Benchmarking Best Practices in Maintenance Management

1. 1Inventory what operations does today against the seven activities in the source: pre-start inspection, filling out work requests, completing work orders, recording breakdown or malfunction data, basic service such as lubrication, routine adjustments, and executing maintenance supported by central maintenance. Mark each activity as done by all, done by some, or not done, per area.
2. 2Measure the work-request base first. Count what percent of operations personnel generate work order requests. The survey bands are 90%+ = 4 pts, 75-89% = 3, 60-74% = 2, 40-59% = 1, under 40% = 0. Run the same count for facilities personnel. This is your baseline.
3. 3Set the limits before you assign tasks. The source says involvement depends on equipment complexity, operator skills, and union agreements. Write down which of those three constrains each task, so you hand over only what operators can legitimately and competently do.
4. 4Hand over the basic tasks in order of difficulty: inspection before start-up, request and breakdown-data recording, lubrication, then routine adjustments. Train and certify operators on each task. The survey gives 1 point per task operators are trained and certified to perform, so certification is the standard, not informal willingness.
5. 5Move work-order priority into a standing joint meeting. The survey scores weekly joint operations/maintenance priority-setting at 4 pts, daily at 2, maintenance deciding with minimal operations input at 1, and random emergency-driven priority at 0. Run the same structure with facilities.
6. 6Establish operator accountability for asset upkeep and performance. The survey scores this yes (4) or no (0), so make it an explicit, documented responsibility, not an implied one.
7. 7Use the PM coordination point as a communication test. Almost three-fourths of organizations have trouble coordinating PM with operations/facilities. The source puts the cause on communication: maintenance has not communicated the need, or operations has not been educated on the benefit. Explain to operations why each PM window is needed and what it prevents.
8. 8Redeploy the freed maintenance hours on purpose. The stated goal is to free maintenance resources for more advanced techniques. The next step is predictive work, for example vibration analysis on rotating equipment and thermography on electrical equipment. If the hours are not redirected, the involvement program has no payoff.

The source supplies the scoring frame, so this walks a plant through it. The scores below are illustrative inputs, not data from the source. A plant scores itself on Section 8. About 45% of operations personnel generate work requests (1 pt, the 40-59% band). Maintenance sets work-order priority with minimal operations input (1 pt). Operators are not responsible for asset upkeep (0 pts). Operators are trained and certified on only one task, inspections (1 pt). The plant compares its Section 8 total to the 22.2 average from more than 100 surveyed companies and sees a gap. It acts in this order: first, a weekly joint priority meeting, which moves that item from 1 pt to 4. Second, it names operators responsible for upkeep, moving that item from 0 to 4. Third, it trains and certifies operators on lubrication and routine adjustments, which adds points and returns maintenance hours. Those hours go to predictive work such as vibration analysis, which the source names as the next focus. The plant re-scores after each change and tracks the section against 22.2.

• If fewer than 40% of operations personnel generate work requests, then fix request-generation first. That item scores 0 pts, and every later step depends on operations reporting problems.
• If maintenance sets priorities with minimal operations input (1 pt) or by emergency (0 pts), then create a weekly joint meeting (4 pts). Daily joint meetings score only 2, so a daily meeting is a step toward weekly, not a replacement for it.
• If union agreements, equipment complexity, or operator skills block a task, then limit handover to inspection, request-writing, and data recording. Document the constraint and revisit it later. Do not push tasks past what the constraint allows.
• If operators are to perform lubrication, then train and certify them first. The source warns that unskilled lubrication staff produce substandard programs and excessive breakdowns despite having a lubrication program.
• If operations resists PM downtime windows, then treat it as a communication failure before an authority problem. Either the need was not communicated or operations was not educated on the benefit.

• Treating involvement as 'operators lubricate', and stopping there. The source says PM is more than lube routes and procedures. The tell: operator lube routes exist but inspections, adjustments, and data recording are never assigned.
• Handing tasks to untrained operators, as with the entry-level oiler problem. The tell: lubrication-related breakdowns continue even though a lubrication program is in place.
• Maintenance dictating priorities with minimal operations input, or ranking jobs by whoever shouts. The tell: random emergency-driven priority and constant disagreements between maintenance and operations, which the survey scores 0.
• Releasing maintenance hours with no plan for them. The tell: operators do more tasks but the PM/predictive workload and the proactive-to-reactive ratio (target 80/20) do not move.
• Leaving operations uninformed about why PM matters. The tell: PM windows are refused or missed because operations does not see the benefit. This is the communication gap behind the three-fourths coordination figure.

• ▸Area / equipment group: name the unit being assessed.
• ▸% operations personnel generating work requests: enter the percent and its band score (4/3/2/1/0).
• ▸% facilities personnel generating work requests: enter the percent and its band score.
• ▸Priority-setting method: weekly joint (4), daily joint (2), maintenance-led (1), or random (0). Record the current method.
• ▸Operator upkeep responsibility documented?: yes (4) or no (0), plus who owns it.
• ▸Tasks operators are certified for: list each (inspection, lubrication, routine adjustment, breakdown recording, etc.). One point per certified task.
• ▸Constraints per task: note equipment complexity, operator skill, or union agreement for each task not yet handed over.
• ▸Maintenance hours freed: estimate hours per week released by operator tasks.
• ▸Redeployment target: name where the freed hours go, e.g. vibration analysis or thermography.
• ▸Section total vs. 22.2 benchmark: sum the scores and compare to the surveyed average.

1. 1Audit how maintenance is described at your site. Compare it against the Preface's list of labels: necessary evil, cost, insurance, disaster repairing function, prima donnas. Record where each label appears (budget meetings, staff talk, capital requests). The labels show the attitude you start from.
2. 2Test the budget history. The Preface says maintenance has received 'little budgetary attention other than a nominal increase or decrease from year to year.' Pull the last several years of maintenance budgets. If every change is a small nominal adjustment with no link to asset condition or risk, management is treating maintenance as a cost to trim, not a function to manage.
3. 3Check resource provision before judging the craft workforce. The Foreword reports fewer than 4 of 8 hours per day on hands-on work, and as low as 2 hours in many organizations. It attributes this to craftsmen not being 'provided the necessary resources by management.' Measure your own hands-on time, then list the missing resources (planners, materials, tools, work orders) and who owns each gap.
4. 4Price the true cost of a repair. Add lost production to labor and materials. The Foreword gives a range of 2 to 15 times the repair cost, averaging 4:1, so a $10,000 repair is closer to $50,000. Re-cost your last five major repairs this way and show management the totals.
5. 5Apply the planned-versus-unplanned ratio. The Foreword estimates a cost ratio as high as 1:5, so a $100 planned job could cost $400 more if unplanned. Check whether you employ a job planner (only about one-third of organizations do). Use the gap to justify planner headcount as an investment.
6. 6Score management support with Chapter 1, Section 14 (Asset Care Continuous Improvement). Rate visible management support, organizational support, past support of improvement efforts, support for ongoing skills training, ROI focus, and the plant management–labor spirit of cooperation. Each item is scored 0 to 4 points. Have people from several levels score independently and compare the results.
7. 7Look at the enablers beyond the numbers. The benchmarking chapter lists leadership, motivated workforces, management vision, and organizational focus as hidden factors that allow best practices. Document specific evidence for each (decisions made, time given, resources released) so you are not relying on stated intent.
8. 8Turn the gaps into a short list of management commitments. Each commitment needs an owner, a resource, and a date. Cover at least visible support, training support, and a long-term asset-health measure kept alongside production targets. Review progress with senior management on a fixed schedule.

The Foreword's own repair example shows the method. A maintenance repair is booked at $10,000 in labor and materials. Counting lost production at the typical 4:1 ratio, the real cost is closer to $50,000. A plant manager who sees only the $10,000 treats the repair as a cost to cap. Seen at $50,000, the same event is a failure that earlier planned work could have prevented. Pair this with the planned-versus-unplanned ratio of up to 1:5, where a $100 planned job becomes a $400-more unplanned one. Then compare it with the labor picture: at about $20.00 per hour and only about 50% utilization, the site pays full price for a resource it uses at half capacity because management has not supplied planners, parts, or systems. Presented together, these three numbers move the discussion from 'maintenance spends too much' to 'management has not funded the conditions that make maintenance spending effective.' Section 14 then gives a scored baseline: if visible management support rates 'weak' (2 points) instead of 'strong' (4), that gap becomes the first item on the action list.

