Maintenance Work Management Processes Maintenance Strategy Series

1. 1Define the coverage universe first. List all plant and facility equipment, including special back-up and redundant units. The source states that all of it must fall under a complete, cost-effective PM program, so an asset left off the list is a gap.
2. 2Rank the equipment by criticality and confirm the PM program gives the critical equipment proper coverage. Add each asset's PM tasks to this list, so every asset is shown as covered or uncovered.
3. 3Build each asset's PM from a cross-section of four task types: inspections, adjustments, lubrication, and proactive replacement of worn components. If an asset's PM is only inspections, or only lubrication, the cross-section is incomplete.
4. 4Issue every PM task as a work order charged to the equipment, functional location, or building/floor/room. The source says even a PM task is generated as a work order. This builds the history file and carries labor, parts, and contractor costs to the asset, so you can later judge whether the PM is cost-effective.
5. 5Classify every work order by lead time. Any maintenance activity performed with less than one week of advance planning is unplanned (reactive). Everything with a week or more of planning is planned.
6. 6Measure the share of total maintenance manpower spent on unplanned work. Apply the 80%/20% test: at most about 20% unplanned and about 80% planned. Note that the source passage's wording ('less than 80%') is garbled. Its later references to the '80%/20% rule' and the '80% planned mode' show the intent is 80% planned.
7. 7If the test fails, re-evaluate the PM program (coverage, task mix, critical-asset focus) before doing anything downstream. Do not move on to MRO stores and purchasing, planning and scheduling, or predictive work until the program passes. The source says to hold progress until PM is successful.
8. 8Staff the support the program needs. The source gives one planner per 15-20 craft technicians. It reports that a third of surveyed organizations had no planners at all, and the rest had the wrong ratios.

The source offers no single case with measured hours, so this is a structured application of its rules, and the hours are hypothetical. Suppose a plant covers its critical pumps with only quarterly inspections. Over a month, work orders are tagged by lead time and 'planned' means a week or more of advance planning. Say 320 of 1,000 craft hours went to jobs started with less than a week of planning. That is 32% unplanned, which fails the 80%/20% test. Under the source's rule, the plant stops work on MRO stores, scheduling and predictive tools and re-evaluates PM. It checks whether back-up pumps were left out, then adds lubrication, adjustment and worn-component replacement tasks to the critical-asset PMs. Why this matters: the source warns that compensating for a reactive organization (bending best practices in later processes to make them work reactively) reinforces negative behavior and sub-optimizes everything built afterward. The source also contrasts wrench time of about 20% in a reactive organization with 60% or more in a proactive, planned, scheduled one. That gap is the payoff for fixing PM first.

• If unplanned work (less than one week of planning) exceeds roughly 20% of manpower, then re-evaluate the PM program before progressing to MRO, planning, or scheduling processes.
• If any equipment, including redundant or back-up units, has no PM, then add it to coverage. The program must be designed to eliminate all unplanned failures, so exempt assets are not allowed.
• If a critical asset's PM consists of one task type only, then rebalance it across inspections, adjustments, lubrication, and proactive replacements.
• If PM tasks are not issued as work orders against the asset, then fix that first. Without a work order, cost and history data cannot be used to judge whether the PM is cost-effective.
• If the team is tempted to work around best practices in later processes because 'we are too reactive,' then treat it as a PM-effectiveness failure. Return to the PM program rather than adapting later processes to reactivity.

• Measuring PM by tasks completed rather than by unplanned workload. The tell: PM compliance reports look strong while 'emergency' work still dominates labor hours.
• Excluding back-up or redundant equipment from the PM list because it is not running. The tell: the first outage of a standby unit is a surprise breakdown.
• Starting advanced predictive or reliability techniques before PM is effective. The source says these will fail when the organization is not mature enough to use them. The tell: new tools produce data nobody acts on, and the crew is still firefighting.
• Defining 'planned' loosely. The tell: jobs planned the same shift are counted as planned, so the unplanned percentage looks better than the definition (less than one week of planning) allows.
• Building later processes around a reactive culture. The tell: stores, scheduling, and purchasing rules have exceptions for 'urgent' work, which reinforces negative behavior.

• ▸Equipment list and coverage: list every asset, including back-up/redundant units, and mark whether it has a PM.
• ▸Critical assets: mark the critical assets and confirm each has proper PM coverage.
• ▸Task mix per critical asset: record whether it has inspections, adjustments, lubrication, and proactive replacements (Y/N for each).
• ▸PM work order link: confirm every PM task is issued as a work order charged to the asset or location (Y/N).
• ▸Review period: state the dates of the measurement window.
• ▸Total craft hours in the period: enter all maintenance manpower hours.
• ▸Unplanned hours: enter hours on work with less than one week of advance planning.
• ▸Unplanned % and gate result: divide unplanned by total hours and compare to the 80%/20% target. Pass or Fail.
• ▸If Fail, corrective actions and owner: list PM gaps to close, who owns each, and the re-test date before moving to MRO or planning processes.

1. 1Rewrite the maintenance mission first. A mission like "Fix it Fast When it Breaks" commits the department to reactive work. Tie the statement to corporate objectives, such as maximizing throughput at lowest cost, highest quality, and within safety standards, and state that maintenance needs a proactive vision to focus its activities.
2. 2Put every job on a work order, including PM tasks, which are generated as work orders. Charge each order to an equipment item, functional location, or building/floor/room. Without equipment history you cannot tell which breakdowns repeat or which PM work is paying off.
3. 3Split incoming work into its sources: preventive/predictive inspections, operational requests, project activities, and emergency/breakdown work. The inlet is not under maintenance's control, so you manage proactivity by seeing what is flowing in. Treat inspection-generated work as the intended inlet and breakdown work as the one to shrink.
4. 4Measure the craft backlog as a tank with an inlet (work generated) and an outlet (work completed). Track it in weeks of work. Backlog rises when more work is generated than completed and falls when more is completed than generated. Without an accurate backlog figure you cannot schedule ahead.
5. 5Control the backlog through the outlet, meaning applied resources. When it climbs, add overtime or contract labor to test whether the rise is temporary. If the rise is a permanent trend, hire craft technicians. If it falls within two weeks, remove the overtime and then reduce contract labor. Never change labor levels without accurate backlog figures.
6. 6Run the planning-to-scheduling chain so scheduled work is executable. Plan the job, kit the parts at stores (or deliver them to the site), build the weekly schedule, and have supervisors pick daily work from it. Confirm equipment availability with operations before technicians show up, not when they arrive.
7. 7Add operator involvement to catch problems early. Operators can clean, inspect, service, use visual controls, and initiate work requests for current or developing problems. Enter those requests into the work order system so they become planned work rather than breakdowns.
8. 8Review proactivity through the KPIs, comparing results to guidelines and a permissible variance. Use the work process indicators and the planning and scheduling performance indicators. Investigate any drift toward emergency work.

