Maintenance planning - 2

Maintenance planning
Begin with the basics.
By Gunnar Gustafsson


Evaluating the maintenance requirement
It’s not sufficient to evaluate only the units needing maintenance. Study the system, the working environment in which the units operate and how the unit or system might affect the operation of the plant itself. Identify bottlenecks that can harm the whole system — or the whole plant. Check bottlenecks with extra care and evaluate their condition to ensure that no unforeseen problems can threaten plant operation.
Keep in mind that the plant’s designers probably didn’t have maintenance foremost in their mind. Instead, they probably had instructions to design and build it as cheaply as possible. Also, the plant probably was erected by workers having little knowledge about maintenance. Most plants can show horrendous examples of the effect of this lack of knowledge. There’s no doubt that plant maintenance often can benefit from a bit of a redesign, an option that shouldn’t be ruled out.
Maintenance requirements typically come in three varieties: condition-based, time-based or run to failure. These classifications are by no means exhaustive and they’re only meant to provide guidance when evaluating the kind of maintenance suitable for each unit evaluated.
Condition-based maintenance (CBM)
This can be the most economical choice, but only if the cost of the monitoring devices isn’t too high and extensive dismantling to check the condition of the equipment isn’t required. CBM works best when simple checks are sufficient to get an indication of the equipment’s condition.
A simple visual inspection can detect leaks or other mechanical faults. You can use touch to detect heat and vibrations. You can listen for damaged bearings and even smell to detect overheating or oil leaks. Also, trending the recorded process variables can be used to detect the need for maintenance. However, for vital or costly equipment, use instruments because they detect faults long before an operator can using only the five senses.
Time-based maintenance
This is the type that manufacturers normally recommended. It’s based solely on the number of operating hours or calendar days a unit has been in operation. This method normally is used when condition monitoring is too difficult or when there’s a clear correlation between operating time and mechanical failure. When using time-based maintenance, consider how the unit is used. The recommended maintenance intervals normally apply to units in fairly constant operation. It’s not useful for units that are mainly in standby mode.
For the latter type of equipment it’s better to multiply the number of starts by an “equivalent” number of operating hours. For example, an emergency generator is tested every week (52 starts per year) and each start can be considered equal to 20 running hours. Multiplying the number of starts by 20 gives a value that should be added to the true running hours to arrive at the maintenance trigger point. This method is recommended because each start causes more wear than an hour in normal operation.
Run to failure
This maintenance approach is used for highly reliable equipment, when it’s difficult to perform condition monitoring or when instrumentation is costly in comparison to the equipment value.
Although not normally recommended, run to failure can be used if the unit in question won’t interfere with plant operation should it cease to function, assuming it can be replaced easily and rapidly. If not, it might be possible to alter the installation to enable a rapid replacement. This maintenance type can be tied to the unit — if the unit needs to be included in the maintenance system at all.
The design of any vital system should be reconfigured if some run-to-failure component can fail often and without warning.
Writing the procedures
After the relevant data for each applicable unit has been captured and its maintenance requirement analyzed, it’s time write procedures. Enter them directly into the computerized maintenance system or write them by hand on a note pad — the important thing is to have the procedures written.
Although it’s essential to involve plant personnel in the equipment analysis, it’s probably unfair to force them to enter data into the computer system. As one operator put it, “If I wanted to sit and write all day, I would have gotten me an office job.” Management should consider outsourcing the initial maintenance procedure drafting and development using a competent technical writer with knowledge about maintenance and producing suitable procedures.
If plant personnel follow the advice given above, the result will be a valuable information resource that forms the basis of the comprehensive set of maintenance procedures. It would reduce the labor required to write the procedures, thus saving a substantial amount of money.
Gunnar Gustafsson is educated as a Marine Engineer with a long experience from operating and maintaining ships, oil rigs and power plants of various types, He now puts his education and experience to good use as a technical writer. Contact him at gunnar.gustafsson@gmail.com.