• If the maintenance budget changes only by a nominal increase or decrease each year, then treat maintenance as managed as a cost, and rebuild the request around lost-production-inclusive repair costs.
• If craftsmen are at or below 4 hands-on hours per day, then look first at management-provided resources (planners, materials, work orders, tools) before blaming individual performance.
• If lost production is excluded from repair cost reports, then re-cost major repairs with it (typical 4:1, range 2:1 to 15:1) before any budget discussion.
• If Section 14 shows strong stated support but weak past support or no tie to ROI, then treat the support as rhetorical and require specific resource commitments.
• If management support is weak or moderate, then set the enabler-building (visible support, training, vision) ahead of buying new tools or systems.

• Accepting management's stated support at face value. The tell: leaders endorse improvement but past efforts were poorly supported, training is cut, or the Section 14 'past support' score is Poor.
• Blaming craftsmen for low productivity. The tell: only 2 to 4 hands-on hours a day, while planners, parts, or work order systems are missing and no one owns fixing them.
• Reporting maintenance cost as labor plus materials only. The tell: repair costs look modest and nobody can say what a breakdown cost the plant in lost production.
• Using production or facilities methods to control maintenance. The tell: maintenance is managed by generic cost-cutting targets that the Preface says 'has not and will not be successful.'
• Overstocking or cutting inventory as a quick fix instead of managing it. The tell: stockrooms are inflated (carrying cost over 30% a year, over $300,000 on a $1 million inventory) or someone outside maintenance dictates stocking.

• ▸Labels in use: list how maintenance is described (necessary evil, cost, insurance, etc.) and where each was heard.
• ▸Budget history: record the last several years of maintenance budget and the size and reason for each change.
• ▸Hands-on hours per day: enter measured craftsman hours against an 8-hour day, and the number of resource gaps behind the shortfall.
• ▸Resource gaps and owners: list each missing resource (planner, materials, tools, work orders) and the manager accountable.
• ▸True repair cost: for each of your last five major repairs, enter labor plus materials, lost production, and the resulting ratio.
• ▸Section 14 scores: enter 0–4 points for visible support, organizational support, past support, training support, ROI focus, and cooperation.
• ▸Enabler evidence: for leadership, vision, workforce motivation, and organizational focus, note one concrete action as evidence.
• ▸Commitments: state each management commitment with its resource, owner, and due date.
• ▸Review date: set the next senior management review and the measure of long-term asset health to be shown there.

1. 1Baseline the reactive state with four numbers from your own records: overtime as a share of total hours worked (the surveyed average is 14.1%, which the source calls almost three times what it should be), the proactive-to-reactive work split, hands-on hours per craftsperson per day, and equipment downtime percentage. These four figures set your starting point and your targets.
2. 2Set the target ratio explicitly: 80 percent proactive (planned, scheduled, preventive) to 20 percent or less reactive. Other maintenance practices only become effective once you reach this ratio, so sequence everything else behind it.
3. 3Build the PM program first, because it is the lever that shrinks reactive work. Only about 22% of surveyed companies say PM meets their needs, and only 20 percent of U.S. companies believe their PM programs are effective, so assume yours has focus problems and audit it against actual failures.
4. 4Treat lube routes as the starting point only. Almost three-fourths of organizations have lube routes, yet many stop there. Extend PM to inspections and services (preventive and predictive) on equipment, and give the lubrication position real training, since an untrained entry-level lubricator produces breakdowns even when a lube program exists.
5. 5Install a job planner and plan work several weeks ahead. Only about one-third of organizations employ a planner. Planned work can cost as little as 1/5 of unplanned work (a ratio as high as 1:5), and planning several weeks out is what lets stores and procurement be optimized.
6. 6Align stores with planned work: raise service levels from the common sub-90 percent to 95 to 97 percent with 100 percent data accuracy, so planned jobs are not derailed by stockouts and technicians stop stockpiling personal stores.
7. 7Put a cost on downtime (utilities, idle production and operations personnel, and so on) and report it alongside maintenance cost. Downtime of 10 to 20% or more is the larger financial prize than the maintenance budget itself.
8. 8Plan a three-to-five-year transition and manage the cultural shift. The technical change is not difficult; the time goes into moving the organization from viewing maintenance as a necessary evil to treating it as a core business process. Track overtime and the proactive/reactive ratio quarterly to show progress.

The source's savings illustration starts with a maintenance labor budget of $3M annually. Studies show the materials budget is a similar amount, so total maintenance spend is roughly $6M. The source's claim is that if labor efficiency improves by about one-third (through less overtime, fewer outside contractors, and deferred maintenance done without added spend) and materials are also cut by one-third, total savings could approach $2M per year. These are expense dollars not spent, so they flow to profit. The savings are not headcount cuts and are not immediate: moving from a reactive to a proactive, best-practice organization takes three to five years. The larger payoff is availability. If equipment downtime runs 10 to 20% in a company, recovering that capacity outweighs the pure maintenance savings, and avoids buying redundant backup equipment that hurts return on net assets.

• If overtime is well above the low-single-digit range implied by the 14.1% average being roughly three times too high, then treat the shop as reactive and fix PM and planning before buying tools or software.
• If you have lube routes but breakdowns persist, then check lubricator skills and training and expand PM beyond lubrication before adding more routes.
• If the proactive share is below 80 percent, then defer other improvement practices that depend on it and put effort into PM and planning first.
• If stores service level is under 90 percent, then fix inventory controls to 95 to 97 percent with accurate data before scheduling work weeks out; otherwise planned jobs stall on parts.
• If someone claims downtime has no cost, then calculate it (utilities, idle personnel, lost deliverable product) before approving any maintenance budget cut.

• Equating PM with lube routes: the program is declared done once routes exist. Tell: lubrication is staffed by untrained entry-level people, other PM tasks are missing, and breakdowns remain high.
• Expecting immediate savings: leadership cuts the program when costs do not fall in year one. Tell: the three-to-five-year timeline is absent from the plan.
• Defining savings as headcount reduction: the culture turns defensive. Tell: savings are tracked only as heads cut, not overtime, contractor use, or deferred work.
• Having PM without planners or parts: PM tasks are written but not scheduled or supported. Tell: no job planner, stockouts above 10 percent, and technicians hoarding personal stock.
• Ignoring downtime cost: maintenance is judged only on its own budget. Tell: no downtime cost calculation exists, or backup equipment is purchased to cover failures.

• ▸Overtime % of total hours worked: record current value; compare against the 14.1% surveyed average and set a reduction target.
• ▸Proactive vs. reactive work split (%): record current split; target is 80/20 or better.
• ▸Hands-on hours per craftsperson per day: record actual (surveyed norm is under 4 of 8); note the main resource gaps.
• ▸PM coverage beyond lubrication: list equipment classes with inspection/service PM versus lube-only.
• ▸Lubricator skill and training status: note training level and any breakdowns traced to lubrication.
• ▸Planner count and planning horizon: record planners per hourly employees and how many weeks ahead work is planned.
• ▸Stores service level and data accuracy: record current % against the 95 to 97 percent and 100 percent targets.
• ▸Downtime % and cost per hour: include utilities, idle production personnel, and lost product.
• ▸Transition milestones (years 1 to 5): list the quarterly review dates and the cultural changes to be tracked.