The source's tank illustration (Figure 1-16) shows the backlog logic. Work flows in from PM/PdM inspections, operational requests, and projects, and flows out as completed work. The figure labels the inlet at 4 weeks and the outlet at 2 weeks, so more work is arriving than is being finished and the backlog tank rises. A reactive shop would let this run, and each aging job would become a candidate breakdown. A proactive shop opens the outlet. It first applies overtime or contract labor to see whether the surge is temporary. If the rise persists, it treats the trend as permanent and hires craft technicians. If the backlog later falls within two weeks, the overtime is removed and contract labor is reduced or eliminated. At no point does the shop adjust labor on a hunch. The source warns that changes made independently of backlog figures are arbitrary and dangerous. The same discipline lets the weekly schedule be built on work that is planned, kitted, and ready, rather than on whatever broke that morning.

• If backlog is rising, then add overtime or contract labor first to test whether the increase is temporary. If it persists as a trend, then hire craft technicians.
• If backlog falls within two weeks, then eliminate overtime and reduce contract labor. If it keeps declining, then examine staffing and reposition employees.
• If you lack accurate backlog figures, then do not change maintenance labor levels. Fix the work order data first.
• If equipment availability is unconfirmed before the scheduled day, then the schedule is not proactive. Confirm with operations and supervisors ahead of time rather than discovering it on arrival.
• If a problem is found by an operator, then route it as a work request into the work order system so it is planned, not left to become a breakdown.

• The mission is "Fix it Fast When it Breaks." The tell is that speed of repair is the proudest metric and no one can state a proactive vision tied to corporate objectives.
• Work orders are not used or only partly used. The tell is incomplete equipment history, so you cannot say which assets consume the most breakdown effort or cost.
• Labor is added or cut without backlog data. The tell is staffing changes that follow complaints or budget pressure, and a backlog that swings unpredictably.
• Planned work is not execution-ready. The tell is technicians arriving to find the equipment unavailable, or hunting for parts that were never kitted at stores.
• Reorganizing in place of building the process: restructuring, downsizing, outsourcing, centralizing, or decentralizing without a business plan. The tell is staff who expect another change in six months and only pay lip service to it.

• ▸Mission statement (current): copy the wording; mark it reactive or proactive and note which corporate objective it ties to.
• ▸Work inlet by source: list this period's work arriving from PM/PdM inspections, operational requests, projects, and emergency/breakdown work.
• ▸Work completed (outlet): record work orders closed this period, by the same sources.
• ▸Backlog in weeks: record the current figure and whether it rose, fell, or held versus last period.
• ▸Resource response: state overtime, contract labor, hiring, or reduction, and the backlog evidence that justifies it.
• ▸Trend call: mark temporary or permanent, and the date you will re-check (the two-week test).
• ▸Weekly schedule readiness: note whether jobs are planned, parts kitted, and equipment availability confirmed with operations.
• ▸Operator-initiated requests: count requests entered into the work order system, and note cleaning, inspection, or servicing activities covered.
• ▸KPI variance: record the key indicators against their guidelines and note any result outside permissible variance with the owner.

1. 1Fix the definition before you measure: wrench time is the share of paid hours a craft technician has hands on tools doing direct, value-added work. Anything in the Figure 4-5 'nonproductive' list is excluded: work-rule discussion, idle time, traveling and transportation, too many technicians per job, starting late and quitting early, waiting for materials, receiving assignments and instructions, and cleaning up and putting away tools.
2. 2Establish your baseline against the source ranges: a reactive organization runs about 20-30% (the passages cite 20%, 25%, and 30%), and a proactive, planned, scheduled organization reaches 60% or a little more. Place yourself on that scale before setting a target, and do not promise 60% from a reactive start without the planning structure behind it.
3. 3Check the prerequisites first. Wrench time depends on a working work order system, because hours must be tracked per work order and per craft. Confirm that work orders carry hours by craft (not only dollars) and that craft backlog is tracked in hours. If backlog by craft is unknown, you cannot tell whether technicians are 'right sized,' and a wrench-time number will not be actionable.
4. 4Check planner coverage. The proper ratio is one planner to 15-20 craft technicians. A third of surveyed organizations used no planners, and many of the rest had the wrong ratio. Count planners against technicians, and confirm planners write technically complete job plans rather than doing clerical work.
5. 5Break the gap into its Figure 4-5 categories and assign each a fix. Define the work so the right number of technicians goes to the job. Stage and route jobs to cut travel. Notify the storeroom when materials will be needed. Sequence jobs so crews start and quit on time. Give assignments and instructions in a timely way, backed by clear job plans. Pre-arrange tools, rigging, and scaffolding.
6. 6Pre-clear the delay sources named in the effectiveness review: waiting on or looking for parts, rental equipment, equipment shutdown, and drawings. Lockout and safety permits that depend on operations should be arranged before the crew arrives.
7. 7Plan only work where the return exceeds the cost of planning. Set parameters such as work requiring spare parts, work over a set hour limit, or work requiring shutdown and lockout. Keep the job plans flexible so technicians do not push back.
8. 8Convert the gain into hours and dollars: paid hours × wrench-time rate = productive hours, and the difference between the current and target rate × labor rate = potential savings. Monitor as a KPI inside the control loop: when it trends outside permissible variance, identify the cause, implement a corrective action, evaluate it, and adjust.

The source's Figure 4-4 example uses 100 maintenance technicians working 2,000 hours per year each, so the company pays for 200,000 hours of labor. At a 30% rate of productivity (reactive), the technicians accomplish 60,000 hours of actual work. At 60% (planned), they accomplish 120,000 hours. The difference is 60,000 additional productive hours. At a labor rate of $20 per hour, that is 60,000 × $20 = $1.2 million of potential savings. The source ties the gap to the Figure 4-5 categories: clearly defined jobs cut work-rule discussion, better scheduling cuts idle time, staging cuts travel, and advance storeroom notification cuts waiting for materials. Timely assignments and instructions and pre-arranged tools do the same. The source frames the saving as potential, and it shows up as less overtime and outside contracting, not as a cash line item that appears by itself.

• If you have no planners, or the ratio is far from 1:15-20 technicians, then fix planner staffing before launching any wrench-time target or tracking effort.
• If hours by craft are not captured on work orders or craft backlog is not tracked, then repair work order discipline first; the wrench-time figure and the staffing conclusions drawn from it will be unreliable.
• If measured wrench time sits near 20-30% and most work is reactive, then target planning and scheduling of 80-90% of activities weekly and treat 60% as the ceiling-level goal, not the next-quarter number.
• If the dominant delay is waiting on materials, tools, rigging, drawings, or shutdowns, then fix job-plan content and advance coordination with stores and operations, not technician supervision.
• If a job is small, fast, or needs no parts, shutdown, or special equipment, then skip detailed planning, since planning must return more than it costs.

• Treating wrench time as a technician discipline problem: supervisors push people to work harder while the delays come from missing parts, unclear instructions, and unprepared shutdowns. The tell: delays cluster in materials, instructions, and travel categories, not in effort.
• Letting supervisors or technicians do their own planning because planning 'has no value.' The tell: low wrench time, heavy reactive work, and no planner role or an oversized planner ratio.
• Over-rigid planning that dictates every step. The tell: craft pushback, resentment, and job plans that never get refined.
• Planning everything regardless of return, or using clerical planners without technical skill. The tell: planners add paperwork on trivial jobs and technicians still call with questions the job plan should have answered.
• Roles blur: planners, supervisors, storeroom, and operations each step into one another's work. The tell: finger pointing about who caused the delay, with wrench time not improving despite a planning program.

