Developing and Implementing RCM for a Limited Staffed Facility

by Tim Jackson, Generation Asset Engineer, Florida Municipal Power Agency and Todd A. Cooper, Senior Consultant, Cohesive Information Solutions, Inc.

Background

FMPA is a wholesale power agency owned by municipal electric utilities. FMPA’s membership today includes 30 municipal electric utilities serving approximately 1.8 million Floridians. Fifteen members purchase all their power requirements from the Agency. Five municipals purchase some of their power requirements. In all, FMPA supplies nearly 50% of its members’ total power needs and has ownership interests in 12 operating power plants.

The Treasure Coast Energy Center was constructed by FMPA and went commercial in May of 2008. This 300 MWe Combined Cycle generating plant has been designed to be operated and maintained by a small maintenance and operations staff.

The need was identified to establish a mature maintenance process early in the plant life to allow the staff to maximize the effectiveness of their program, minimize unnecessary activities, and increase the reliability and therefore worth of the plant equipment.

Challenge

For the new plant, a relatively small size staff was designated: 4 Shift Operations Leads, 4 Operators, and 4 Maintenance Technicians (3 Instrument and Electrical and 1 Mechanic / Welder), 1 Plant Manager, 1 Operations & Maintenance Manager, and 1 Administrative / Warehouse person.

Goal

To develop a process that reduces the need for technical involvement in maintaining and improving the implemented maintenance program.

To effectively manage and optimize the preventive maintenance program, a streamlined RCM approach should be taken. This looks at a failure mode and prevention strategy at an equipment type level and applies different factors based on the functional location where the equipment is installed. For instance, a motor might have ten different failure modes and six different predictive/preventive maintenance tasks intended to monitor or prevent those failures from occurring. However, some functional locations where the motor might be installed will have no effect on the plant operations if a failure occurs. In these instances, it is more cost effective to run the motor until a failure occurs and not invest in its prevention. In a different functional location, the failure might constitute a single point of failure vulnerability for plant operation. In this case, almost any preventive/predictive maintenance that can be performed to prevent a functional failure is cost effective.

Plan

To develop the RCM process, the following steps were taken:

  1. Prioritize the plant components by functional location. Criticality is normally broken down into three categories (five in a nuclear environment). In other words, by the impact the failure of the equipment would have on the plant:
    - The first and highest is “critical” meaning that a failure at the functional location involves a single point of failure that could bring the unit off-line.
    - The second is “important” meaning that the functional location is critical to the operation of the unit but a backup or other strategy exists such that a functional failure does not immediately bring the unit offline but significantly reduces the safety margin leading to the unit coming off-line.
    - The third is typically “run to failure” meaning that a failure does not affect plant operation and can be dealt with in a scheduled fashion.
  2. For each piece of equipment identified as Critical or Important, develop a list of significant failure mechanisms. Both manufacturer and industry information will be useful in the development of this list. For each vendor recommended PM activity, a failure mechanism should be clearly delineated. Once all of the maintenance strategies have been developed and applied to functional locations, all PMs should be tied to a maintenance strategy. Any PMs that are left over at that point should be evaluated as to why they should be kept.
  3. For each Critical component, a time-based PM is designed to prevent each identified failure mechanism. The initial frequency will be based on manufacturer recommendations or regulatory requirements. After a period of operation, age-exploration will be used to optimize the PM frequencies.
    The implementation of a Reliability Centered Maintenance program begins with a solid and practical Preventive Maintenance (PM) program. To adopt a RCM approach, the PM program needs to be dynamic, not remain static. In other words, as more knowledge of the individual equipment is obtained, change the frequency of performance (or change to a condition based initiation) to optimize the benefits of performing the PM. During the testing for the development of the RCM philosophy, it was determined that performing PMs too often contributed to failures. When the frequency was decreased, random failures decreased also.
  4. For each Important component, Condition Monitoring is used to determine when PM activities are needed to be performed. The Condition Monitoring uses metrics designed to determine degradation that could lead to the significant failure mechanisms for the Location.
  5. No PM program is used for the Run-to-Failure components. Since these components have no discernible impact to the facility if failure occurs, the maintenance strategy is to wait until failure to perform maintenance. Even if there are signs of degradation, do not perturb the schedule of maintenance on more significant items, allow the component to fail prior to performing work.
  6. Mechanisms are implemented in the Work Order process to allow the technicians to be aware of the design function of the components and the justification for the PMs and frequencies. Their feedback on the Work Orders are solicited to allow changes to the program based on the personnel most familiar the equipment.