1. 1Fix the formula before you measure. Wrench time = hands-on hours ÷ paid hours. Use the paid shift (8 hours) as the denominator, and count only direct, hands-on work in the numerator. Waiting for parts, waiting for instructions, and travel are not hands-on.
2. 2Set the baseline from the source benchmark. Wireman reports that craftsmen spend fewer than 4 of 8 hours on hands-on work (under 50%), and in the majority of organizations as little as 2 hours (25%). Treat 25-50% as the range you expect to find, then measure your own site rather than assuming.
3. 3Record every activity against a work order. The source says work must be tracked from request through completion, or the data is 'fragmented and useless.' Require each technician to log time by work order, and log the non-hands-on time (waiting for parts, waiting for instructions, travel) as separate entries.
4. 4Sort the non-hands-on hours into the three loss buckets in the definition: waiting for parts, waiting for instructions, and traveling. Total each bucket per week. The largest bucket is your first target.
5. 5Convert the loss to dollars. Multiply non-hands-on hours by the labor rate. Wireman uses about $20.00 per hour, and at 50% utilization that is $80 per technician per 8-hour day. Use your own rate, and show the figure to management in dollars.
6. 6Attack the parts bucket through stores service. The survey scores 'materials in stores when required' at more than 95% as top marks and under 50% as zero. Measure your stores hit rate, and add material requirements to PM tasks so parts are identified before the technician starts.
7. 7Attack the instructions bucket through planning. Wireman's targets are at least 80% of work planned weekly and at least 90% weekly schedule compliance, with a job planner doing the planning. Only about one-third of organizations have one. Put safety information, detailed instructions, material requirements, and labor estimates on the job package.
8. 8Re-measure on the same definition and check estimate accuracy. The survey asks whether PM actuals are reviewed annually against time and material estimates. Use that review to correct labor estimates, then compare the new hands-on percentage to the baseline.

Wireman's foreword gives the figures. A craftsman is paid for 8 hours a day at roughly $20.00 per hour, but spends fewer than 4 of those hours on hands-on work. 4 ÷ 8 is 50% wrench time, so about $80 of each $160 daily labor cost buys no direct work. In the majority of organizations the figure drops to 2 hours: 2 ÷ 8 = 25%, which leaves $120 of $160 unproductive per technician-day. Wireman is explicit that these craftsmen 'are not lazy' and are 'not provided the necessary resources by management.' So you do not start by pressuring technicians. You split the lost hours into parts waits, instruction waits, and travel, then fix the biggest bucket. The source ties the fixes to planning and scheduling: 80-90% of work planned weekly, 90% schedule compliance, and a planner role. It also notes that a reactive organization with good planning and scheduling may raise wrench time, though the passage cuts off before giving the target figure. The measurable result is the percentage rising from the baseline and the daily dollar loss falling.

• If your measured wrench time is under 50%, treat it as the benchmark-typical condition and look for resource failures (parts, instructions, planning) before looking at individual behavior.
• If waiting for parts is the largest loss bucket, then measure the stores hit rate first and work toward more than 95% of materials in stores when required, before changing anything else.
• If waiting for instructions is the largest bucket, then check whether work is planned: under 80% planned weekly or under 90% schedule compliance means planning is the constraint, and without a planner the first move is to create the role.
• If time entries are not tied to work orders from request through completion, then do not report a wrench-time number yet; fix the recording discipline first, because fragmented data cannot be analyzed.
• If PM tasks lack labor estimates and material requirements, then add them and compare actuals to estimates, so scheduled time reflects real hands-on time.

• Blaming technicians. The tell: the improvement plan contains pressure or monitoring but no changes to parts, planning, or job packages. Wireman states low hands-on time reflects missing resources, not laziness.
• Counting any time at the job site as wrench time. The tell: the figure comes out near 100% or does not match the 2-4 hour benchmark. Only direct hands-on work counts; waiting and travel are excluded.
• Incomplete work order recording. The tell: logged hours fall well short of paid hours, or some activities never appear on a work order. Without complete tracking from request to completion, the data is fragmented and useless.
• Fixing parts availability by overstocking. The tell: stores availability improves but inventory grows. The source warns carrying costs exceed 30% of item price per year, so a $1 million inventory costs over $300,000 annually. Keep inventories low while still avoiding stock-outs.
• Having a CMMS but using a fraction of it. The tell: the system holds data but it is inaccurate or incomplete. The source reports most U.S. companies use under 50% of CMMS capabilities, and it must be used completely with accurate data to be effective.

• ▸Technician / crew and week: identify who and when the data covers.
• ▸Paid hours (denominator): total paid hours for the period, for example 8 per day.
• ▸Hands-on hours (numerator): hours of direct, hands-on work recorded against work orders.
• ▸Wrench time % : hands-on hours ÷ paid hours, to compare against the 25-50% benchmark range.
• ▸Hours waiting for parts: time lost to materials not in stores or not staged.
• ▸Hours waiting for instructions: time lost for lack of job plan, detail, or direction.
• ▸Hours traveling: time lost moving between jobs, stores, or locations.
• ▸Labor rate ($/hr): your own rate (source example: about $20.00).
• ▸Weekly cost of non-hands-on time: (paid hours − hands-on hours) × labor rate; show the total to management.
• ▸Planning and stores indicators: % of work planned weekly (target at least 80%), schedule compliance (target at least 90%), and % of materials in stores when required (target above 95%).

1. 1Size the stakes first. Materials run 40% to 60% of the maintenance budget, second only to labor. Pull your site's annual maintenance spend and multiply by both ends of that range. That gives the dollar band you are managing and tells you whether the work deserves close scrutiny.
2. 2Settle who owns the stocking decision. In about half of organizations, someone outside maintenance decides what to stock, how many, and how many to issue. Write down who decides each of the three today. If it is not maintenance, move at least the what and how-many decisions to maintenance, because maintenance is the customer that feels the stock-out.
3. 3Charge every issue to a work order. Use the survey's scale: 100% of materials charged to a work order number at issue scores 4, 75% scores 3, 50% scores 2, 25% scores 1, and under 25% scores 0. Measure your current percentage, then close the gap. Without work-order-level issue data you cannot see which jobs consume which parts.
4. 4Tie each work order to an equipment/asset number. Parts demand only becomes stockable information when it is linked to the asset that consumed it. Check what share of your work orders carry an asset number, and treat anything below 100% as a gap in the parts history.
5. 5Compare estimated to actual materials on completed work orders. Only about 10% of organizations do this. For each closed job, record the materials planned versus the materials actually drawn, and review the variances. Frequent unplanned issues mean the job plan or the stocked list is wrong.
6. 6Cost every stock-out at the full production-loss multiple, not at the part price. Repair labor plus materials understates the cost. Lost production runs 2 to 15 times the repair cost, and the average is 4:1. Record each stock-out delay with the downtime it caused, so storeroom decisions are judged against that cost.
7. 7Cost the other side of the ledger, which is carrying cost. Inventory carrying cost exceeds 30% of item price per year, so a $1 million inventory costs over $300,000 a year. Stock to a satisfactory service level and avoid stock-outs, but do not overstock to compensate for poor control.
8. 8Track the symptoms monthly: craft hands-on time (fewer than 4 of 8 hours is typical), overtime (U.S. average 14.1%, almost three times what it should be), and work orders delayed waiting for materials. Falling overtime and rising hands-on time are your evidence that availability is improving.

The source's own figures can be run as a two-sided test. Suppose a breakdown repair costs $10,000 in labor and materials. At the 4:1 average, the real cost with lost production is about $50,000. If the repair stalls because a $200 part is not in the storeroom, the delay is judged against that $50,000 exposure, not against $200. Now take the opposite fix, overstocking. A $1 million inventory costs over $300,000 a year to carry (30%+). The tradeoff is therefore not 'more stock is safer.' Each added dollar of stock costs about 30 cents a year, and each avoided stock-out saves a multiple of the repair cost. The expert move is to identify which items sit on the critical assets and stock those to a service level. Then use work-order issue history (100% of issues charged to a work order, 100% tied to an asset) to find the rest. Reduce the remaining inventory while watching stock-outs, overtime, and hands-on time.

• If maintenance does not decide what to stock, how many, and how many to issue, then move those decisions to maintenance first. A support function should not run its customer's business.
• If less than 100% of material issues are charged to work orders, then fix issue discipline before adjusting stock levels. Any stocking decision made on partial data is a guess.
• If a stock-out delays a repair, then cost it at the repair cost times the production-loss multiple (average 4:1, range 2 to 15), not at the part price.
• If the storeroom is being overstocked to cure shortages, then price the carrying cost (over 30% a year) and stock only to a satisfactory service level.
• If overtime is well above the 14.1% U.S. average or craft hands-on time is under 4 of 8 hours, then check materials first. Craft are not lazy; they are not being given the resources to do the job.

