Surrounded by technology but pressed for time, even top guns get off-target.

The goal of any maintenance team is to maximize production output. When this doesn't happen, what's the cause? In nearly every case, it's a departure from maintenance basics.

Sometimes, the very nature of the assigned work is at fault, such as making a maintenance crew pull double duty as a construction crew. Consider a plant that hosts the division's corporate offices for 13 plants. The division vice president assumed the maintenance department was her private labor force for building pilot projects, making signs, and doing odd jobs. She misunderstood a core principle of assigning work: You measure the dollar value of a maintenance worker's time in terms of production time saved, not dollars saved by not outsourcing.

Overloading maintenance people to save a few dollars on a construction project forces them to cut corners. A quick visual inspection misses something a thorough inspection would have caught. This results in 3 hr of downtime at $250,000 an hour, but you saved $171.32 on a sign-making project.

Let's look at another example. Squeezed for time after attending two days of seminars, a maintenance engineer quickly reviews test data, missing a voltage imbalance problem that toasts air compressor motors. This error results in a 12-hr plant shutdown.

Doing things right is not enough. You must focus instead on doing the right things. Sometimes, that means questioning management. You must enforce the principles of maintenance, and review requests for work against them. Then assign and prioritize work accordingly. Let's look at what you can control.

Visual inspections. How can you ensure good visual inspections? You can't do this by edict. We all know youprobably won't see a problem. The chances of penalty and pressure for unfinished work is high. Thus, the usual decision is, "This was fine the last time I checked it, so it's okay now." A typical inspection form asks you to check a box that says "okay" for something like transformer oil level. To eliminate inaccurate assessments, your form should ask for the actual level in units of measurement. The sidebar outlines how to make a good inspection form. Remember, maintenance people are not interchangeable. You must respect individual skill levels, and actively develop craft capability.

Safety. Since we plan maintenance activities, how is an accident possible? If you wade through reams of accident reports, you see certain causes repeatedly. Building safety steps into maintenance procedures reduces maintenance-related accidents. Training also helps, but the most important factor is management commitment to personally ensure workers avoid unsafe acts; one on one, face to face.

Grounding. Problems attributed to harmonics, bad power from the utility, bad equipment, and gremlins are usually due to poor grounding and bonding. Common errors include neutrals bonded at multiple places, using earth as a bonding jumper, and grounding system deterioration. Therefore, make inspection of the grounding and lightning protection system routine maintenance.

Testing and trending. Today's powerful test equipment is bargain-priced. Yet, hundreds of case histories reveal an amazing lack of basic testing. But, don't go overboard; evaluate each possible test with a cost benefit analysis. One consulting engineering company I know, Iris Power Engineering, has about 80 different tests for a motor. Most of these are impractical for a 10hp mixer motor, but many apply in other cases. After baseline testing, you should plot trends to look for emerging problems.

Predictive maintenance (PdM). A natural result of testing and trending, PdM allows you to assess equipment health and determine when a failure is likely to occur. This "condition-based maintenance" maximizes production uptime and the economic life of the equipment that makes production possible. PdM reduces maintenance costs, unplanned downtime, and unnecessary intrusion into equipment. PdM can get you off the "repair and replace emergency breakdown merry-go-round."

Preventive maintenance (PM). Before PdM, we scheduled PM by time-related methods (e.g., calendar days or hours of service). These resource-wasting methods couldn't address equipment wear. By basing PM tasks on PdM results, you spend your maintenance efforts fixing problems before they happen. You also reduce what was once the leading cause of downtime: intrusive maintenance work.

Reviewing procedures. Procedures don't necessarily reflect best practices. So, make it a policy to get feedback on each procedure as workers perform it. Make the feedback process easy and quick. For example, handwritten notes may do the trick. The less formal the process, the more effective it is. Don't rely on a procedure review task force, because historically they don't accomplish much. Everyone should look for ways to eliminate nonessential tasks and improve ways of doing essential tasks. Maintenance managers can improve procedures by getting out from behind desks and foregoing non-essential meetings to observe work in the field.

Training. "What you don't know won't hurt you" doesn't apply to maintenance. Many of us assume training is a function of how much we've budgeted for expensive offsite classes. Think beyond this mindset. For example, using On the Job Training (OJT) enables you to assign a trained person as a tutor to someone who needs specific training on a certain task. To realize the full potential of OJT, assign specific goals and checklists to both workers.

Among your other options are online courses, assigned reading of technical books and articles, and vendor classes. You may find another company needs training on subject X (which your people know inside out), but has an expert on subject Y (which you need training on). Networking inprofessional organizations, such as the IEEE and NETA, helps you find such opportunities.

Analysis. Testing is useless if you don't use the results. If you can't understand the results, don't run the tests. Employees capable of understanding test results have left many downsized companies. If this is a problem you struggle with, an engineering consulting firm can help you. Use common tools such as Pareto charts to analyze test results. Then, form an action plan.

Action. One power monitor consultant has a list of cases where facilities went down for a problem that the power monitor had identified ahead of time. When you treat a power monitor as a wall ornament rather than as a tool for deciding what actions to take, you miss the point of having one. And you miss the point of having a maintenance department. "Paralysis by analysis" happens when you spend too much time managing information rather than acting on it. Instead of being distracted by technology, use it to help focus your efforts on those actions most likely to improve production. Such a philosophy is a basic principle of effective maintenance.