• ▸Number of craft technicians: count of technicians in scope.
• ▸Paid hours per technician per year: e.g., 2,000; multiply to get total paid hours.
• ▸Current wrench-time %: measured, not assumed; note the method and date.
• ▸Target wrench-time %: set against the 20-30% reactive and ~60% planned reference points.
• ▸Hourly labor rate: use the actual blended rate for the crafts in scope.
• ▸Potential savings: paid hours × (target % − current %) × labor rate.
• ▸Planner-to-technician ratio: planners ÷ technicians; compare to 1:15-20.
• ▸Top three delay categories: pick from the Figure 4-5 list (materials, instructions, travel, etc.) with estimated hours each.
• ▸Corrective action, owner, and review date: one fix per delay category, plus when you will re-measure.

1. 1Classify your current state first, using the three groups in the Figure I-1 survey: not using work orders, partially using them, or fully using them. The first two groups collect information that is incomplete and unsuitable for sound financial decisions about assets. Name which group you are in per site or per craft, because that sets the scope of the fix.
2. 2Make the work order the single hub for all maintenance information. Every repair, service, and PM task, including PMs the system generates, exists as a work order. If work happens without one, it never reaches history, and that gap is your biggest completeness loss.
3. 3Require every work order to be charged to an asset identifier: an equipment item, a functional location, or a building/floor/room indicator. History files build from these identifiers. A work order with no valid identifier cannot feed analysis for any asset.
4. 4Post all cost categories to the work order, not just labor: spare parts, contractor, rental equipment, miscellaneous costs, and rebuild costs. Check each closed order for the categories that applied. Cost data feeds the 'most costly equipment to maintain' report, so any missing category understates that equipment.
5. 5Standardize problem and cause descriptions with controlled lists. The book's appendices provide an Equipment Types list, a Problem Code Master, and a Cause Code Master as models. Free-text entry cannot be rolled up into a 'top ten equipment problems' list, so give technicians a pick-list.
6. 6Tie closure to posting. When the work order is completed, its cost information and repair history are charged to the appropriate history file. Do not let orders sit open after the work is done; closing them is what makes the data timely. Review each order at closure (Chapter 11, Work Order Closure and Analysis) for asset, codes, and costs.
7. 7Test the data with the reports the CMMS/EAM must produce: top ten equipment problems, most costly equipment to maintain, percent reactive vs. proactive maintenance, and cost tracking of all parts and costs. If the reports look implausible or cannot be produced, the data is deficient. Ask the book's evaluation question: is the data we are collecting complete and accurate?
8. 8If the data fails those tests, specify new work management process steps that correct the problems. The source says the re-evaluation may show weak system utilization, and the fix is a process step, not only a system change. Re-test after the change.

The source gives a mechanism rather than a numbered case, so this walk-through follows its work order flow with no invented figures. A PM task is generated as a work order and charged to a specific piece of equipment. A technician performs it and finds a problem. The work order records labor, a spare part drawn from stores, and a contractor charge. At completion, the cost information and repair history post to that equipment's history file. In an organization with full work order utilization, that file is complete, so a 'top ten equipment problems' report and a 'most costly equipment' report reflect this asset accurately. In a partial-utilization organization, the contractor invoice never reaches the work order and the problem is typed as free text. The same asset then looks cheaper than it is and drops out of the problem rollup. The data feeding life cycle cost decisions is incomplete in exactly the way the Figure I-1 survey warns about.

• If work is performed without a work order, then treat it as a data-quality failure, not a paperwork lapse. Anything outside the work order is missing from history and cannot inform asset decisions.
• If a cost category (parts, contractor, rental, miscellaneous, rebuild) applied but is not on the closed order, then reject the closure, because the asset's cost history is understated.
• If the top-ten or cost reports cannot be produced, or do not match what you know about the plant, then re-evaluate the data collection and the system's utilization before trusting any analysis.
• If problems and causes are entered as free text, then move to master code lists (problem, cause, equipment type) so records can be aggregated across assets.
• If you are tempted to adopt advanced analysis before basic posting is reliable, then stop. The source says advanced techniques deployed before the organization is mature enough to use them will fail, so fix the basics first.

• Partial work order utilization: only some jobs get orders, or orders carry only some costs. The tell is that history files look thin for assets you know are worked on often.
• Orders without a valid asset, location, or building/floor/room charge. The tell is costs or problems that cannot be traced to an item when you pull a history file.
• Treating the CMMS flow as the whole process. The vendor determines most of the system flow, but the source says true work management also needs disciplined, repeatable organizational processes. The tell is a configured system with inconsistent human entry.
• Uncontrolled problem and cause descriptions that defeat reporting. The tell is that a top-ten problem list is impossible to build, or is dominated by inconsistent wording for the same failure.
• Trusting reports without checking the inputs. The tell is that the most costly equipment list ranks assets by what happened to be posted rather than by what was spent.

• ▸Work order number: identify the record being reviewed.
• ▸Charged to (equipment / functional location / building-floor-room): confirm a valid identifier exists so the order posts to a history file.
• ▸Origin (breakdown, PM, planned, other): confirm even PM tasks exist as work orders.
• ▸Problem code: select from the Problem Code Master, not free text.
• ▸Cause code: select from the Cause Code Master.
• ▸Costs posted (labor / spare parts / contractor / rental / miscellaneous / rebuild): mark each category that applied and confirm its amount is on the order.
• ▸Completion date and posting date: confirm the order is closed and posted promptly after the work finishes.
• ▸Posted to equipment history (Y/N): confirm cost and repair history reached the history file.
• ▸Reviewer and exceptions: record who checked the order and which fields were corrected.

1. 1Make the work order the single cost-capture document. Charge every job, including PM tasks, to an equipment item, functional location, or building/floor/room, and post labor, spare parts, contractor, rental, miscellaneous and rebuild costs to it. Without complete history you cannot make sound life-cycle cost decisions.
2. 2Check where you sit on the work-order survey scale: not utilized, partially utilized, or fully utilized. If you are in the first two groups, fix work-order discipline before any other cost initiative, because your cost data is incomplete.
3. 3Measure the share of unplanned work, defined as any maintenance performed with less than one week of advance planning. Treat the target as the 80%/20% rule and re-evaluate the PM program if the limit is exceeded. The passage's wording is internally inconsistent: it says 'less than 80%' in one place and 'less than 20% unplanned weekly' in another. Resolve it by treating the 80% planned / 20% unplanned split as the intended target, and do not advance to later process improvements until you meet it.
4. 4Staff planners at the stated ratio of one planner to 15-20 craft technicians, and track backlog in hours by craft. If backlog is not tracked by craft, you cannot tell whether craft headcount is right-sized, so you cannot size labor to the work.
5. 5Use planning and scheduling to attack waiting time. Eliminate delays from waiting on or looking for parts, rental equipment, equipment shutdown, drawings and tools. Track wrench time, which runs about 20% in an average reactive organization and 60% or a little more in a planned and scheduled one.
6. 6Tune MRO inventory against service level, meaning the percent of time a part is in stock when needed. Target 95-97% service level, 100% data accuracy, more than 1 turn per year, elimination of non-moving spares, reduction of slow-moving spares, controlled access, consignment and strategic supplier partnering. Stores cannot be cost effective while demand is mostly reactive: they either miss service level or carry excess stock.
7. 7Run the improvement loop: find an area, set a target through a gap ('As Is – To Be') analysis, implement, then measure with accurate and timely data. Repeat it, and computerize the system (CMMS/EAM) once paper volume buries the workforce.