• Fixing shortages by overstocking. The tell is a growing inventory value with stock-outs still occurring, and carrying cost above 30% a year that nobody has calculated.
• Costing a stock-out at the part price. The tell is that a missing low-cost item is treated as minor even though the delayed repair carries a 4:1 average production-loss multiple.
• Leaving stocking authority outside maintenance. The tell is that someone else sets what is stocked, how many, and how many are issued, and maintenance is blamed for shortages it did not control.
• Issuing parts without a work order or asset number. The tell is a low materials-charged-to-work-order percentage (below 100%) and no way to say which asset consumed which parts.
• Never comparing planned to actual materials. The tell is repeated unplanned trips to the storeroom mid-job and no variance review, which only about 10% of organizations do.

• ▸Annual maintenance budget and materials share (%): enter total spend and the share spent on materials; compare with the 40-60% range.
• ▸Stocking authority: record who decides what to stock, how many, and how many to issue (maintenance or other).
• ▸% of material issues charged to a work order: measure and score against 100/75/50/25/<25%.
• ▸% of work orders tied to an asset number: measure to confirm parts demand can be traced to equipment.
• ▸Planned vs. actual materials variance: for closed jobs, list planned cost, actual cost, and the cause of any difference.
• ▸Stock-out log entry: date, part, asset, hours the job waited, repair cost (labor + materials).
• ▸Production-loss multiple: enter your site multiple (average 4:1, range 2-15) and compute the true stock-out cost.
• ▸Inventory value and annual carrying cost: enter inventory value and multiply by 30%+ to get the yearly carrying cost.
• ▸Symptom trend: record monthly overtime %, craft hands-on hours per 8-hour day, and number of jobs delayed for materials.

1. 1Name the key assets, processes and facilities whose availability you will track. The survey scores 'overall equipment availability calculated on key assets, processes, and facilities' in bands: 90% or more scores 4, 60 to 89% scores 3, 30 to 59% scores 2, and under 30% scores 0. Coverage is the first thing to fix, and the passages do not define a coverage-first cutoff beyond these bands.
2. 2Make every job traceable to an asset. Score yourself on the share of work orders tied to an equipment/asset number, the share of jobs covered by a work order, and the share of man-hours reported to a work order (100% scores 4, 75% scores 3, 50% scores 2). Without this, uptime and failure counts cannot be attributed to an asset.
3. 3Record required downtime on each work order. The survey awards a point for tracking it, alongside craft hours, materials and requestor. Downtime per asset is the raw input for any availability figure.
4. 4Capture failure cause, not just the repair. Score what share of an asset's work order history accurately tracks causes of failure and clearly identifies failures (90% or more scores 4, under 40% scores 0). The source warns that chronic problems stay hidden, and that only accurate records reveal them and allow the root cause to be solved using maintenance and financial data together.
5. 5Run failure analysis with a defined tool (why tree, fishbone, Pareto) so the work is accurate and consistent. Use the results to adjust the PM/PdM program through RCM. RCM needs complete, accurate asset data: 90% or more of assets scores 4.
6. 6Match inspection technique to criticality. Predictive maintenance covers vibration analysis, thermography, oil analysis and sonics, scored 4 for critical and non-critical assets, 2 for critical only, and 0 for none. Tie the data to the CMMS (integrated 4, interfaced 3) so findings generate PM or corrective work orders, and schedule predictive work in the weekly plan.
7. 7Put the cost in front of decision-makers. Price each outage as repair cost plus lost production (the source's average is 4:1). Communicate this to everyone, and use availability in operational decisions (90% or more of the time scores 4).
8. 8Hold a joint operations/maintenance priority meeting weekly (scores 4; daily scores 2). Include maintenance in production scheduling meetings, and check that the share of emergency/urgent work orders stays under 20% (scores 4).

The source's own example: a maintenance repair costs $10,000 in labor and materials, but the real cost is closer to $50,000 once lost production is counted. The gap, roughly $40,000, is lost production, so the outage costs about five times the invoice. That matches the source's statement that the average ratio is usually 4:1. Suppose the repair sits on a work order tied to an asset number, with required downtime recorded and the failure cause logged. Then it can feed a Pareto of failures for that asset, and the root cause can be pursued with a why tree. A manager who sees only the $10,000 cuts the inspection that would have caught the fault. A manager who sees $50,000 compares it with the cost of vibration, thermography or oil analysis on that asset. The source adds that downtime is more costly than maintenance expenditure, and that any time gained in repairs becomes reduced downtime or increased availability.

• If the emergency/urgent share of work orders is 20% or more, treat availability as a planning problem first. Fix priority-setting and PM coverage before buying more monitoring. Under 20% is the top survey band.
• If a repair is justified only against its labor-and-materials invoice, restate it at the lost-production-inclusive cost (average 4:1) before approving or rejecting preventive or predictive spend.
• If asset history lacks failure causes or equipment numbers, do not compute or compare MTBF-style figures yet. Fix the records first, since chronic problems stay hidden without them.
• If maintenance reports to production, check whether supervisor pay is based only on throughput. If so, expect maintenance to turn into firefighting, unless availability or life-cycle cost targets are added to their compensation.
• If maintenance reports to engineering and a project is running late, check whether PM labor is being diverted to project work. The source warns that existing equipment then suffers from lack of maintenance.

• Counting only the repair invoice. Tell: maintenance cost is cut or PM deferred without any lost-production figure in the business case.
• Availability is calculated on a handful of assets, or not at all. Tell: the survey item scores in the lower bands (under 60% coverage) and operational decisions ignore availability.
• Failure records say what was replaced but not why. Tell: repeat failures on the same asset, no why-tree or Pareto output, and chronic problems found only at breakdown.
• Predictive data sits outside the work system. Tell: vibration, thermography or oil analysis findings are not in the CMMS, do not generate work orders, and are absent from the weekly schedule.
• The reporting structure or incentives pull maintenance off PM. Tell: production-centric 'fix it when it breaks' behavior, or PM labor repeatedly diverted to projects.

• ▸Asset / process / facility name and asset number: confirm every work order can be tied to this number.
• ▸Availability coverage: record whether availability is calculated for this asset, and the share of key assets covered (90%+, 60-89%, 30-59%, under 30%).
• ▸Downtime events logged: record each event with its required downtime from the work order.
• ▸Failure cause recorded: enter the cause per event and the share of asset history with accurate cause data.
• ▸Repair cost (labor + materials): enter the direct cost from the work order.
• ▸Lost production cost: enter the production lost during the outage, then add it to repair cost for total event cost.
• ▸Cost ratio: divide total event cost by repair cost and compare with the 4:1 average cited in the source.
• ▸Detection method in place: record vibration, thermography, oil analysis, sonics or condition monitoring, and whether critical only or all assets.
• ▸Linked to CMMS / work orders generated: state integrated, interfaced or not connected, and whether findings create work orders.
• ▸Emergency/urgent share of work orders and next action: record the percentage (under 20% is the top band) and name one corrective action with an owner.

1. 1Fix the denominator first. For a plant, use total maintenance cost as a percentage of total process, production, or manufacturing cost, calculated as a total and not per production unit. For a facility, use maintenance cost per square foot maintained. Both are described as stable, which makes trending easy.
2. 2Build the numerator from complete cost capture. Pull labor, materials, and contractor spend by asset, and also energy, quality, and stores and purchasing costs. The survey scores 4 points only when maintenance costs are tracked for all assets, not just key assets, and when all contributing costs are available for analysis.
3. 3Calculate the ratio monthly and compare it with the monthly maintenance budget and the year-to-date total. Trend it. The source says that if the percentage fluctuates, you examine the efficiency and effectiveness of maintenance to find the cause.
4. 4Measure labor utilization as the first waste test. Record hands-on hours per craftsperson per day out of 8. The source benchmark is fewer than 4 hours, and as low as 2 in most organizations. At roughly $20.00/hour, 50% utilization is a direct cost waste.
5. 5Test the planning lever. Compare planned and unplanned work, since the cost ratio may be as high as 1:5 (a $100 planned job versus about $500 unplanned). Check whether a planner exists, since only about one-third of organizations employ one, and check the share of work orders spent on preventive versus emergency and normal maintenance.
6. 6Close the estimate loop. Report actual time, material, and downtime on every completed work order. Run a planning effectiveness report showing the jobs closed out 20% over or under the labor or material estimates, by planner and supervisor.
7. 7Score yourself against the benchmark. In the survey, Financial Optimization averages 23.9 and Maintenance Reporting 13.4, out of a 640-point survey with a 366 average total. Treat low scores in cost tracking and reporting as the first fix, because you cannot manage the ratio without data.
8. 8Assign an owner. Name a dedicated individual or team to analyze financial costs, and make financial information available on demand (4 points) rather than monthly (1 point).

