some type of documentation system, such as a Maintenance Effectiveness Review, or MER, to ensure that PM’s are not removed or dismissed out of hand.

A present concept, or methodology, is PM Optimization.  This process helps streamline existing PM’s, with some processes looking at the effectiveness of the PM’s and removal of redundancies, but the usual requirement is to address them all.  This becomes very ineffective as the personnel required to evaluate the PM’s should include the technicians themselves.  This draws time away from their schedule.  The result of a standard PMO, then, is a streamlined set of PM’s that are in place regardless of their effectiveness.

The other area of concern is, of course, Key Performance Indicators, or KPI’s.  These are developed in a similar way to ‘balanced scorecards’ or ‘performance dashboards’ used within operations and business.  The concept here is to develop measurements of the efficiencies of different components of the maintenance program.  The problem with this approach is that it drives the analyst towards addressing the efficiencies themselves and not the system constraints.  If a maintenance dashboard is used, it should address the overall system and not individual components or maintenance processes.

We will discuss these issues more in the weeks to come.

For Comments, please email , President SUCCESS by DESIGN and Member National Writers Union (UAW Local 1981) and the International Federation of Journalists.