written by William C. Worsham and Charles J. Latino
and presented at the 1999 API Pipeline Conference

It does not show up on a company¡¯s balance sheet and is not usually recognized as an asset, but the data that a business possesses is an important asset to the company. A company with good business values can benefit from a reliability program because it generates lots of important data. This data can be used for reliability purposes to improve equipment and process performance, reduce operating costs, and dramatically improve profits. To gather and manage this data, a good Computerized Maintenance Management System (CMMS) is extremely helpful.

To satisfy facility reliability needs the Computerized Maintenance Management System (CMMS) must provide for the collection of data to facilitate Reliability study and analysis. The question becomes how to capture this data and what type of data is needed. The purpose of this paper is to shed some insight into the type data needed and why. From a reliability perspective a CMMS must satisfy the following broad requirements:

Help identify reliability deficiencies.

• Provide data to aid in the analysis of reliability deficiencies.

• Provide reports that measure the effectiveness of reliability corrections.
  

Help Identify Reliability Deficiencies

For the reliability professional to identify maintenance and operational deficiencies certain information is essential. Much of this information can be obtained by introducing shift notes that are kept by mechanics and operators into the CMMS and analyzing tasks depicted therein for frequency and impact. While there are tools, like computers and software that now make it possible to sort through unstructured data in useful ways, data manipulation and analysis is much easier when a logical coding method is used. A brief example of what is meant by logical coding is illustrated below.

Logical Coding

Note: Drop Down Windows Facilitate Logical Coding
Compare this to the illogical coding that follows:

Illogical Coding

Logical coding will require some training for those individuals who write shift notes. Once training is complete and a logical coding method is adopted, deficiencies can be determined. Some examples of the type deficiencies that can be delineated in this manner are:

Repeat pump failures and the amount of production lost.

• Repeat sprocket and/or chain failures and amount of production lost.

• Chronic or repetitive leaks by service line (e.g., steam, acid, condensate, etc.) and dollars expended to correct these leaks.

• Repeat failures of systems or equipment and the amount of production lost in total or per occurrence.

• Stores substitutes made and probable production lost if the substitute fails.

• Trips to stores to secure parts and dollars expended in this effort.

• Number of safety lockouts that occur and the money expended in securing the lockout plus the amount of work or workers delayed while the lockout is secured.

In identifying deficiencies like those above, it is important to not only identify the deficiency, pump failures, but to also identify the cost associated with that deficiency, the amount of lost production. Also, in perusing the above deficiencies, it can be seen that some of the deficiencies are process, leaks, some are equipment related, equipment and system failures, and some are administrative in nature, stores substitutes, trips to stores, and safety lockouts. Typically, process and equipment problems are considered reliability issues whereas administrative situations often are not. This is a fallacy. Administrative issues like substituting for parts in Stores because a substitute part is more easily obtained or costs less often creates reliability issues and should be tracked. Non-value-added situations like trips to stores and safety lockouts, while necessary, are reliability issues and should be analyzed to minimize costs and process interruptions. In addition, there are other reliability deficiencies that must be considered.

Other ways that reliability deficiencies can be identified and analyzed is to keep track of these items in the CMMS. The CMMS should:

Identify and statistically sample the number and type of key strokes made by control room operators. On the surface this activity may seem unnecessary, but it is something that should be examined periodically. The goal should be to maximize operator attention time to the process and minimize keystrokes. The keystrokes may be necessary to operate the process and/or equipment and to provide information on how well the equipment and process is performing. Regardless, keystrokes should be minimized because every keystroke is a chance for error.

• Identify the frequency and type of adjustments made to specific machinery and equipment. It goes almost without saying that frequent and/or large adjustments to machinery and equipment indicates that the process is either being changed frequently, or that the machinery and equipment is operating or was designed unreliably. Frequent adjustments are sometimes taken for granted and go unreported for ages.
  

• Identify the time it takes to ramp down, stay down, and ramp up when an outage or production interruption occurs. Standards for ramping down (shutting down) and ramping up (starting up) should exist or be established. If possible, deviations from standard should be noted and explained. Serious deviations should be investigated. For the length of time the equipment or process stays down, standards by reason (i.e. felt change, filter change, maintenance, etc.) should also be established and deviations noted.

