Many maintenance departments today “fight fires” instead of approaching their problems systematically. Prevention is a far better goal than trying to solve problems as they arise. While this strategy may be a little costly at first, it is not nearly as expensive as allowing problems to occur.
Maintenance problem-solving is primarily concerned with four areas: maintaining critical systems, fixing the problem quickly and faster than the last time, determining what is causing the breakdown to happen so frequently, and identifying the 20 percent of breakdowns that are consuming 80 percent of your resources.
This article focuses on the four common types of maintenance problems with the ultimate goal of helping you to prevent or at least minimize each type.
A problem is a situation that can be characterized by a gap between your existing circumstances and where you do or do not want to be. The gap cannot be eliminated or maintained through obvious methods. Some analysis and creativity are required to define a situation as a “problem.” Visualizing a problem as a gap can be a useful technique. Usually you want to overcome the gap, but sometimes you wish to maintain it. An example would be painting an object to prevent deterioration.
If you can see a solution and all it takes is good planning, then the situation confronting you should be termed a “difficulty” rather than a problem. Of course, if you are experiencing many of these difficulties, there may be a common root cause that could define a problem.
Issues are caused by your goals or a lack of them. You may have an overall goal of wanting your plant to run efficiently with few interruptions, but unless you translate that general goal into viable subgoals, you will experience problems. Establishing specific subgoals is essential if you wish to control the magnitude and number of the inevitable problems. Otherwise, having no goals or only general ones will magnify those problems. Often a disturbance (problem) will force you to ask, “What (unrecognized) goal do I have that is being thwarted by this situation?” Asking this question may cause you to reassess the goal.
The four common types of maintenance problems can be categorized as identification, cause/effect, means and ends. Let’s discuss each of these in turn.
When you don’t understand a natural phenomenon, a question or a method of doing things, your natural inclination is one of curiosity. Industrial maintenance is the same way. You must identify (understand) everything in your department or plant or have someone on staff who does. When a problem occurs, you need to identify where and when it happened as well as where and when it did not. More importantly, you need to identify why you do things a certain way while always on the hunt for a better approach.
In school, you are taught the canned approach to solving problems. While this is important, it only covers problems that are recognized. What about the real-world situations? Industrial maintenance often presents situations that are so confusing that problems are camouflaged. Sorting out the mess means finding the basic problem that spawns all the other effects. This is not easy, as you may solve the wrong problem or try to alleviate symptoms caused by the basic problem. For example, you may put coolers on hot hydraulic systems instead of locating the valve or cylinder that is allowing fluid to flow back to the tank.
Identification problems become relevant not only when trying to understand a situation but also when confusion reigns and the problem is hidden by a mass of effects. The former should be attacked by curiosity and the latter by analysis. These types of problems can also appear when a manager finally asks the question, “What are we spending most of our time on and how could we minimize it?”
To properly solve cause-and-effect problems, you must first learn how to distinguish between cause and effect. Effects are things you perceive with your senses or detect through condition monitoring techniques. They accompany or precede a machine failure.
Typical effects are excessive heat, vibration and noise. A failed bearing or gear is also an effect. Simply changing the component is concentrating on the effect. While this often must be done to restore operation, forgetting about the reason for the failure is neglecting the cause. For instance, excessive heat in a hydraulic system is an effect and a predictor of problems. Concentrating on cooling the system rather than discovering the cause of the excessive heat is an invitation to problems but an all too common solution. Attack the symptom, but don’t forget to unearth the root cause. Remember, symptom is a synonym for effect.
Means problems are generally characterized by questions beginning with “how” such as “How can I accomplish that?” or “How can I improve that?” They leave the choice of means open-ended. With a means problem, you are trying to decide how to achieve a goal. The problem of selecting a goal or end has already been solved, so you are now focusing on how to achieve it.
Typical questions that characterize means problems include how to reduce excessive lubricant failures, how to decrease lubricant costs while maintaining good quality, how to lessen machine downtime, how to improve safety and how to change the department mindset to prevention mode. Solving a means problem often involves finding an expert, but you should never assume the current method is the final answer. Improvement is always possible.
Problems of ends or goals can be characterized by the question, “What goal should I pursue?” As mentioned previously, your goals may be very general at first but must be translated into detailed subgoals to truly matter. Common questions to ask might include which metrics should be used to gauge progress, which 20 percent of the problems are generating 80 percent of the efforts, what are the critical parts of systems that must be constantly monitored, and how are problems categorized (critical, important and projects for correction).
In addition to recognizing the four problem types, you must also be aware that problem-solving can be divided into four levels of sophistication:
These four levels merely describe approaches that can be used on maintenance problems. One is not better than the others but must be selected based on the severity of the problem. Of course, if a maintenance department always focuses on reaction, it might consider moving to a higher level for recurring problems.
Your objectives will determine the problems you experience. Just as there are different levels of sophistication in problem-solving, there are different levels of objectives. These objectives are the ones you set for yourself or your department. The farther down you move on the following list, the smaller the resultant problems should be.
Routine objectives include maintaining things as they are, handling normal (expected) problems, reacting quickly, having lots of spares and adapting to the problem (learning to live with it).
Corrective objectives usually involve the elimination of accepted problems or modifying a design to solve an inherent problem.
Improvement objectives might consist of requesting new equipment, changing the way things are done, concentrating on prevention and providing better training.
Most problems have an immediate phase (or crisis) and must be addressed now. However, managers who want to move to the leadership objectives will try to prevent or minimize a recurrence. While supervisors and management are concerned with doing things right, leadership concerns itself with doing the right things. Remember, setting objectives determines the problems you will encounter. Setting the right objectives will minimize those problems. In the typical plant, supervisors and management trump leadership.
Your prevention efforts must be comprehensive and cover all areas from which problems may arise, such as personnel, maintenance practices, hardware and systems. These categories are most useful when solving cause/effect problems. However, they may also be used to keep a manager focused on all aspects of maintenance.
An example of an Ishikawa diagram of lube oil system strainer basket failures
Two important techniques for establishing a problem’s true cause are the Ishikawa diagram and the Kepner-Tregoe method. These techniques are especially useful with cause/effect problems that defy solution.
The Ishikawa diagram helps you focus on the different aspects of a problem so the listed causes will not be concentrated in one or two areas. For instance, most problems can be broken down into four areas: personnel, maintenance practices, hardware and systems. Some problems may be divisible into more than four, but with some imagination, most should yield at least these four. These categories force you to look at a situation from multiple perspectives to generate possible causes.
Some refer to these diagrams as fishbone diagrams or cause-and-effect (C-E) diagrams. They encourage you to list as many causes as possible. To do this, you must withhold judgment until the listing is complete to assure no one jumps to conclusions.
By contrast, the Kepner-Tregoe method relies on describing what the problem is, what it is not, where it occurs and where it does not. In effect, you are building a fence around the problem to keep important information inside (and under review) while keeping out extraneous information. Your main thrust is to identify what has changed. The true cause will account for all effects. If one effect could not be caused by the selected cause, that cause must be discarded.
Prevention requires maintenance management to develop a new mindset and make a conscious decision to move away from fighting fires. By understanding the four basic types of maintenance problems, the different levels of problem-solving and the three categories of objectives, you will be better prepared to achieve this new mindset.