The source's own figures give a worked waste calculation. A craftsperson paid about $20.00/hour is on the clock 8 hours, but hands-on time is under 4 hours, and often 2. At 4 hours of use the resource runs at 50%, so about $80 of an $160 day is non-productive. The cause is not laziness; the source says craftsmen are not given the resources to do the tasks. Next, apply the planning ratio: a $100 planned job could cost as much as $500 unplanned, so moving 10 jobs from unplanned to planned would save up to $4,000 on the source's 1:5 ratio. Then check the macro indicator. If maintenance runs at a stable percentage of production cost and it suddenly moves, you examine maintenance efficiency and effectiveness for the cause. The source also notes that one-third of the 1979 US spend of over $200 billion was wasted, and that the waste ratio likely still holds. Treat that as the baseline assumption for an unaudited program.

• If maintenance cost as a percentage of production cost (or cost per square foot) moves from its stable trend, then audit efficiency and effectiveness before touching the budget.
• If hands-on time is under 4 hours per day, then fix resource availability (planning, parts, scheduling) before considering headcount cuts.
• If you have no planner, then add planning first, because planned work may cost one-fifth of unplanned work and planning is among the largest savings areas.
• If costs are tracked only for key assets, then extend tracking to all assets and include contractors, energy, and stores costs before you judge cost-effectiveness.
• If estimated and actual costs differ by more than 20% on many closed work orders, then address estimating and planning quality by planner and supervisor.

• Using per-production-unit maintenance cost, which swings with output volume. The tell: the ratio moves whenever volume moves. The source says to use total calculations.
• Treating maintenance as a cost to cut, with nominal year-to-year budget changes. The tell: budgets rise or fall by a flat percentage with no link to waste sources.
• Blaming craftsmen for low productivity. The tell: low wrench time alongside missing parts, jobs, or planning. The source says resources are the cause.
• Tracking costs only for key assets or leaving out contractors and stores. The tell: the totals fail to reconcile and you cannot explain a change in the percentage.
• Applying generic production or facilities cost controls to maintenance. The tell: controls are in place but waste persists. The source says this has not worked and will not.

• ▸Period: month and year-to-date window being measured.
• ▸Total maintenance cost: labor + materials + contractors, plus stores and purchasing costs, for the period.
• ▸Denominator: total process/production/manufacturing cost (plants) or square feet maintained (facilities).
• ▸Maintenance cost ratio: numerator divided by denominator, with the prior 12-month trend and the variance.
• ▸Budget versus actual: monthly maintenance budget and year-to-date total.
• ▸Hands-on hours per craftsperson per day (out of 8) and hourly rate: use to estimate the labor waste.
• ▸Planned vs. emergency/normal work orders: counts and share; planner in place (yes/no).
• ▸Work orders closed 20% over or under estimate: count and percent by planner/supervisor.
• ▸Cost tracking coverage: percent of assets with maintenance costs tracked, and named owner of financial analysis.

1. 1Name the two profit levers and size each separately: expense reduction (labor and materials) and capacity gain (availability and efficiency). Treat the capacity lever as the larger one. The source says the pure maintenance cost contribution is 'dwarfed' by the savings from added asset capacity and efficiency.
2. 2Estimate the waste pool. Take the annual maintenance labor budget, then apply the source's estimate that 1/3 of maintenance spend is wasted. Assume a materials budget of similar size. A $3M labor budget gives about $1M labor savings, and the same 1/3 on materials brings the total near $2M per year. State this as a ceiling to approach over time, not a first-year number.
3. 3Decide where labor savings come from before promising headcount cuts. The source lists reduced overtime, reduced outside contractors, and deferred maintenance done without added spend. Tag each savings line to one of these.
4. 4Trend two stability indicators: maintenance cost as a percentage of total process/production/manufacturing cost (total, not per unit), and maintenance cost per square foot maintained for facilities. When either fluctuates, examine maintenance efficiency and effectiveness to find the cause.
5. 5Calculate a real cost of downtime. Include utility costs, idle production/operations personnel, late-delivery cost, and overtime to make up lost production. Anchor the figure on the true cost: lost sales for product not made on time. Publish it so operating decisions use it.
6. 6Measure efficiency, not just uptime. Compare running speed and output to design speed and capacity, because equipment can run at only 50 to 60% of capacity while uptime looks fine. Efficiency losses are always greater than pure downtime losses and are rarely measured.
7. 7Capture cost and failure data at equipment level, in dollars and in hours by craft, through work-order control (hours × labor rate, summed over the period). Without it you cannot budget, run life-cycle costing, or find chronic problems. Record small events too, such as a repeated breaker reset.
8. 8Set a 3–5 year horizon for a reactive-to-proactive transition. Avoid dependence on technical change alone. The source says the work is cultural: moving from seeing maintenance as a cost to treating it as a core business process.

The source's savings arithmetic: a company has a $3M annual maintenance labor budget. Applying the 1/3-waste estimate gives roughly $1M saved over time. The materials budget is typically a similar $3M, and a 1/3 reduction there brings total savings close to $2M per year. These are expense dollars not spent, so they become profit dollars. The savings need not come from headcount; they can come from less overtime, fewer outside contractors, or deferred maintenance done without extra spend. Capacity is the bigger lever. If equipment downtime runs 10 to 20% or more, product that could reach the market is lost. A company that buys backup or redundant equipment to cover that downtime lowers its return on net assets and its investment rating. In the source's real case, a compressor overhaul paid back in 28.1 days after restart, and $5.4M of increased production was realized in the next 12 months.

• If maintenance cost percentage or cost per square foot moves from its stable trend, then investigate maintenance efficiency and effectiveness before changing the budget.
• If someone says downtime has no cost, then force the calculation: utilities, idle personnel, late delivery, overtime to recover schedule, and lost sales.
• If you are tempted to buy redundant equipment to cover downtime, then first test whether availability and efficiency gains on the existing assets close the gap, since redundant purchases hurt return on net assets.
• If uptime goals are met but output is short, then check running speed and capacity against design. Efficiency losses exceed downtime losses and often go unmeasured.
• If the maintenance organization is reactive, then forecast savings over 3–5 years, not the current budget cycle, and fund the cultural change.

• Counting only labor and material savings and ignoring capacity gains. The tell: the profit case is a small share of what the downtime cost would show.
• Assuming the $2M savings means layoffs, which sparks resistance. The tell: the plan lists headcount and omits overtime, contractors, and deferred work.
• Tracking uptime alone. The tell: availability looks good while equipment runs at 50 to 60% of design speed or capacity.
• Budgeting without equipment-level cost history. The tell: a target like 'cut labor 10%' appears with no data on how labor was allocated this year, or labor is recorded only in dollars with no hours by craft.
• Expecting immediate returns from a reactive shop. The tell: the benefit is judged in the first year, and the effort is abandoned before the 3–5 year culture shift lands.

• ▸Annual maintenance labor budget ($): enter the current figure.
• ▸Annual maintenance materials budget ($): enter the current figure, expected to be similar to labor.
• ▸Waste estimate (%): start with the source's 1/3 and adjust only with evidence.
• ▸Savings sources: list overtime, outside contractors, and deferred maintenance, with a dollar target for each.
• ▸Maintenance cost as % of total production cost: record the current value and trend.
• ▸Maintenance cost per square foot maintained: record the current value and trend (facilities).
• ▸Downtime % and cost per downtime hour: sum utilities, idle personnel, late delivery, overtime, and lost sales.
• ▸Running speed vs. design speed/capacity (%): record separately from uptime.
• ▸Time horizon and owner: state the 3–5 year transition plan and who is accountable.