• Identify the frequency and impact of shortages of parts and operating supplies, such as spare parts, raw materials, additives, filters, blanks, hoses, etc. These items, commonly known as stock-outs, can have a serious impact on production. If the stock-out is for a critical item the impact can be devastating. Therefore, stores stock items should be assigned criticality codes and stocked accordingly. All stock-outs should be noted. Care should be taken to designate a stock-out whenever a substitute part is used because the item wanted was not available. A situation comes to mind where an electrician wanted a twenty five amp heater and because it was not available in stores used a 23.5 amp heater as a temporary fix. The piece of equipment involved in that temporary fix experienced numerous resets for months before that condition was uncovered and corrected.
  

• Identify and keep track of the runtime for each process and each piece of operating equipment before a shutdown occurs. Runtime must also be tracked for vehicles, fork trucks, cranes, tow motors, and other types of motorized equipment. Runtime is essential for calculating Mean Time Between Failure (MTBF), an important reliability indicator, for equipment that is not run continuously and useful for calculating MTBF for equipment that is run continuously.

• Identify each human interface and make potential inferences when problems occur as to which interfaces may have contributed to the problem. For example, in an area that experiences heavy failures, the supervisor to hourly employee interface may be contributing to the problem.
  

• Identify product flow slowdowns due to ramping, equipment failure, and administrative order. It is a reliability concern whenever production rates are curtailed for any reason. Whenever this happens, the time the slowdown started, the reason for the slowdown, and its duration should be captured.

Provide Data to Aid in the Analysis of Reliability Deficiencies

Once reliability deficiencies have been identified, it is imperative that these deficiencies be analyzed to determine root cause and corrective measures initiated to improve reliability. The analysis and subsequent solutions can be aided by such CMMS data as discussed in these five major categories:

Process Supporting Data

• Machinery Supporting Data

• Other Supporting Data

• Direct Failure Information

• Financial Information

Process Supporting Data

Depending on the deficiency, certain process supporting data is necessary. In order to cover the full range of possible deficiencies, the list of supporting data becomes fairly large. Included in this list would be a list of processing exemplars and where to obtain actual specimens or samples. Also included would be the results of job audits, quality parameters that must be met to assure a first quality product, and the data and analyses of previous process failures. In addition, a continually updated list of experts in various disciplines, and on various machinery and processes, should be maintained and readily available in the CMMS. Finally, a common problem list should be maintained. The problem list should include a list of chronic problems within the same process such as yield deficiencies, pluggage problems, quality problems, and deteriorating heat transfers.

Machinery Supporting Data

A significant amount of machinery supporting data must be maintained in order to analyze and resolve reliability deficiencies. This data should be stored in the CMMS and should include machine histories that delineate all past repairs and adjustments. It should also include all previous failure analyses on the same equipment. It should also include a listing of the clearances for the machine in question, plus a listing and pictorial representation of the spare parts used in that particular machine. Additionally, for critical machinery the CMMS should include graphics capable of showing 3D cutaway views of the machine with parts installed. Finally, the CMMS should include a list of quality parameters for the machine and a list of past deficiencies in part quality.

Other Supporting Data

Besides process and machinery data, certain other data is necessary to completely resolve reliability deficiencies. This data should be stored, maintained, and retrievable from the CMMS. In this category of data would be such things as: a listing of the applicable operating, maintenance, and safety procedures for the area in which the reliability deficiency exists, a list of trained failure analysts for the site or company, and area training records showing past training experiences. For example, the training records should include who was trained in what subject or skill and particulars about the training. Finally, this category should include a vendor list including phone numbers and who to contact for each vendor.

Direct Failure Information

Many reliability deficiencies involve chronic process or machinery failures. Until resolved chronic failures occur over and over again When they do occur, it is important that direct failure information be gathered and maintained in the CMMS. This data must include operating data for the period of time prior to and at the time of failure, and observations of witnesses prior to and at the time of failure. Gauge readings prior to and at the time of failure should be recorded, and observed liquid spills or gas releases, their timing, volume and/or size should also be recorded. If the failure involved an explosion, the exact location of parts immediately after the explosion occurred should be recorded. Sketches and photographs of the failure site, and eyewitness accounts of people interviewed immediately after the failure should also be recorded. Failed parts should be tagged, bagged, and saved for the failure analyst, and a listing of the failed parts and their present location should be recorded in the CMMS. Finally, operational paradigms should be identified and recorded. For a variety of reasons, operational paradigms are usually best secured by outside experts.