The passages give the mechanism rather than a plant case, so this example uses their benchmarks. Assume a reactive shop with 20% wrench time. Planning and scheduling can raise that to about 60%, so the same craft hours deliver roughly three times the productive work. The saving shows up as fewer delay hours spent waiting on parts, tools, drawings and shutdowns, not as headcount cut by decree. At the same time, moving from mostly unplanned work toward the 80% planned mode lets stores plan demand. Stores can hold 95-97% service level without carrying excess inventory. Contractor and rental spend is charged to work orders, so you can see which assets drive it. Cost falls for the same level of service because waste is removed at three points: idle labor, excess stock, and emergency purchasing.

• If work orders are not used or only partially used, then fix work-order capture first, because cost decisions on incomplete data are unreliable.
• If unplanned work (under one week of planning) exceeds the 80%/20% limit, then re-evaluate the PM program and hold off on further process development, because every later process costs more and takes longer in a reactive environment.
• If backlog is not tracked in hours by craft, or planners are missing or outside 1:15-20 craft technicians, then you cannot claim craft are right-sized. Fix planning before adding or cutting labor.
• If service level is below 95-97% or inventory is bloated, then check the proportion of reactive demand before changing stocking rules.
• If paper work and filing are consuming the workforce and nobody uses the data, then computerize the work order system.

• Compensating for reactivity: teams bypass planning best practices to make them work in a reactive environment. The tell is a planning process that exists on paper while most jobs still start with less than a week of notice.
• Partial work-order use: costs live in spreadsheets or invoices, not on equipment history. The tell is that you cannot state total cost by asset or by cost type.
• Planners in name only: planner titles exist but the ratio is off or planners do not drive scheduling. The tell is that craft backlog by hours is unknown.
• Fixing stores before fixing PM: you raise stock to protect service level. The tell is rising inventory value with turns not above 1 and slow-moving or non-moving spares persisting.
• Measuring wrench time without removing delays: you report the number but still tolerate waiting for parts, tools, drawings and shutdowns. The tell is the same delay causes recurring week after week.

• ▸Work-order utilization (none / partial / full): record the current state and the evidence for it.
• ▸Unplanned work % of manpower hours: hours on jobs with under one week of planning divided by total hours; compare to the 80%/20% rule.
• ▸Planner-to-craft ratio: planners divided by craft technicians; target 1:15-20.
• ▸Backlog hours by craft: list hours for each craft; mark 'not tracked' if unknown.
• ▸Wrench time %: share of time actually working versus delayed or waiting; baseline against 20% reactive and 60% planned.
• ▸Top delay causes: count waiting on or looking for parts, rental equipment, shutdown, drawings, tools.
• ▸MRO service level %: parts in stock when needed; target 95-97%.
• ▸Inventory turns and non-moving/slow-moving spares: record turns per year (target above 1) and spares to eliminate or reduce.
• ▸Gap, target and next action: state As-Is vs To-Be for the weakest item, an owner and a re-measure date.

1. 1Split the asset's performance into two separate measures. Availability is the share of time the equipment can produce, so unplanned downtime drives it. Efficiency is the rate of production while in service, so idling and minor stoppages drive it. Track each on its own, because the fixes differ.
2. 2Log every stop as a work order charged to the specific equipment item, functional location, or building/floor/room. The source says a partial or absent work order system leaves asset information incomplete and unfit for financial decisions. Without equipment history you cannot see which assets lose uptime.
3. 3Attack availability first by eliminating unplanned equipment downtime. Use PM and PdM inspections and services to cut the number of unexpected breakdowns and service interruptions. Find the repetitive failures and treat them with reliability techniques (RCM, FMEA), because eliminating repetitive failures is what raises availability.
4. 4Attack efficiency separately by finding idling and minor stoppage losses. The source says these usually come from small problems developing in the equipment. Target them with a good PM program plus selected predictive activities.
5. 5Remove repair-time delays through planning and scheduling. Eliminate waiting on or looking for parts, rental equipment, shutdown windows, drawings, and tools. Any time gained in a repair translates into reduced downtime. The source pictures the target as a NASCAR pit stop.
6. 6Check the reporting structure for conflicts. In a production-centric organization, supervisors paid on throughput sacrifice PM. In an engineering-centric one, project slippage pulls maintenance resources off PM. If you find either, add equipment availability or life-cycle cost numbers to the supervisors' compensation.
7. 7Roll the two measures into OEE for a holistic view. Compare against an understood upper limit, because chasing the maximum OEE can cost more than the investment recovers. Once availability and efficiency goals are met, use OEE in equipment-related decisions.
8. 8Tie results to money. Added throughput on fixed overhead lowers cost per unit and raises return on asset. The source cites studies in Japan and the U.S. that capacity-driven profit exceeds expense-cutting benefits by four to one.

The source gives no plant-level data, so this example follows its stated logic. A line is losing output, and the maintenance manager separates the loss into two buckets. Bucket one is unplanned downtime, which hits availability. Bucket two is idling and minor stoppages while running, which hits efficiency. For downtime, work orders charged to the equipment history show repetitive failures on the same assets. These go to RCM and FMEA, and PM and PdM tasks are added to catch wear early. For the minor stoppages, small developing problems are found and covered by PM. Planning and scheduling then removes waits for parts, tools, drawings, and shutdown windows, so each repair returns the asset to service faster. Overhead stays fixed while throughput rises, so cost per unit falls. Under the source's four-to-one finding, that capacity gain is worth more than an equal expense cut.

• If unplanned downtime is the dominant loss, then work on availability first: PM/PdM and elimination of repetitive failures.
• If the asset runs but below rate, then look for idling and minor stoppages caused by small developing problems, and respond with PM plus selected predictive tasks.
• If repairs are long, then audit planning and scheduling for waits on parts, tools, drawings, rentals, and shutdowns before changing the maintenance policy.
• If production supervisors are paid only on output, then add availability or life-cycle cost measures to their compensation, or expect PM to be sacrificed.
• If OEE is being pushed toward its maximum, then stop at the understood upper limit, because beyond it the cost may exceed what the investment can recover.

• Reporting a single uptime figure and missing efficiency losses. The tell: the asset is counted as available but throughput stays below design rate.
• Incomplete work order use, so downtime is not charged to equipment. The tell: you cannot name the repeat-failure assets from history.
• Letting projects or output targets pull maintenance off PM. The tell: PM tasks are deferred whenever projects slip or production targets tighten, and breakdowns follow.
• Buying predictive technology because it is advanced. The tell: PdM tools are in use but downtime has not dropped and PM intervals are unchanged.
• Chasing maximum OEE without understanding its upper limit. The tell: spending climbs while payback on the equipment investment does not.