Actions Taken

To support the Maintenance program, a reduced Total Productive Maintenance process has been combined with the RCM process. Plant operators are authorized and trained to perform “tool-pouch” maintenance activities as they are identified. Toolpouch Maintenance is the process for performing work that does not require pre-planning. It is important to capture the cost (material and labor hours) and fact the work was completed. Toolpouch maintenance is simple in nature, within the basic understanding and training (Skillof- Craft) of the personnel.

The purpose of the Toolpouch Maintenance process is to:

UNPLANNED WORK ORDER APPLICATION

UNPLANNED WORK ORDER APPLICATION

Reactive Maintenance will occasionally have to be implemented on a piece of equipment that is not inconsequential, but overall the goal of the program is to detect degradation before it occurs and implementing preventive maintenance to prevent the unexpected failure. When the program is implemented well, it is common for reactive maintenance to only account for 10 to 20 percent of the total maintenance performed. When combined with reduced numbers of Preventive Maintenance activities, these results in a more efficient, effective, and cost saving maintenance program.

There are four major components of the Reliability Centered Maintenance program: Reactive Maintenance (Corrective Maintenance), Preventive Maintenance, Predictive Maintenance (Condition Monitoring), and Proactive Maintenance.

Four components of reliability centered maintenance program

Since the installed EAM program at TCEC already supports the Reactive, Preventive, and Predictive Maintenance functions, FMPA and Cohesive have developed a business solution which includes the information and processes necessary to support a RCM program. The solutions address the needs at the Location level in the component structure. The Location defines the significance (Criticality) of the equipment, since it is the level at which the function is defined. A single model of pump, for example, may be used in functions that have no impact if removed from service or it may be used in functions that result in a site shutdown if removed from service. As a result, the use of Location in the RCM process is appropriate. Each Location is defined with:

In the Preventive Maintenance application, a tab has been entered which includes the purpose of the PM, the justification for the frequency that the PM is performed at, and the exact Failure Modes that the PM is designed to prevent from occurring. This information is fed to the Locations application to provide the Maintenance Strategy table information on the RCM tab. Conversely, by selecting the associated Location in the PM application, the failure class information is provided to the PM application.

Personnel at multiple levels are engaged in the process, including the technicians in the field, who are providing feedback whenever a maintenance activity is performed. This information is needed to effectively optimize the preventive maintenance activities and frequencies (PM Optimization). The PM and Frequency Bases are provided to the technician in the Work Order Post-Work tab from the PM application. This provides a benchmark for the technicians to judge effectiveness and recommend changes. When Corrective Maintenance is performed, the design function of the Location is provided on the Post-Work tab for the technician to assess the significance of the failure and to determine recommendations to prevent recurrences.

Expanded failure code lists have been developed for the Failure Reporting sub-tab that are used to identify human error, organizational, and programmatic failures in addition to equipment failure modes. If a new failure mode is identified during maintenance, the EAM software will notify the responsible personnel to address the necessary changes to the failure modes and effects analysis (FMEA). This greatly increases the effectiveness of Root Cause and Failure Analysis (RCFA) activities. The Cohesive solution is designed to implement the proactive maintenance requirements of the RCM process.

RCM Tabs

Anytime a PM is performed, feedback is gathered from the Craft personnel on the as-found condition of the equipment. If there is continued degradation discovered whenever PM is performed, the PM becomes a candidate for frequency reduction. If after a number of performance with no noticeable degradation, the PM becomes a candidate for frequency extension. This process will continue until a frequency is identified where a change is noted. The frequency will then be shortened to the last one where no changes were noted. Monitoring of the equipment will then continue. This PM Optimization methodology is known as “Age Exploration”.

WORK ORDER POST-WORK RCM PREVENTIVE MAINTENANCE TAB

WORK ORDER POST-WORK RCM PREVENTIVE MAINTENANCE TAB

In addition, when a piece of equipment has failed or degraded and corrective maintenance is performed, feedback is solicited from the craft on the failure mechanism and their thoughts on what could have been done to prevent the issue. The assumption is the craft personnel, being the ones who work most closely with the specific equipment in the plant will have formed relevant ideas on how to improve performance. This information will be incorporated into changing maintenance strategies to prevent further failures. Failure modes and causes will be coded on corrective maintenance Work Orders. These codes will be trended for opportunities for improvement.

WORK ORDER POST-WORK RCM CORRECTIVE MAINTENANCE TAB