Financial Information

Applying scarce resources to analyze and resolve reliability deficiencies must be cost effective. To assess cost effectiveness and to determine the impact of failures and slowdown on plant financials, certain financial information is necessary. Although there is a reluctance by some to put financial information in the CMMS, it works best for the failure analyst when this is done. The following financial information is suggested: production rates, parts usage rates, product output, unit cost, expected return on investment, downtime cost per unit of time, fully loaded hourly rates or a standard hourly rate for assessing repair costs, value added costs by department, cost accounting data used internally to evaluate costs, and overhead costs.

Provide Reports that Measure the Effectiveness of Reliability Deficiency Corrections

Once a solution to a reliability deficiency has been devised it has to be tested and tracked on two levels, operational and financial. To do this certain additional data must be available in the CMMS. Briefly, the following operational and financial data is required:

Operational Level

On the operational level, the following data should be stored, maintained, and retrievable from the CMMS:

Failure Rates - Failures/year

• MTTR - Mean Time To Restore after failure

• All Quality Parameters

• Quality Reject Rates of Parts and Supplies
  

• Parts Usage Data

• Unavailability of Parts from Stores

• Vibration Data
  

• Infrared Thermography Readings:
  - Production Rates
  - Operational Pluggage Rates
  - Rates of Loss of Heat Transfers
  - Flow Rates
   
  
• Slowdowns or Rate Reductions

Most of the above items have been discussed earlier in this paper or are self explanatory. However, a few items are worth mentioning again. For example, whenever an equipment or process failure occurs, the equipment or process that failed, the date and time of the failure, and its runtime to failure should be recorded. Additionally, the time to restore the equipment or process should be recorded. Using this input, the CMMS should tabulate the number of failures, the mean time between failures, and the mean time to restore after failure.

Other than to count incoming stores stock items to ensure against shortages and a visual inspection to spot obvious damage, many companies do not inspect stores stock items against specifications. Where possible and feasible this should always be done. In addition, there should be a provision in the CMMS for recording stock item quality defects.

In addition to quality defects of parts, quality parameters for all products manufactured at the facility should be retrievable through the CMMS. These parameters and how well the products are meeting the parameters should be stored in the CMMS or retrievable through an interface with other plant systems, for example the Quality Control system. This concept also applies to production data for such things as production rates, flow rates, pluggage rates, etc., and to non-destructive testing data for things like vibration data, infrared thermography readings, etc. This type data, which is useful to the failure analyst, does not have to reside in the CMMS, but should be retrievable from the CMMS.

Financial Level

On the financial level, trends of certain financial information should be tracked and maintained in the CMMS or retrievable from the CMMS through an interface with other plant systems. Trends of product output, unit cost, customer complaints, on-time deliveries, stores inventory, and raw material inventories (intermediate and final product) are very useful to the reliability professional for assessing the impact of reliability deficiencies as well as evaluating the solutions to these deficiencies.

Summary

A properly used and set up CMMS is a powerful tool for enhancing reliability efforts. A word of caution though, an improperly setup and used CMMS can have an equally negative impact on reliability. The trick is to do it right in a consistent manner. Like any tool, CMMS must be maintained and used properly to achieve the greatest potential possible.

William C. Worsham is a Senior Consultant for Reliability Center, Inc. Mr. Worsham has over 30 years experience in the field of Maintenance and Reliability program management. He has participated in and led teams in the development, design and implementation of three separate maintenance management systems. He has also participated in the design and implementation of specialized reliability inspection programs such as lubrication scheduling, vibration monitoring, instrument inspection and preventive maintenance. Mr. Worsham is a practitioner of root cause analysis in the field with his clientele as well as an educator. He can be contacted at 804/458-0645 or bworsham@reliability.com.

Charles J. Latino is president & founder of Reliability Center, Inc. Mr. Latino is a chemical engineer with a background in psychology and human factors engineering. He is a leader in the development of an integrated approach to achieving greater reliability in manufacturing and industrial systems and processes. He has served as consultant to many companies in the United States and abroad. He is the author of Strive For Excellence...The Reliability Approach. Mr. Latino can be contacted at 804/458-0645 or info@reliability.com