• ▸Asset / functional location: the equipment item the work orders are charged to.
• ▸Period reviewed: the time window for the analysis.
• ▸Unplanned downtime events and hours: count and total hours of unexpected stops from work orders.
• ▸Repetitive failure modes: failures recurring on this asset, to send to RCM/FMEA.
• ▸Idling and minor stoppage events: stops while in service, with the small problem behind each.
• ▸Repair delays: time lost waiting on parts, tools, rentals, drawings, or shutdown.
• ▸PM/PdM tasks in place or added: the tasks aimed at each loss above.
• ▸OEE and its understood upper limit: current value and the ceiling beyond which cost outruns recovery.
• ▸Cost-per-unit effect: expected change in cost per unit from added throughput on fixed overhead.

1. 1Write the ROA equation for your plant: profit divided by asset valuation. Then list every maintenance lever by which side it moves. Expenses (labor, materials, contracting, energy) and throughput/availability move the numerator. The investment in the asset moves the denominator.
2. 2Set a mission and vision for maintenance and reliability, then link them through a performance management system to the corporate objective, usually profitability. Without that financial linkage, reorganizations (downsizing, centralization, outsourcing) are random and fail.
3. 3Break maintenance expense into labor, materials, and contracting. Add energy cost as a fourth line, because deferred preventive work such as cleaning heat exchangers or coolers raises energy use. Track all four against the equipment history.
4. 4Ensure every cost lands on a complete work order posted to equipment history. Partial or absent work-order use gives incomplete data that cannot support life cycle costing decisions.
5. 5Collect the data total cost management needs per critical asset: MTBF, MTTR, downtime or lost-production cost per hour, a Pareto of failure causes, initial cost, replacement cost, and complete work order history.
6. 6Include lost production and energy waste when you time a PM task or decide on a repair. Cutting maintenance spend alone is not the goal; the decision must weigh all costs the equipment decision touches.
7. 7Raise capacity by lifting wrench time through planning and scheduling. Eliminate delays waiting on or looking for parts, rental equipment, shutdowns, drawings, and tools. Reactive shops run around 20% wrench time, planned and scheduled shops up to 60% or more.
8. 8Run the improvement loop: pick an area, set targets, implement, measure, repeat. Review ROA drivers each cycle using accurate, timely data.

The source describes energy as a profit lever. Take a plant whose heat exchangers and coolers are not cleaned at the proper frequency. Energy usage rises, which is an expense line the maintenance function directly influences. Doing the preventive cleaning at the right interval cuts energy cost, and as the source puts it, expense dollars not spent drop to the bottom line and become profit dollars. That raises the ROA numerator. The same logic applies to timing PM tasks: the interval decision must count cost of lost production or downtime, not just labor and parts. To run it properly you pull MTBF, MTTR, downtime cost per hour, the failure-cause Pareto, and the work order history for the exchanger. The source warns that without this data the analysis cannot be done. The caution runs the other way too: cutting headcount, spares, or contracting raises profit only if all other factors stay the same, and the source does not claim that holds in practice.

• If a proposed cost cut (headcount, spares, contracting) is justified only by lower maintenance expense, then test it against downtime, lost production, and energy cost before approving, because the profit gain assumes all other factors stay equal.
• If work orders are missing or partially used, then fix work order data capture before attempting financial optimization or life cycle costing; the data will otherwise be unsuitable for financial decisions.
• If a maintenance initiative is a reorganization (structure, centralization, outsourcing) with no business plan tied to corporate profitability, then stop and build the mission, vision, and performance linkage first.
• If wrench time is near the reactive 20% level, then prioritize planning and scheduling to remove waiting-for-parts, tools, drawings, and shutdown delays before adding labor.
• If PM or cleaning intervals are being set, then include downtime and energy costs in the calculation, not just task cost.

• Treating profitability as expense reduction only. Tell: the plan counts headcount, spares, and contract cuts but never mentions throughput or ROA's denominator.
• Missing maintenance mission and vision. Tell: nobody can state why the maintenance department exists or how it ties to corporate objectives.
• Running financial optimization on incomplete data. Tell: MTBF, MTTR, downtime cost per hour, or work order history are absent or guessed.
• Ignoring energy and lost-production costs in PM timing. Tell: PM intervals are justified only by labor and parts cost, and exchangers or coolers go uncleaned.
• Random reorganizations without a business plan. Tell: repeated structure changes with no performance measures linking to profitability, and little success.

• ▸Asset/plant and period: name the scope being analyzed.
• ▸Profit (numerator) baseline: record current profit for the scope.
• ▸Asset valuation (denominator) baseline: record the investment value in the assets.
• ▸Maintenance expense by category: list labor, materials, contracting costs separately.
• ▸Energy cost linked to maintenance: note items such as heat exchanger or cooler cleaning frequency and current energy impact.
• ▸Downtime/lost-production cost per hour: enter the value used in all repair and PM decisions.
• ▸Data readiness check: mark MTBF, MTTR, failure-cause Pareto, initial cost, replacement cost, work order history as present or missing.
• ▸Wrench time current vs. target: record the measured percentage and the planned target.
• ▸Proposed action and expected ROA effect: state which side (numerator or denominator) moves and the measure that will confirm it.

1. 1Name the business reasons for improving maintenance before you design anything. The source lists poor ROI on total plant or facility valuation, poor throughput versus design, inability to meet production demands, high cost of occupancy, excessive downtime, and production inefficiencies. Write down which of these your site actually has, and attach a measured baseline to each.
2. 2Define the maintenance mission and goals as a business, with ROA (return on assets) as the target. Every goal must trace to profitability. Reject goals that stop at technical activity, such as 'complete all PMs', unless they also name the financial outcome they serve.
3. 3Assign roles and responsibilities at every level, from senior executive to shop floor. The source says maturity requires the whole organization to share a focus on maximizing the investment in the assets. List who owns the PM program, planning, scheduling, execution, closure, and cost review, and name one person for each.
4. 4Evaluate and train maintenance managers on business skills, not only technical skills. As people move from technician to supervisor to superintendent to maintenance manager, managerial and business skills matter more and technical skills matter less. Add business and managerial training to the development path before the promotion.
5. 5Make the work order the single data hub so the business can be managed. Charge every labor, spare-part, contractor, rental, rebuild, and miscellaneous cost to an equipment item, functional location, or building/floor/room. Post it to the history file at closure, including PM tasks, which also run as work orders. Without full work order utilization, the data cannot support life-cycle costing decisions.
6. 6Build financial decision-making on that data. For Total Cost Management, collect MTBF, MTTR, downtime or lost-production cost per hour, a Pareto of failure causes, initial cost, replacement cost, and a complete, accurate work order history. When setting PM timing, include lost production cost. When cleaning heat exchangers or coolers, include wasted energy cost.
7. 7Understand OEE (overall equipment effectiveness) and its upper limit before you chase it. Too low means the return on the equipment investment is not being maximized. Pushing for the maximum OEE may cost more than can ever be recovered. Aim for the financial balance that maximizes ROA.
8. 8Sequence the build-out and expect it to take time. Start with PM coverage of all equipment, including back-up and redundant units, then add advanced techniques only as the organization can use them. The source says full maturity may take up to four years, and advanced techniques deployed too early will fail. Run continuous improvement as a standing habit, because 'Yesterday's Excellence is Today's Standard and Tomorrow's Mediocrity.'