1. 1Set the coverage target at 100% for all three measures the audit scores separately: man-hours reported to a work order, materials charged to a work order number at issue, and total jobs performed covered by a work order. A 75% result on any of them scores 3 of 4 points, not full marks. Measure all three, because one can look healthy while another leaks.
2. 2Run every job through the same five-step path: review the submitted work, approve it, plan it, schedule it, record the completed work. Name one person for each step. Work that skips a step is untracked work, and the book states that data is lost and true analysis can never be performed.
3. 3Tie every work order to an equipment/asset number. The audit scores the percentage of work orders tied to an asset (100% = 4 points). Without the asset field, history cannot be analyzed by equipment and failure analysis has nothing to work from.
4. 4Configure the work order to capture the categories the audit asks you to track, one point each: required downtime, required craft hours, required materials, and the further categories on that list. Record estimates at planning time and actuals at close-out so you can compare them. Only about 10% of organizations compare work order labor and material estimates with actual results.
5. 5Assign a qualified person to check completed work orders for quality and completeness before closing them. Set the target at 100% checked (4 points). Track the share of orders closed within eight weeks of the request date, again targeting 100%, so work does not sit open.
6. 6Set planning and scheduling targets: at least 80% of all maintenance work planned on a weekly basis, and weekly schedule compliance of at least 90%. Report both every week. Planned work can cost as little as 1/5 of unplanned work, so a $100 planned job could save as much as $400 if it were unplanned.
7. 7Keep the data complete and accurate in the CMMS and use its full capability. The source cites a U.S. survey in which most companies used less than 50% of their CMMS capabilities, and says a CMMS works only if it is used completely and its data is fully accurate. Make the work order history available for analysis (target 100% of orders), and add a craft backlog view so staffing decisions rest on projected work per craft.
8. 8Control the emergency mix and the PM feed. Target fewer than 20% of work orders opened as emergency or urgent (4 points). Target 80-100% of work orders generated from preventive maintenance inspections (4 points). Review both figures each period as a sign of how reactive the operation is.

The book's own audit (Section 3, Maintenance Work Orders) works as a scorecard. Suppose a plant finds that about 75% of man-hours are reported to work orders (3 pts), 50% of materials are charged to a work order number at issue (2 pts), and 75% of jobs are covered by work orders (3 pts). Every work order is tied to an asset number (4 pts), but 35% are opened as emergency or urgent (2 pts, the 30-39% band). The plant then sees a pattern: the missing materials and the missing hours mean the cost of a job can never be fully known. The plant closes the materials gap first by requiring a work order number at the storeroom before any issue. It then moves the jobs and man-hours toward 100% by refusing time entries without a work order number. Next it adds a quality check on every completed order, and it records downtime, craft hours and materials as tracked categories. Over time the emergency share should fall below 20% as planning reaches the 80% weekly target and schedule compliance reaches 90%. The audit is then repeated to confirm the scores moved.

• If any of the three coverage measures (man-hours, materials, jobs) is below 100%, then fix that gap before adding analysis or reports, because fragmented data is useless for analysis.
• If emergency or urgent work orders exceed 39% of the total, then treat the work order system as reactive. This band scores 0 points, so move effort into planning and PM-generated work first.
• If weekly planned work is under 80% or schedule compliance is under 90%, then check that the review, approve, plan, schedule and record steps each have an owner before changing software.
• If the CMMS is installed but you use less than half its capabilities, then do not buy more modules. Complete data entry and accuracy first, because the benefit comes from complete use.
• If work orders carry no asset number or no estimates and actuals, then do not attempt failure analysis or performance monitoring. Fix the fields first, since you cannot compare what you did not record.

• Partial coverage: only major jobs get work orders. The tell: the sum of man-hours on work orders is well below payroll maintenance hours, or the storeroom issues parts with no work order number.
• Work orders opened but never analyzed: the system collects data but nobody compares estimates to actuals or does failure analysis. The tell: breakdowns end in a part change with no recorded cause, as in the majority of companies the book describes.
• Orders left open or unchecked: no qualified person checks quality and completeness, and orders drift past eight weeks. The tell: a large backlog of aged open orders and closed orders with blank or implausible fields.
• CMMS purchased but underused: less than 50% of capabilities in use and data of doubtful accuracy. The tell: staff keep side spreadsheets or paper logs, so the system cannot answer basic questions.
• Everything flagged as emergency: priorities become meaningless and planning collapses. The tell: more than 30-39% of orders are urgent, overtime stays high (the book cites a U.S. average of 14.1% of time worked), and planned-work and schedule-compliance targets are missed.

• ▸Man-hours reported to a work order (%): total work order hours divided by total maintenance hours; target 100%.
• ▸Materials charged to a work order at issue (%): issues carrying a work order number divided by all issues; target 100%.
• ▸Jobs covered by work orders (%): jobs with a work order divided by all jobs performed; target 100%.
• ▸Work orders tied to an asset/equipment number (%): target 100%.
• ▸Emergency/urgent work orders (%): below 20% scores full points; above 39% scores zero.
• ▸Orders checked for quality and completeness, and closed within 8 weeks (%): record both figures; target 100% each.
• ▸Weekly planned work (%) and schedule compliance (%): targets at least 80% and at least 90%.
• ▸Work orders generated from PM inspections (%): target 80-100%.
• ▸Categories tracked per work order: list each (downtime, craft hours, materials, others); 1 point each; owner and due date for every gap.

1. 1Baseline first. Score your planning and scheduling against the Section 4 audit questions: percent of non-emergency work orders completed within four weeks (90% or more scores 4 points), who plans the work, how often schedules are issued, and how often the maintenance-operations scheduling meeting is held. The low-scoring answers show where to start.
2. 2Assign a dedicated maintenance planner. The audit scores a dedicated planner at 4 points, a maintenance supervisor at 2, and a craft technician or anyone else at 0. Only about one-third of organizations employ a planner, so this is the largest single gap to close.
3. 3Define a four-part plan on every planned work order: craft required, materials required, tools required, and specific job instructions or a job plan. The audit gives 1 point per element, so a complete plan scores 4. Anything missing is a predictable delay.
4. 4Train the planner formally. The audit ladder runs from no program (0) through on-the-job only (1), one month of 1-on-1 on-the-job training (2), a written program (3), to public seminars (4). Cover work order planning and execution, material planning, scheduling practices, and project planning, which earns 1 point each.
5. 5Issue the job schedule weekly. Weekly issuance scores 4 points; biweekly 3; every 3 to 6 days 2; daily 1. Build the schedule from planned work orders whose materials and tools are confirmed.
6. 6Hold the maintenance and production/facilities scheduling meeting weekly, which is the top score. Use it to agree on when equipment will be available for the planned jobs.
7. 7Track delays caused by poor plans: the percent of planned work orders delayed by poor or incomplete plans. Under 10% scores 4; over 50% scores 0. Feed each delay back to the planner as a correction to the plan.
8. 8Compare estimates to actuals for labor and materials on closed work orders, and watch overtime. Only about 10% of organizations monitor performance this way, and overtime averages 14.1% of time worked, almost three times what it should be.

Wireman's survey statistics show why planning pays. Craftsmen spend fewer than 4 of 8 hours per day on hands-on work, and in many organizations as little as 2 hours. The cause is that management does not provide the resources they need, not laziness. At roughly $20.00 per hour, a craftsman used at 50% wastes about half of that paid hour. Only about one-third of organizations employ a job planner, even though experts see planner use as one of the largest cost-saving opportunities in maintenance. Planned work versus unplanned work may have a cost ratio as high as 1:5. In practice, this means a job that costs $100 when planned can cost up to $500 when handled reactively. An organization scoring itself on Section 4 might find that craft technicians plan their own jobs (0 points), schedules are issued daily (1 point), and plans cover only craft and materials (2 points). Moving to a dedicated planner, weekly schedules, and a full four-element plan would lift each of those items to 4 points.

• If craft technicians or supervisors are planning their own work, then appoint a dedicated planner first, because the audit scores that arrangement 4 points against 0 to 2.
• If planned jobs are delayed because parts or tools are not ready, then check which of the four plan elements (craft, materials, tools, job instructions) is missing from the plan and fix the template, not the technician.
• If more than 20% of planned work orders are delayed by poor plans (3 points or lower on the audit), then pause scheduling expansion and tighten plan quality and planner training.
• If overtime sits well above the 14.1% average, then treat it as a sign of reactive work and push more jobs through planning and weekly scheduling, along with preventive maintenance.
• If the scheduling meeting with operations is irregular, then set it weekly and align it with a weekly schedule issue, because both score highest at weekly.

