Appendix B Benchmarking and Best Practices
B e n c h m a r k Definitions:
a point of reference from which measurements may be made; b: something that serves as a standard by which others may be measured or judged; c: a standardized problem or test that serves as a basis for evaluation or comparison. a:
Verb a: to study (as a competitor's product or business practices) in order to improve the performance of one's own company.
Source: Merriam- Webster's 11 th Collegiate Dictionary.
The benchmarking method is very much based on performance measurement. Independent performance measures are chosen before the study and the best way of achieving this performance is analyzed through investigating companies that are achieving high levels of performance in these areas. Best practice is then identified based on the performance differentials. Benchmarking is investigative as, in most cases, it has no hypothesis of which practices are "best." It analyses performance based on specified performance measures and investigates the reasons for performance differentials. These differentials typically result in the identification of best practices. This form of study is not "blind" to the effects of other practices, within the 261
MaintenancePlanning and Scheduling
area under investigation, on performance. In this sense, the study is creating new knowledge of best practices. The disadvantage of this type of study is that the results are very general. It is useful in terms of indicating which practices seem to work, but it does not indicate the sequence in which these practices should be implemented, the time scale for impact to occur, implementation aspects of practices, and, in most cases, the extent to which practices impact the desired performance measures. It also focuses on companies in similar industries to facilitate ease of comparison. In an attempt to define the North American Manufacturing Industry "Best Maintenance Practices," a Benchmarking Survey of "Assessments of Maintenance Practices" of more than 200 manufacturing companies was performed by Life Cycle Engineering, Inc., a company specializing in Maintenance Engineering. The assessments evaluated the companies' practices in 21 separate, but interrelated, areas.. Following in Table B-l, are the compiled data from the benchmarking survey of maintenance effectiveness assessments. Doing some quick comparative analysis of the data in Table B-l, it is clear from the average scores that maintenance organizations do relatively well in the elements of Organizational Structure, Supervision, and providing craftsmen with adequate Facilities and Equipment. They also do well at Materials Management. These are all elements that do not necessarily require a well-integrated process. The average scores are much lower in the elements of Status Assessment, Master Planning, Work~Job Planning, Work Measurement and Scheduling and Coordination. These are elements that require a more sophisticated, well-designed process with the appropriate level of discipline to follow it.
B.2 BEST MAINTENANCE PRACTICE STANDARDS
The previous section on Benchmarking did indeed refer to practiceshow the art of maintenance is practiced. In order to differentiate that topic from Best Maintenance Practice (BMP) Standards, it is necessary to recognize that standards are neither arbitrary nor variable. The BMP Standards are established values of performance. You must measure your company's performance in order to compare them against the standards. Additionally, you must repeatedly measure your performance if you are to gauge your improvement. In addition to trending improvement, your performance measures are indicators of which policies are successful and which are not.
(.1 (~ (~ (~'~,. C ~) .i-,
~ ; mC
Appendix B: Benchmarking and Best Practices
0 uu -->' ~
c .0 ._V)
Maintenance Planning and Scheduling
For that reason, it is advisable not to institute a multitude of changes simultaneously lest you mask what works and what does not. Following are standards, together with their metrics, that are generally accepted as Best Practice Standards. As performance indicators, many have only a partial influence from "maintenance practices," while others are completely dependent on how the art of maintenance is practiced and how well it is performed. 9 Preventive/Predictive Maintenance (PM/PdM) Application of PM/PdM PM/PdM hours worked + Total hours worked > 30% PM/PdM Compliance PM/PdM Completed + PM/PdM Scheduled > 95% 9 Planned & Scheduled Maintenance Work Planned Work Planned Work (hours) + Total Work (hours) > 80% Planning Accuracy Number of WOs completed within esti mate (+ 15%) + Number WOs completed Schedule Compliance Scheduled Work Completed (hrs) + Scheduled Work (hrs) > 85% Labor Efficiency Estimated Hours + Actual Hours > 85% WO Aging Average age of Work Orders by Priority
(refer to Chapter 6) 9 Maintenance Labor (management, effectiveness and optimization [Maintenance Engineering]) WO Management Percentage Labor Covered by WO = 100% Resource Management Overtime + Total Time < 5% Labor Utilization Labor Productive Time (wrench time) + Labor Paid Time > 65% MTBF Mean Time Between Failures Increasing Trend MTTR Mean Time to Repair ~ Decreasing Trend 9 Total Plant Performance (includes maintenance effectiveness) Asset Availability Hours Asset Performs Primary Function + Hours Asset Scheduled to Perform Primary Function > 95% Asset Performance Number of units produced (per period) + N u m b e r of units scheduled to be produced > 95%
Appendix B: Benchmarking and Best Practices
Quality of Output
Number of units produced at quality standard + Total number of units produced > 95% Operational Equipment Effectiveness Asset Availability • Asset Performance x Quality of Output > 85%
9 Stores Management and Budget and Cost Control Inventory Stockouts Rare (e.g., less than 1 per month-dependent on total inventory size) Inventory Accuracy Items Cycle Counted as Correct + Total Stock Items Cycle Counted > 98% Repair Factor Total Maintenance and Repair Costs + Total Asset Replacement Cost Decreasing Trend Maintenance Costs Total costs are within + 2% of budget CPU Cost per Unit (produced) = Total Plant Costs + Total Units Produced Decreasing Trend Maintenance CPU Total Maintenance Costs + Total Units Produced ~ Decreasing Trend