The source describes the survey in Figure I-1, which splits organizations into three groups: those not using work orders, those partially using them, and those fully using them. Take a plant that falls in the middle group and wants to cut excessive downtime and raise ROA. Because its work orders are only partly used, its asset information is 'incomplete and unsuitable for making good financial decisions.' The plant first makes the work order the hub. Labor, parts, contractor, rental, and rebuild costs are all charged to the equipment and posted to the history file when the order closes. PM tasks are issued as work orders too. Once the history is complete and accurate, the plant can assemble the Total Cost Management inputs for a critical asset: MTBF, MTTR, downtime cost per hour, a failure-cause Pareto, and initial and replacement cost. It then sets PM timing with lost production included, rather than judging only on maintenance spend. The maintenance manager is judged on these business results, not on technical skill alone. The plant should expect the gains to build over time, since the source says the full benefit comes only with organization-wide focus, which can take up to four years.

• If work orders are unused or only partly used, then fix work order discipline and cost capture before attempting financial optimization, because the data will be incomplete and unusable.
• If you are setting a PM interval or deciding to clean equipment, then include lost production and wasted energy costs in the calculation, not just labor and parts.
• If OEE is low or not clearly understood, then train on OEE first. If it is understood, then aim for the OEE that balances cost against return and maximizes ROA, not the highest possible number.
• If someone is being promoted toward maintenance manager, then require business and managerial training and evaluate them on business skills rather than technical skills.
• If leadership wants advanced techniques now, then hold to the sequence (PM program and basic processes first), because advanced techniques deployed before the organization can use them will fail.

• Judging maintenance managers mainly on technical skills. The tell: promotions go to the best technician, and the manager cannot discuss cost, ROA, or downtime economics.
• Incomplete work order use. The tell: costs and repair details are missing from equipment histories, and no one can produce MTBF, MTTR, or a failure-cause Pareto for a critical asset.
• Optimizing only maintenance cost. The tell: PM timing and cleaning decisions ignore lost production and energy waste.
• Chasing maximum OEE. The tell: spending grows to hit the target, but the investment is never recovered and ROA does not improve.
• Expecting a quick win and distributing a flow chart as the strategy. The tell: early gains fade into a short-term flash, with no cultural change and no organization-wide focus on the investment in the assets.

• ▸Business driver: Name which of the six drivers apply (poor ROI, poor throughput, unmet production demand, high cost of occupancy, excessive downtime, production inefficiency) and the baseline for each.
• ▸Maintenance mission and ROA-linked goals: State the mission and each goal, with the profitability outcome it serves.
• ▸Role owners: Name the owner for the PM program, planning, scheduling, execution, closure, and cost review.
• ▸Manager business-skills plan: Record the business and managerial training each supervisor, superintendent, and manager candidate will receive and the date.
• ▸Work order utilization level: Mark none, partial, or full, and list the cost types not yet charged (labor, parts, contractor, rental, rebuild, misc).
• ▸Total Cost Management data check: Mark available or missing for MTBF, MTTR, downtime cost per hour, failure Pareto, initial cost, replacement cost, and work order history.
• ▸OEE status: Record the current OEE, whether it is understood, and the cost-balanced target that supports ROA.
• ▸Maturity sequence and timeline: List the current stage, the next stage, and the checkpoint date within the multi-year horizon.
• ▸Continuous improvement mechanism: Name the review cadence and the owner for finding small improvements.

1. 1Classify where you stand first. Using Wireman's three groups, decide whether you are not using work orders, using them partially, or using them fully. Full means people complete data on all maintenance activities, including costs, and post it to an equipment history. Anything less produces information that is incomplete and unsuitable for financial decisions about the assets.
2. 2Make the work order the hub. Require that every repair, rebuild, and generated PM task exists as a work order. The source states that even a preventive maintenance task is generated in the form of a work order. Work done off-system never reaches a history file.
3. 3Force a target on every work order. Each order is charged to a piece of equipment, a functional location, or a building/floor/room indicator. Reject any order without one. This is what builds a history file for each item.
4. 4Define what may be charged to an order. Labor, spare parts, contractor, rental equipment, miscellaneous costs, and rebuild costs can all be charged. Make each a required line, even when its value is zero, so a missing cost is visible rather than silent.
5. 5Post on completion. When the order is completed, all cost information and repair history are charged to the appropriate history file. Close the order only after this posting. Closure is a process step, not a courtesy.
6. 6Standardize the codes. Use standardized codes for equipment, stores, and task/craft assignments. The source also lists a Problem Code Master and Cause Code Master as appendices. Without codes, the technical details cannot be analyzed later.
7. 7Staff the discipline. The source ties poor work order use to a lack of planners. The proper ratio is one planner to 15-20 craft technicians. Add planners before adding system features, because planners keep the data accurate and sustain the process.
8. 8Audit the four known failure points on a schedule: failure to fully implement the CMMS, incomplete utilization, inaccurate data input, and failure to use the data once it is in the system. Treat the CMMS as a tool, not the goal.

Wireman's survey (Figure I-1) shows the majority of organizations in the first two groups: not using work orders, or using them only partially. Figure I-3 shows why. A full third of the organizations surveyed did not use planners at all. The other two-thirds claimed to use planners, but their planner-to-craft ratios were inappropriate. The correct ratio is 1:15-20. Figure I-4 shows the consequence: when planners are not properly used, craft backlog by hours per craft cannot be tracked, and staffing indicators cannot be used. The corrective sequence follows from the source. First, put planners in place at the proper ratio. Second, require every job, including generated PMs, to be a work order charged to an equipment item, location, or building/floor/room. Third, charge labor, parts, contractor, and rental costs to it. Fourth, post cost and repair history to the history file at completion. Wireman's conclusion is that organizations must reach full utilization of work orders before they can successfully implement work management systems and process flows.

• If a job is done without a work order, then count it as unrecorded. The asset's history is incomplete, and you should not use that history for life-cycle cost decisions.
• If an order cannot be tied to an equipment item, functional location, or building/floor/room, then do not release it. Find the asset first.
• If planner-to-craft ratios are far from 1:15-20, or a third of the shop has no planner, then fix staffing before tuning the system. Work order discipline depends on it.
• If the CMMS is in place but no one analyzes or acts on the data, then evaluate usage, not the software. The system is only a tool.
• If a KPI does not point to a clear action, then drop it. Keep only those relevant to the business, measured regularly and consistently.

• Partial utilization: some work is ordered, some is not. The tell is equipment histories with gaps and cost totals that do not match what the shop actually spent.
• Treating the CMMS implementation as the finish line. The tell is that the system is live, yet no one uses its data for equipment decisions.
• Inaccurate data entry. The tell is histories that cannot be trusted and are not used for repair-or-replace decisions, so the data stays unused.
• Buying equipment on low bid for lack of maintenance-cost data. The tell is purchasing and accounting choosing the lowest-cost item because no cost history exists to support a life-cycle choice.
• Claiming to have planners without the right ratio. The tell is that craft backlog by hours cannot be reported, even though planner positions exist on the org chart.