• Naming a planner who also supervises or does craft work. The tell is that plans are written after the job starts and the planner is regularly pulled into breakdowns.
• Plans that list only craft hours. The tell is crews stopping mid-job to hunt for materials, tools, or instructions, which shows up as low hands-on time.
• Scheduling without operations. The tell is jobs on the weekly schedule that cannot proceed because equipment is not released.
• Never comparing estimates with actuals. The tell is no feedback on labor and materials, so plan accuracy never improves (the case in roughly 90% of organizations).
• Untrained planners. The tell is no written program or mentoring, and a high share of planned jobs delayed by incomplete plans.

• ▸Work order number and equipment: identify the job and asset.
• ▸Craft required: list the trades and number of people.
• ▸Materials required: list parts and confirm each is in stock or on order.
• ▸Tools required: list special tools and equipment and confirm availability.
• ▸Job instructions or job plan: attach or reference the step-by-step procedure.
• ▸Estimated labor hours and material cost: record the estimates to compare with actuals later.
• ▸Operations release / downtime window: note the date and time agreed in the weekly scheduling meeting.
• ▸Scheduled week and issue date: record when the job appears on the weekly schedule.
• ▸Actual hours/cost and delay cause: after closing, record actuals and any delay due to a poor or incomplete plan.

1. 1Baseline the program before you change it. Score yourself against the survey facts: only about 22% of surveyed companies say their PM satisfies their needs, and over three-fourths show a need for major improvement. Record whether you sit inside the 22% or the three-fourths.
2. 2Define PM broadly in writing. PM covers inspections, lubrication, and adjustments, not lube routes alone. Almost three-fourths of organizations have lube routes and procedures, and many stop there. List every PM activity type you run and flag any equipment that gets lubrication only.
3. 3Fix the lubrication position separately. The oiler or lubricator is generally an entry-level job, and weak skills and training produce substandard lubrication and excessive breakdowns even when a program exists. Define the skills, train the person, and check the work.
4. 4Make PM inspections feed the work order system. The assessment awards its top score when 80-100% of work orders are generated from PM inspections, and 0 points below 20%. Count your current share and set 80% as the target. PM cannot be measured or controlled without a working work order system.
5. 5Plan and schedule every PM job. Each should name the craft, materials, tools, and specific job instructions, and be assigned to a dedicated planner where possible. Planned work can cost as little as 1/5 of the same job unplanned, so a $100 planned job can save up to $400.
6. 6Agree the PM schedule with operations/facilities before issuing it. Almost three-fourths of organizations have trouble coordinating PM with operations/facilities. The cause is communication: either maintenance has not explained the need, or operations has not been taught the benefits. Hold a short session on why each PM window is needed.
7. 7Run failure analysis on breakdowns and feed the findings back into PM tasks. Only about 10% of companies perform any failure analysis, and most just change parts.
8. 8Track overtime as the outcome check. US maintenance averages 14.1% overtime, almost three times what it should be, and high overtime signals a reactive operation. Falling overtime shows PM is shifting work from emergency to planned.

The source's own survey figures show the pattern. About 22% of surveyed companies say PM meets their needs, and over three-fourths need major improvement. Almost three-fourths have lube routes and procedures, which looks positive, but the source warns that many organizations treat lubrication as the whole of PM and stop developing the program. Take a plant in that position. It has a lube route run by an entry-level oiler with little training, so breakdowns stay high despite the program. Following the source, the plant first widens PM to include inspections and adjustments. It trains the lubricator. It then routes inspection findings into work orders, aiming for 80-100% of work orders generated from PM inspections (the top score in the assessment). It plans each job with craft, materials, tools, and instructions, because a $100 planned job can save up to $400 against the unplanned version. It briefs operations/facilities on the benefits, since coordination fails in almost three-fourths of organizations. Overtime, 14.1% on average, is the check on whether the work is moving from reactive to proactive.

• If your PM consists mainly of lube routes, then you have a start, not a program: add inspections and adjustments before declaring PM done.
• If breakdowns stay high despite a lubrication program, then check the lubricator's skills and training first.
• If less than 20% of work orders come from PM inspections, then PM is not driving your work (0 points on the assessment); raise inspection-generated work toward 80-100%.
• If operations/facilities resist PM downtime, then treat it as a communication gap and explain the benefits before rescheduling or dropping the PM.
• If overtime stays near or above the 14.1% average, then you are still reactive; review whether PM tasks are planned and completed.

• Equating PM with lube routes: the tell is that development stops once lube routes exist and breakdowns continue.
• Untrained entry-level lubricators: the tell is excessive breakdowns despite a documented lubrication program.
• PM scheduled without operations/facilities buy-in: the tell is equipment not released for PM and recurring coordination disputes.
• No work order system behind PM: the tell is that you cannot show what PM was done or what it found, so nothing is measurable.
• Replacing parts without failure analysis: the tell is the same failures recurring and PM tasks that never change.

• ▸PM activity types in use: list inspections, lubrication, adjustments, and any others actually performed.
• ▸Equipment with lubrication-only PM: name each item that gets no inspection or adjustment.
• ▸Lubricator skills/training status: record the training received and how the work is checked.
• ▸% of work orders generated from PM inspections: calculate from work orders; target 80-100%.
• ▸Planning completeness per PM job: mark craft, materials, tools, and job instructions as present or absent.
• ▸Operations/facilities agreement: record the date PM windows were agreed and who agreed them.
• ▸Failure analysis on breakdowns: record yes or no and the action taken from findings.
• ▸Overtime % of total hours: compare against the 14.1% average.
• ▸Next actions and owner: one fix per gap found above.

1. 1Baseline the storeroom with the ten Section 6 audit questions from Chapter 1. Measure each as a percent: materials in stores when required (4 pts above 95%), items in the catalog (90%+), aisle/bin specified (above 95%), issues charged to a work order or account (above 95%), max/min set (above 95%), critical material accessible (above 95%), inventory updated daily on receipt (above 95%), items with at least one issue in six months (90%+). Score the answers, then work the lowest-scoring ones first.
2. 2Settle who controls stocking before touching anything else. The audit gives maintenance 4 points for controlling what is stocked and anyone else 0. In about half of organizations someone outside maintenance decides what to stock, how many, and how many to issue. Move all three decisions to maintenance, and let purchasing execute the buy.
3. 3Build a complete stores catalog so that every item physically in inventory appears in it. Produce it in both alphabetic and numeric listings (4 pts; either one alone scores 2; incomplete or missing scores 0). Assign an aisle/bin location to every item, targeting more than 95%.
4. 4Set a maximum and minimum level for each stocked item, targeting more than 95% of the inventory. Do not fix the problem by overstocking. Carrying cost runs over 30% of item price per year, so a $1 million inventory costs over $300,000 per year. Keep levels as low as still delivers satisfactory service and avoids stockouts.
5. 5Identify critical maintenance material and stock it in the warehouse or a readily accessible location, targeting more than 95% of it. Critical items get protected first when you trim levels elsewhere.
6. 6Enforce transaction discipline. Issue every item to a work order or account number as it leaves the store (target above 95%). Update inventory levels on the day materials are received (target above 95%). These two controls are what make the balances trustworthy.
7. 7Review item activity. Check every item for at least one issue in the last six months (target 90%+). Items with no issue become candidates for review against their cost of carrying.
8. 8Install the four inventory reports and four purchasing reports from the Chapter 1 reporting audit. Inventory: stock catalog, valuation report, performance report (stockouts, level of service, turnover), and where-used report. Purchasing: vendor performance (promised vs. actual delivery), price performance (quoted vs. actual), buyer performance, and a non-stock report of direct buys. Run the inventory and purchasing functions in the CMMS (90%+ of them performed in the system scores 4 pts).

