One of the most cost-effective approaches for system-level reliability improvement is to develop strategies for critical equipment. Equipment strategies may include preventive maintenance (PM), predictive maintenance (PdM), continuous monitoring, commissioning, and redesign. Each of these strategies can be used individually or combined with one or more of the other strategies, with the ultimate goal to optimize the lifecycle cost (LCC).
The development of these strategies for a chemical plant can be very laborious, as well as time and resource intensive. One way to minimize the effort and time required to develop plant-wide critical equipment strategies is to utilize a modular approach. This article will outline how to do that. However, before diving into the modular approach, let’s make sure the basic concepts are covered.
The Basics
For repairable equipment that has been and will be operating for many years, the most commonly used equipment strategies are PM and PdM. For the rest of this article, equipment strategies refer to PM and PdM only. Depending on the failure mode, the cost of failure and the cost of performing maintenance tasks, the benefit ratio of implementing PM/PdM could be as low as 0.1 to 0.3 (lower is better). Here, the benefit ratio is defined as the ratio between the lifecycle cost with the implementation of an equipment strategy and that of running equipment to failure.
Preventive maintenance aims to preserve and maintain equipment reliability by repairing or replacing the worn parts before they actually fail. If not associated with any predictive maintenance, PM is usually performed at a fixed interval, such as every three years of operating time. Predictive maintenance uses techniques to determine the condition of in-service equipment and predicts when a PM is necessary. PdM enables condition-based maintenance, as the PM interval is not only based on the equipment age, but also the equipment-specific configuration and operating conditions. The effectiveness of PdM depends on the timeliness and accuracy of being able to detect problems that will eventually lead to equipment failure.
When developing the equipment strategy for critical equipment, it is important to include both the critical equipment itself and the supporting equipment necessary for the critical equipment to function properly. For example, when considering a centrifugal compressor whose function is to compress a certain type of fluid before it reaches the next chemical processing step, it is important to realize that:
- In order for the compressor rotor to rotate, it must be driven by a motor through a coupling that connects the shafts of both equipment;
- For the compressor bearing to be continuously lubricated, it needs one or two oil pumps;
- To achieve the desired compression ratio, the fluid needs to be cooled by an interstage cooler (e.g., heat exchanger);
- Control panel, instrumentation and control valves are essential for normal and safe operations.
The equipment strategy for the compressor system is not complete unless an equipment strategy for all the previously mentioned supporting equipment (e.g., motor, coupling, pump, cooler, control valves, etc.) is also developed. As the failure of all supporting equipment is tied to the functional failure of the compressor, their reliability cannot be neglected.
Before performing a cost optimization and/or risk mitigation, the following framework for the equipment strategy should be established:
- Location – this can be a single asset or a multiple level of assets following a system hierarchy;
- Function – expectations of the asset by its users; for each location defined at the level above, there needs to be one or more functions defined;
- Functional Failure – ways in which the asset can fail to fulfill the expected function(s); for each function defined at the level above, there needs to be at least one functional failure defined. A functional failure refers to the failure mode (e.g., overheat, high vibration, insufficient lubrication, etc.);
- Cause – each functional failure defined at the level above should have at least one cause associated with it; a cause refers to the failure mechanism (e.g., fatigue, corrosion, wear, etc.) that induces the corresponding functional failure (i.e., failure mode). For each cause, the following details need to be defined:
a.Failure distribution – the commonly used Weibull distribution with two (shape and scale) distribution parameters can be fitted using the available equipment failure data;
b. Corrective maintenance (CM) – when equipment fails unplanned, the CM task specifies the action, duration and cost to bring it back to running; labor, spares and tools associated with the CM task also should be defined;
c.PM and PdM – to ensure equipment capability is maintained or restored when it deteriorates, PM and PdM tasks are the vital lifeline; for each cause, multiple PMs and PdMs can be created. In addition to cost and duration, a detailed PM/PdM procedure helps the planner and technicians perform their tasks more effectively. Labor, spares and tools associated with each PM and/or PdM should be defined.
d. Effect – each cause also should be assigned one or more failure effects; the effects may include production loss in dollars, safety and environmental severity.
Once the equipment strategy framework is completed, simulations can be performed to determine the optimized frequency for performing an individual PM/PdM task or a group of tasks. It also determines statistically what the optimal component(s) of the equipment strategy will be: PM only, PdM only, or PM and PdM.
Now that the basic concepts have been reviewed, let’s explore the modular approach.
Going Modular
It is evident that developing equipment strategies for a complex system like a centrifugal compressor is time-consuming. Given that it is not uncommon for a chemical plant to have hundreds of critical equipment with similar complexity as the centrifugal compressor, the efficiency of equipment strategy development becomes vital to the success of the plant’s reliability improvement program.
A modular approach is proposed to improve the efficiency of equipment strategy development. The key idea is to divide each complex system (e.g., centrifugal compressor) into smaller blocks and develop a generic equipment strategy template for each of the unique types of blocks. Since the majority of block types are common for multiple systems, significant amounts of time spent on developing the same block for all systems can be avoided. The procedure for the proposed approach is as follows:
- Establish a list of plant equipment by class. For a chemical plant, this list may include compressor, pump, motor, drive system, heat exchanger, etc.
- For each equipment class, one or more levels of equipment types can be generated. As an example, for a pump, two types can be added: centrifugal and positive displacement. You can further divide the centrifugal pump into three subtypes: conventional pump, canned motor pump and magnetic drive pump.
- Once equipment class and types (or subtypes) are determined, the generic equipment strategy can be developed for each lowest level equipment type. For a generic equipment strategy, all information in the equipment strategy framework is required, except for failure distribution and effect. At this step, it is important to include as many possible functional failures and/or causes as possible to make the template general. This will save time when applying the generic template to a specific system.
- A specific, system-level equipment strategy can be created by simply adding all required component-level templates together. For example, a CO2 compressor (centrifugal) system equipment strategy requires the following generic templates to be fully developed: compressor, motor, coupling, pump, cooler, control valves, etc.
- For a specific equipment strategy built on generic component templates, the following information needs to be specified:
- Remove functions, functional failures and /or causes that do not apply to the specific equipment from the generic template;
- Establish failure distributions for all applicable failure causes based on historical data or known operating conditions;
- Assign appropriate effect information to all applicable failure causes; this may include operational, safety and environmental, and should consider the specific impact of the equipment on the system in terms of production loss and potential environmental, health and safety (EHS) risk;
- Remove PM/PdM tasks that are not applicable for the specific equipment;
- Tasks across different components may be grouped together to further simplify the optimization process.
The benefits of using the proposed modular approach include:
- Significant time savings: For a generic equipment strategy template that will be used in 20 different systems, the total time spent on developing all 20 blocks is significantly reduced if one generic block is copied 20 times with minor adjustments.
- Information sharing: When a component is added to the system, the resource (e.g., labor, tools, spares) information is automatically added to the system resource and can be shared by other components in the system.
- Simple modification: If one component in the system is updated, the corresponding equipment strategy can be easily updated by replacing the component template with a new one.
- Independence: All blocks in the system are parallel to each other, with no interconnection. Adding a new block or deleting an existing block does not affect the system’s structure.
- Better use of resources: An individual who is most knowledgeable about the operation and maintenance of a certain type of equipment will be assigned the task to develop its equipment strategy.
Conclusion
Efficient equipment strategy development is crucial to the success of a large-scale reliability improvement project. The proposed modular approach, assisted by properly selected reliability-centered maintenance (RCM) software, has been proven to be effective in achieving efficiency, reducing development time and maximizing limited resource utilization.