• ▸Utilization group: Mark not using, partial, or full, based on a sample of recent jobs.
• ▸Asset/location link: Record the share of sampled orders charged to an equipment item, functional location, or building/floor/room.
• ▸Cost lines present: For each sampled order, mark whether labor, parts, contractor, rental, and misc/rebuild costs were charged or explicitly zero.
• ▸PM as work order: Confirm that generated PM tasks exist as work orders (yes/no).
• ▸History posting: Confirm cost and repair history posted to the history file at completion (yes/no, with exceptions).
• ▸Code standardization: Note whether problem, cause, equipment, stores, and craft codes are standardized (yes/no).
• ▸Planner ratio: Planners divided by craft technicians, compared with 1:15-20.
• ▸Data problems found: Tick failure to implement, incomplete utilization, inaccurate input, or data not used.
• ▸Corrective action and owner: Record one action per gap, with a named owner and due date.

1. 1Confirm the work order system is the base before you add planners. The passages say data must be complete, accurate, timely, and usable, and that without data that passes this test 'it is impossible to make further progress.' Run those four questions against your equipment history first.
2. 2Check the preconditions. Stores must have spares identified, tagged, and stored in a known location with accurate on-hand and usage data. Purchasing must be able to buy parts to meet the schedules. Treat the benchmark as a service level of 95%-97%, meaning a part is in stock when needed.
3. 3Assess how reactive you are. The passages say the stores and purchasing groups cannot be cost effective unless unplanned work is under 20% weekly. If you are above that, work on the planning enablers (failure analysis, PM, operator reporting) alongside hiring planners, so the planner is not simply absorbing breakdowns.
4. 4Define the planner role as a decision-making position that prepares every job before it is scheduled. For each job the planner determines the scope, tasks, labor skills, materials, tools, and instructions. Write this into the role description.
5. 5Route all planned work through the work order. The planner charges the order to an equipment item, functional location, or building/floor/room indicator, so that labor, parts, contractor, and rental costs post to a history file at closure.
6. 6Split work by complexity using the book's structure: simple planning, complex planning, PM planning, and shutdown/turnaround/outage planning. Match the planner's depth of effort to each category rather than planning everything the same way.
7. 7Pass the planned jobs to scheduling. The passages state that planning leads to scheduling, and that scheduling needs a backlog and prerequisites in place first. Do not let the planner double as the scheduler by default.
8. 8Measure the result with planning and scheduling performance indicators, and with wrench time. Reactive organizations average about 20%, while planned and scheduled organizations reach 60% or a little more. Run the continuous improvement loop: pick an area, set targets, implement, measure, repeat.

The passages give a benchmark comparison rather than a single named case. A reactive maintenance organization has about 20% wrench time, meaning craft technicians have their hands on tools only one hour in five. The rest is spent delayed or waiting, for example for parts, instructions, or tools. A proactive organization that is planned and scheduled reaches 60% or more. The goal is for maintenance to run 'like a pit stop in a NASCAR race.' To apply this, you start with the preconditions. The work order data must be complete, accurate, timely, and usable. Stores must hold a 95%-97% service level. Unplanned work must fall under 20% a week. The planner then prepares each job with scope, tasks, skills, materials, tools, and instructions before it reaches the schedule. Technicians arrive to a job with the parts staged and the steps written. The 20% to 60% gap is the productivity the planning capability is meant to recover. The passages do not say how long this takes, so set your own targets and measure against them.

• If work order data fails the complete/accurate/timely/usable test, then fix work order discipline before staffing planners, because nothing further can progress.
• If stores service level is below 95%-97%, then planned jobs will stall on parts; fix MRO first or in parallel, or the planner's work is wasted.
• If unplanned work is above 20% weekly, then treat planning as unsustainable and invest in the planning enablers (failure analysis, planned component replacement, operator problem reporting) to push breakdowns down.
• If a job is routine and repeatable, then use simple planning or PM planning; if it spans crafts or needs coordination, then use the complex planning process; if it is a shutdown, turnaround, or outage, then use the STO process.
• If you want to judge whether planning is working, then look at wrench time and the planning and scheduling indicators, not at the number of planners hired.

• Hiring planners into a reactive operation. Tell: planners are pulled into emergencies daily, and plans are never issued before the work starts.
• Planning without parts support. Tell: jobs are ready on paper but wait on stock-outs, and the service level is under 95%.
• Work done off the work order. Tell: maintenance is performed but not recorded, so equipment history is lost and plans cannot be improved from failure data.
• Treating planning and scheduling as one undefined function. Tell: nobody can say who determines the scope, labor, and materials versus who assigns the week's jobs.
• Expecting fast results from an advanced setup in an immature organization. Tell: the plan is a flow chart handed out with no change in culture or skills, and it fails, as the author's note warns.

• ▸Data check (complete / accurate / timely / usable): rate each pass or fail for your equipment history.
• ▸Weekly unplanned work %: record the current figure; the target in the passages is under 20%.
• ▸Stores service level %: record the percent of time a part is in stock when needed; target 95%-97%.
• ▸Work order / asset: identify the equipment item, functional location, or building/floor/room it is charged to.
• ▸Job scope and tasks: list what is to be done, step by step.
• ▸Labor skills and hours: list the crafts and the hours needed.
• ▸Materials and tools: list parts (with stock status) and special tools.
• ▸Instructions and safety: attach the written procedure and any permits or precautions.
• ▸Planning category: mark simple, complex, PM, or shutdown/turnaround/outage.

1. 1Check the prerequisites before building any schedule. The source's chapter on weekly scheduling lists 'Prerequisites to Effective Scheduling' and ties scheduling to a backlog of planned work. Confirm that jobs reach the scheduler as work orders, because any maintenance not recorded in the work order system is lost. A job that isn't on a work order can't be scheduled or counted.
2. 2Test MRO support first. Measure service level, the percent of the time a part is in stock when needed. The source benchmark is 95%-97%. Below that, stores and purchasing cannot support the schedule, so parts-dependent jobs stay off the schedule until the parts are confirmed.
3. 3Check the planned/unplanned ratio. The source says stores and purchasing cannot be cost effective unless maintenance is proactive, with less than 20% unplanned work weekly. If unplanned work runs higher, expect the schedule to be broken by emergencies and treat the ratio as your first improvement target.
4. 4Schedule only work that is fully planned. For each work order, confirm that parts, rental equipment, drawings, tools, and the equipment shutdown are all arranged. These are the delays the source lists: waiting on or looking for each of them. The goal is a 'pit stop' where the crew arrives and works.
5. 5Coordinate with operations before committing a job to a slot. The point of scheduling is to minimize interruption of the production schedule, so the shutdown window is part of the schedule. Agree on that window with operations, and bring maintenance, operations, and engineering together on one plan.
6. 6Build the weekly schedule from the backlog and assign it to craft labor. Match planned jobs to available craft hours and to asset availability, and sequence them so that no crew waits on another.
7. 7Execute the schedule, then close each work order with complete cost and history data. Post labor, parts, contractor, and rental costs so that the next schedule has accurate job durations and equipment history.
8. 8Measure the result with wrench time and delay causes. The source cites about 20% wrench time for reactive organizations and 60% or more for proactive, planned, and scheduled ones. Each time a crew waits, trace the delay to a missed prerequisite and fix it upstream.