The passages give a cost case. A site has a $1 million maintenance inventory. At a carrying cost of over 30% of price per year, it costs over $300,000 annually to hold. The common reaction to a materials problem is to overstock the storeroom, which raises that cost without fixing the control problem. Instead, the manager scores the storeroom against the Section 6 questions. Suppose the audit shows materials in stores when needed only 70 to 79% of the time (2 pts), max/min specified on under 70% of items (0 pts), and someone outside maintenance deciding what to stock (0 pts). The sequence is: first take stocking control into maintenance, then set max/min on every item, then protect critical items in an accessible location, then verify daily receipt updates and work-order issue coding. Stocking decisions now follow actual issue history, not a blanket increase in stock. The passages show the stakes of getting this right. Materials run 40% to 60% of the maintenance budget, and lost production can cost two to fifteen times a repair, 4:1 on average. A $10,000 repair that waits on a missing part is really closer to $50,000. The stock-versus-carry tradeoff is therefore judged against downtime cost, not just the carrying rate.

• If someone outside maintenance decides what is stocked, how many, or how many to issue, then move those decisions to maintenance first, because the audit scores anything else at 0 points and it is the support function directing its customer.
• If materials are missing when needed, then check max/min coverage, critical-item accessibility, and receipt posting before you raise stock levels, because blanket overstocking adds carrying cost of over 30% of item price per year.
• If an item is critical, then stock it in the warehouse or a readily accessible location even when its issue history is thin, since a stockout converts a repair into lost production at two to fifteen times its cost.
• If an item shows no issue in six months, then review it as a carrying-cost candidate rather than leaving it at its current max/min.
• If the stockout and level-of-service report cannot be produced, then fix issue-to-work-order coding and daily receipt updates first, because the performance report depends on those records.

• Overstocking to fix availability. The tell: service level barely moves while inventory valuation climbs, and carrying cost (over 30% per year) is never mentioned in the discussion.
• Maintenance does not own stocking decisions. The tell: purchasing or accounting sets quantities, and maintenance managers cannot say why an item's max/min is what it is.
• Untracked issues and late receipts. The tell: parts leave the store without a work order or account number, receipts are posted days later, and counts disagree with the system.
• Incomplete catalog and locations. The tell: items on the shelf are missing from the catalog, bins are unmarked, and technicians buy directly for things already in stock (non-stock report full of duplicates).
• Counting only labor and parts when justifying stock. The tell: stock decisions are argued against the price of the repair alone, ignoring that lost production averages 4:1 against it.

• ▸Item / catalog number: record the number as it appears in the alphabetic and numeric catalog; blank means the item is missing from the catalog.
• ▸Aisle/bin location: record the location; blank counts against the 95% target.
• ▸Critical? (Y/N) and location accessible: mark whether the item is critical and whether it is stocked where it is readily accessible.
• ▸Max / Min levels: enter both; blank counts against the 95% target.
• ▸Unit price and carrying cost per year: enter price and multiply by 30%+ to show annual holding cost per unit.
• ▸Last issue date and work order/account number: shows whether at least one issue occurred in six months and was properly charged.
• ▸Last receipt date and date posted: a gap shows whether the daily-update standard is being met.
• ▸Stockouts in period and service level: log times the item was needed and not in stores.
• ▸Stocking decision owner and action (keep / change max-min / review / remove): name the maintenance decision-maker and the action taken.

1. 1Score your current program against the Section 2 'Training Programs in Maintenance' audit in Chapter 1. It has ten questions, each worth 0 to 4 points, so the maximum is 40 (the planner-subject question adds one point per subject, so score the questions as written). Cover supervisory training, planner training, planner subjects, general quality and productivity training, craft training, training intervals, format, instructors, and the rest. Record the letter answer for each question, not just the total, because the letters show where the gap is.
2. 2Fix the planner and supervisor gaps first. The Foreword reports that only about one-third of organizations employ a job planner. It also cites planned versus unplanned work cost ratios as high as 1:5. Move planner training from the 'on the job' rung (1 pt) toward a written program (3 pts) or public seminars (4 pts). Cover all four scored subjects: work order planning and execution, material planning, scheduling practices, and project planning.
3. 3Make supervisor training happen at promotion and repeat on a schedule. The top rung (4 pts) is 'all are trained when salaried and additional training is mandatory on a scheduled basis.' Optional follow-up training scores only 3.
4. 4Tie craft training to a pay and progression program (4 pts). The lower rungs are: formal experience required plus OJT (3), experience required before hire (2), OJT only after hire (1), nothing (0). Define skill levels and link each pay step to demonstrated training completion.
5. 5Set the training interval for ALL craft employees at less than 1 year (4 pts). Between 12 and 18 months scores 3, and between 18 and 24 months scores 2. Training given to only some employees scores 1 at best, so a calendar that covers everyone matters more than depth for a few.
6. 6Build the format as a mix of classroom and lab or workshop exercises (4 pts). All-classroom scores 3, all-lab scores 2, and all-OJT scores 1. Use outside contract experts as instructors where you can (4 pts); staff subject experts score 3.
7. 7Fix entry-level skill gaps, especially in lubrication. The Foreword notes that the oiler/lubricator is generally an entry-level position. Lack of skills and training there produces substandard lubrication, which leads to excessive breakdowns even when a lube program exists. Train lubrication staff before assuming the program works.
8. 8Extend training beyond the maintenance department. Score the general quality and productivity training question (4 pts when it reaches upper management, line supervision, hourly workers, and support personnel). Record whether management supports ongoing skills training (4 pts yes, 0 pts no). Re-score yearly and compare against the prior total.

The Foreword describes a pattern that shows what training addresses. Preventive maintenance is often reduced to lube routes. Almost three-fourths of organizations have lube routes and procedures, but the oiler position is generally entry-level, and the technician's lack of proper skills and training often leads to substandard performance and excessive breakdowns despite the program. Apply the audit to a plant in this position. Lubrication staff have OJT only, which scores 1 on craft training. Planners learn on the job, which also scores 1. Supervisors get optional training, which scores 3. Training reaches only some employees every two or more years, which scores 1 on intervals. Your first moves follow the scoring ladder. Give lubrication technicians a mixed classroom and lab course (4 pts on format), and move planners from 1 pt to 2 pts with at least a month of one-on-one OJT, then to a written program (3 pts). Make supervisor refreshers mandatory and scheduled (4 pts), and schedule all craft training at least annually (4 pts). Re-run the audit after a year to verify the points moved.

• If planner training is OJT only or nonexistent (0 to 1 pt), fix it before broader craft training, because planned work may cost as little as one-fifth of unplanned work.
• If training reaches only some craft employees, then change the schedule before adding content, because the interval question gives at most 1 pt unless ALL employees are covered.
• If a preventive program consists mainly of lube routes and breakdowns remain high, then check the skills and training of the entry-level lubricators before expanding the routes.
• If you must choose a format, then use a mix of classroom and lab (4 pts) over classroom-only (3) or OJT-only (1).
• If management support for ongoing skills training is 'No' (0 pts), then secure that commitment first, because the rest of the program depends on continued funding.

• Treating OJT as a training program. The tell: craft, planner, and format questions all score 1 pt, and no written curriculum or schedule exists.
• Training only the craft and ignoring supervisors and planners. The tell: planners have no seminar or written program, and supervisors received no training after being salaried.
• Irregular, optional, or partial training. The tell: attendance depends on who volunteers, intervals exceed 18 to 24 months, and only some employees are covered.
• Assuming a program exists because the paperwork exists. The tell: a lubrication program is in place but breakdowns stay excessive, because entry-level technicians are unskilled.
• Training with no link to pay or progression or to management support. The tell: skills learned have no effect on pay level, and training is cut when budgets tighten.

• ▸Supervisory training (0-4): record the letter answer and note whether training is mandatory and scheduled.
• ▸Planner training (0-4): record the rung (seminar, written program, 1-on-1 OJT, OJT, none).
• ▸Planner subjects covered (0-4): add 1 point each for work order planning/execution, material planning, scheduling, and project planning.
• ▸General quality and productivity training (0-4): list which groups are included.
• ▸Craft training structure (0-4): note whether it is tied to a pay and progression program.
• ▸Training interval for ALL craft (0-4): enter months between sessions and whether all employees are covered.
• ▸Training format (0-4): classroom, lab, OJT, or mix.
• ▸Instructors (letter answer): outside expert, staff expert, or other.
• ▸Lowest-scoring item and next action: name the item, owner, and due date.