What is asset management?
In a manufacturing context, particularly process manufacturing, how do we handle this interface between operations and maintenance?
The term “asset management” has been broadly applied to cover many facets of a common question, which at its heart can be summarized as, “How do I eliminate, reduce, or at least manage my risk of equipment failure and the associated costs and losses of production?” There are many ways to answer that question, and they are typically lumped into this one category. Effective answers need to address two elements: technical solutions and work processes.
The technical side involves some kind of sensor that can measure whatever attribute is important to the asset in question. For example, on a large turbine or electric motor, there might be vibration sensors or bearing temperature devices to warn of approaching mechanical failure. The number of sensors is normally in proportion to the criticality of the specific asset and the variety of likely failure modes. In this particular example, if a failure of that turbine shuts down the whole operation, that asset is about as critical as they get. Choosing which sensors get installed is driven by expected failure modes, which is usually based on experience.
Some assets have sensors installed externally, or they might be an integral part of the device itself. The typical example of integral monitoring is a smart field device. The vast majority of process instrumentation and controller devices available today have some internal diagnostic capability. The nature and extent of these capabilities varies by the type of device, manufacturer, and communication protocol.
For example, a pneumatically operated smart control valve has the ability to monitor the air pressure required to move the stem. Let’s say over a two-year period, the air pressure required to close the valve is between 30 and 35 psi. But today it takes 50 psi to move the valve, so the controller can trigger an alarm warning that something has changed, which may be a sign of an approaching failure. Similarly, the transmitter of a flowmeter may report that its internal temperature is increasing due to a failing internal component. In either case, the warning comes before a failure, while there is still time for maintenance to select a course of action suitable for the situation. For a critical asset, the action may be immediate.
A smart device has to communicate this information to whatever control system needs to receive it. Since most companies don’t want to install any more wiring than absolutely necessary, most systems use a clever method to send it over the same wiring that is installed to carry the main process variable. HART communication superimposes a digital signal on top of the analog process variable. A fieldbus, such as Foundation fieldbus or Profibus, includes diagnostic information in the package of data that it sends out normally.
Using information profitably
The technical side of the equation works very well. The tougher element is how companies use that information, which brings us to the work processes part of the discussion.
Virtually any plant worth its salt will use some degree of asset management, at least on its most critical pieces of equipment. There will always be some core of instrumented assets that are critical and difficult to repair, even though it may be small. Those are usually selected based on experience after the previous plant manager was canned for losing too much production time.
Once you’re past this critical group, the picture generally deteriorates. Rather than choosing maintenance tasks based on immediate need as indicated by the diagnostic operations and indicators, equipment runs until failure, or maintenance is performed following a schedule. For some equipment, this is appropriate. There are noncritical devices that can run until they quit, and that’s fine. You don’t replace a light bulb until it burns out. There may be critical devices that can be serviced only when the plant is shut down, so every opportunity may be used.
An effective asset management program has two positive results: interruptions to production are minimized, which supports maximum possible production, and unnecessary maintenance operations are reduced. Maintenance people work only on the items that need work, and they are able to perform these operations at a time when production will be impacted the least. Creating work processes that support this approach are not easy, and this is where the system falls apart for most companies.
There are exceptions. Read the article in the HART Communication supplement with this issue about the MOL refinery in Hungary. By implementing an effective asset management program, this facility enjoys operating rates and profitability that are the envy of other companies in the same industry.
Peter Welander is a content manager for Control Engineering. Reach him at firstname.lastname@example.org