The source offers a comparison rather than a single site case, so this example applies its figures. Take a reactive plant where craft technicians have their hands on tools only about 20% of the time. The rest goes to waiting on or looking for parts, rental equipment, drawings, tools, or a shutdown. First, the scheduler checks stores service level against the 95%-97% benchmark and finds parts often missing. Second, the scheduler notes that unplanned work exceeds the 20% weekly ceiling, so emergencies keep displacing scheduled jobs. The scheduler then limits the weekly schedule to work orders whose parts are staged, whose tools and drawings are identified, and whose equipment shutdown is agreed with operations. Each completed job is closed on its work order with actual costs and history. As the delays are eliminated, the plant moves toward the proactive profile the source describes, a wrench time of 60% or more and a pit-stop style of execution that keeps equipment in service.

• If a work order lacks confirmed parts, tools, drawings, or a shutdown window, keep it off the weekly schedule and send it back to planning, because it will create a waiting delay.
• If stores service level is below 95%-97%, fix stocking and purchasing before expecting schedule compliance, since parts availability caps how reliably you can schedule.
• If unplanned work exceeds about 20% of the week, treat emergency reduction as the priority, since scheduling cannot hold and stores either miss service level or carry excess inventory.
• If a job cannot be tied to a work order, do not schedule it; require the work to be recorded first so the history exists.
• If crews report waiting time, classify the cause (parts, rental equipment, shutdown, drawings, tools) and correct the planning step that should have caught it.

• Scheduling unplanned or half-planned jobs: crews show up and wait for parts, tools, or shutdowns. The tell is wrench time stuck near the reactive 20% level.
• Ignoring operations in the timetable: the equipment is not released, so the work slips and production is interrupted. The tell is jobs repeatedly rescheduled because the asset is not available.
• Scheduling around an unreliable MRO system: the plan assumes parts that are not in stock. The tell is service level under 95%-97% and rising excess inventory used as a workaround.
• Work done off the work order system: effort goes unrecorded, so durations and equipment history are wrong. The tell is completed jobs with no cost or history posted.
• Letting emergencies run the week: the schedule is rebuilt daily and never reaches closure. The tell is unplanned work above 20% each week.

• ▸Work order number and asset/location: identify the job and the equipment, functional location, or building/floor/room it is charged to.
• ▸Planned status: record whether the job is fully planned (yes/no) and who planned it.
• ▸Parts confirmed: list parts needed, stock status, and whether service level supports them.
• ▸Tools and rental equipment: list special tools or rentals and the confirmed availability date.
• ▸Drawings and documents: note the drawings or procedures attached to the work order.
• ▸Equipment shutdown window: record the date and time agreed with operations and the name of the approver.
• ▸Craft labor assigned: record the crafts, headcount, and estimated hours scheduled.
• ▸Delay log after execution: record any waiting time and its cause (parts, rental, shutdown, drawings, tools).
• ▸Closure data: confirm that actual labor, parts, contractor, and rental costs and the repair history are posted.

1. 1Establish the work order as the single hub. Every activity, from inspections and adjustments to major overhauls, and every PM task, is generated as a work order charged to an equipment item, functional location, or building/floor/room indicator. Maintenance performed without a work order is lost data.
2. 2Define one flow per work category: emergency/breakdown, simple corrective (simple planning), complex corrective (complex planning), preventive, and shutdown/turnaround/outage (STO). Document each as a flow that says who identifies the work, who plans it, who schedules it, who executes it, and who closes it.
3. 3Separate the documented flow from the CMMS/EAM default flow. Vendors set most of the system flow and allow some configuration. Add the organizational disciplines the system cannot enforce, so the process is repeatable regardless of who is on shift.
4. 4Control emergency work with its own written definition and control process, so breakdown work is not the default route for everything. Treat unplanned work as any activity with less than one week of advance planning.
5. 5Sequence the build. Confirm the PM program is effective (unplanned work under the 80%/20% limit) before adding MRO, planning, scheduling and STO processes. Then confirm MRO service level reaches 95-97%. Do not bend the best-practice steps to fit a reactive environment.
6. 6Assign the planner role: one planner per 15-20 craft technicians. Planners prepare simple and complex corrective work, PM work, and STO work so it can enter the weekly schedule.
7. 7Close every work order with labor, parts, contractor, rental and miscellaneous costs, plus repair history, posted to the equipment history file. Track backlog in hours by craft to test whether craft staffing is right-sized.
8. 8Run closure and analysis on completed orders, using problem and cause codes, to feed top-10 lists and root cause analysis. Review process KPIs against the documented flows.

The source's own diagnostic runs as a sequence. A plant states it has a work management process, but the survey logic in the text shows the usual picture: a third of organizations use no planners, and the rest have the wrong planner-to-craft ratio. Apply the checks in order. First, the PM program: if more than 80% of manpower is spent on unplanned work (less than one week of planning), re-evaluate PM before building anything else. Second, MRO: measure service level, the share of time a part is in stock when needed, and require 95-97%. Third, the work order system: confirm that a PM task, a breakdown repair and an overhaul all appear as work orders charged to an asset, with labor, parts and contractor costs posted to history at closure. Fourth, staffing: with 40 craft technicians you need about two to three planners (one per 15-20), and craft backlog must be known in hours by craft. Only then document and roll out the emergency, simple, complex, PM and STO flows. The text warns that circumventing best practices to suit a reactive shop reinforces negative behavior and sub-optimizes every later process.

• If unplanned work (less than one week of planning) exceeds 80% of total maintenance manpower, then fix the PM program before building further work management processes.
• If parts service level is below 95-97%, then expect planning and scheduling to fail; fix stores and purchasing, or accept excess inventory as the cost of poor proactivity.
• If maintenance work is done but not recorded on a work order, then treat it as lost data and make a work order mandatory for it, including PM tasks.
• If craft backlog is not tracked in hours by craft, then you cannot judge staffing; fix planner coverage first, at one planner per 15-20 craft technicians.
• If the CMMS default flow does not match your documented process, then keep the documented process and configure the system to support it where possible, rather than letting the vendor flow define the practice.

• Partial work order use: only major jobs get work orders, so asset history and life cycle cost data are incomplete. The tell: equipment histories have gaps and you cannot cost a repair to an asset.
• Relying on the vendor's system flow as the process: the CMMS flow is mistaken for a disciplined, repeatable organizational process. The tell: practice differs by shift or supervisor even though the system is the same.
• Adapting best practices to a reactive environment: planning and scheduling are bent to survive constant emergencies. The tell: unplanned work stays above the 80% limit and the processes never mature.
• Missing or mis-sized planners: a planner exists in name only or covers the wrong ratio. The tell: backlog by craft is unknown and the craft headcount cannot be justified.
• Stores unable to support the schedule: parts are not identified, located, or accurately tracked. The tell: service level falls below 95-97% or excess inventory builds up, and spare parts cost cannot be reported by equipment.

• ▸Work category: emergency, simple corrective, complex corrective, preventive, or STO.
• ▸Documented flow exists (Y/N) and location: state where the written flow is kept.
• ▸Work identification trigger: who or what raises the work and how.
• ▸Planning owner and ratio: name the planner and the current planner-to-craft ratio (target 1:15-20).
• ▸Work order charged to: equipment, functional location, or building/floor/room.
• ▸Cost capture at closure: labor, parts, contractor, rental, miscellaneous posted to history (Y/N each).
• ▸Unplanned work share: percent of manpower with less than one week of planning (limit 80%).
• ▸Parts service level: percent of time a part is in stock when needed (target 95-97%).
• ▸Gaps and owner/date: list the missing step and who closes it by when.