SAMPLE — HARTWELL PRECISION PARTS · ILLUSTRATION ONLY
Kaufmann Consulting · Implementation Plan
Hartwell Precision Parts
Reactive maintenance costs far more in lost production than in repair bills
Brief: Hartwell_Precision_Parts_ReactiveMaintenance_Brief.html
Plan issued: May 27, 2026
High Severity
Reactive maintenance drains shipped production
Recovery Range
$169K – $450K
First cash: 60–90 days
Selected Path
Option B: Balanced
CMMS + staged spares · target 40% reactive
This Plan

Plan reference

What follows governs execution from Week 1 through full recovery: the phases, the week-by-week schedule, the metrics that prove progress, the cash model, the operating cadence, the risk register, and the five actions that start this week. It does not restate the diagnosis, compare options, or set terms — the brief does that.

The Four Phases

Phase overview

# Phase Duration Primary focus Key milestone Gate question
1 Foundation Weeks 1–2 Rank downtime, stand up the CMMS account, lock PM windows into the two-shift schedule Five worst machines live in the CMMS Baseline signed off and PM blocks scheduled?
2 Implementation Weeks 3–12 Build PM schedules, log work orders, stage critical spares, attack the worst machine by name — first cash First cash recovered (Week 12) Downtime below 18 hrs/mo for two consecutive weeks?
3 Stabilization Weeks 13–18 Confirm the PM cadence holds without owner intervention; tune spares from real failure history Reactive share holding ≤ 50% Reactive ≤ 50% for two weeks with no owner push?
4 Optimization Weeks 19–26 Reach the 40% reactive target, extend the model to the next machine tier, lock the run rate Annualized run rate confirmed

Phase 2 is highlighted because it is where the first cash appears. Durations are scaled to Option B's stated first-cash window of 60–90 days (CMMS setup 3–4 weeks, downtime results following in 8–12 weeks once the PM cadence holds).

Phase Detail

The four phases, task by task

1
Foundation
Weeks 1–2
Establish the downtime baseline, stand up the CMMS on the five worst machines, and place preventive-maintenance windows on the two-shift schedule as fixed blocks — so Week 3 begins with a system, not a guess.
1. Pull and rank 12 months of downtime by machine.
Owner: Maintenance Lead + Controller · Effort: 6 hours
Done when: a ranked list of every production machine — total downtime hours, failure category, and average repair-wait time — exists in the shared folder, with the five highest-downtime machines flagged.
2. Document the reactive-maintenance baseline.
Owner: Maintenance Lead · Effort: 3 hours
Done when: a one-page baseline records the current 68% reactive share, 22 downtime hrs/month (264/yr), $148,000 annual maintenance spend, and the count of repairs that waited 4+ hours on parts — signed off by Tom Hartwell.
3. Open and configure the CMMS for the five worst machines.
Owner: Tom Hartwell + Maintenance Lead · Effort: 8 hours
Done when: a low-cost CMMS account is live, the five machines are entered with asset IDs, and the work-order and PM-schedule modules are switched on. The rest of the floor is deliberately deferred.
4. Identify the failure modes that caused 4+ hour repair waits.
Owner: Maintenance Lead · Effort: 4 hours
Done when: a list of the specific failure modes on the top five machines that historically stretched past four hours (waiting on parts) exists, each tagged with the exact spare part(s) required.
5. Lock PM windows into the two-shift schedule as fixed blocks.
Owner: Tom Hartwell + Shift Leads · Effort: 3 hours
Done when: recurring PM blocks for the top five machines are placed on the production schedule as fixed, named events — the same way a customer job is scheduled, not "when we get to it" — and are visible to both shifts.
6. Add a downtime-logging step to the existing shift handoff.
Owner: Maintenance Lead + Shift Leads · Effort: 2 hours
Done when: every breakdown and repair on the five machines is logged at the shift handoff, and the first week of downtime entries appears in the CMMS — no new meeting created.
Machines under management0 5 worst by downtime
Downtime baselineUndocumented signed-off one-pager (22 hrs/mo)
PM windows on the scheduleAd hoc / skipped fixed recurring blocks, both shifts
Gate question: Are all five target machines live in the CMMS with PM blocks scheduled, and is the 12-month downtime baseline signed off?
→ YES: Advance to Phase 2.
→ NO: Extend 1 week and re-assess. Trigger: any of the five machines not configured, or the baseline not signed by the Owner, by end of Week 2.
2
Implementation
Weeks 3–12
Run the PM cadence, capture failure history, stage spares only for the failures that cause long waits, attack the worst machine by name — and reach the first measurable, valued reduction in downtime.
1. Build PM schedules in the CMMS for all five machines.
Owner: Maintenance Lead · Effort: 10 hours
Done when: each of the five machines has a defined PM task list, interval, and assigned operator in the CMMS, and the system is auto-generating scheduled work orders.
2. Stand up the work-order flow at shift handoff.
Owner: Maintenance Lead + Shift Leads · Effort: 6 hours over two weeks
Done when: every reactive repair and completed PM on the five machines is logged as a CMMS work order and reviewed at the handoff — two consecutive weeks show 100% of events captured.
3. Pre-stage critical spares for the 4+ hour failure modes only.
Owner: Maintenance Lead + Controller · Effort: 8 hours
Done when: spares for the specific failure modes identified in Phase 1 are on the shelf with min/max levels set in the CMMS, and the total spend is documented and approved by the Controller — no broad spares buy added to the existing $1.085M inventory.
4. Run the PM cadence and capture failure history for 6–8 weeks.
Owner: Maintenance Lead · Effort: ongoing
Done when: six consecutive weeks of completed PM work orders and logged failures exist in the CMMS for all five machines, with PM completion at or above 90%.
5. Attack the #1 downtime machine by name.
Owner: Maintenance Lead + Tom Hartwell · Effort: 6 hours
Done when: the highest-downtime machine has a root-cause note drawn from its CMMS failure history, and a specific corrective PM or repair is scheduled and completed.
6. Produce the first machine-uptime report from CMMS data.
Owner: Maintenance Lead + Controller · Effort: 3 hours
Done when: a report showing downtime hours by machine for Weeks 3–10 against the baseline exists, with the reactive share visibly trending below 68%.
7. Verify the first downtime reduction and value it in cash.
Owner: Controller + Bruce Kaufmann · Effort: 3 hours
Done when: monthly downtime hours for the five machines are compared to the 22 hr/mo baseline, the reduction is valued at $2,024/hr, and the first-cash figure is documented and routed to the line of credit.
Unplanned downtime (hrs/mo)22 ≤ 17
Reactive maintenance share68% ≤ 55%
PM completion rateUntracked ≥ 90%
Gate question: Has unplanned downtime on the five machines dropped below 18 hrs/month for two consecutive weeks, with PM completion at or above 90%?
→ YES: Advance to Phase 3.
→ NO: Extend 2 weeks and re-assess. Trigger: downtime still at or above 18 hrs/mo, or PM completion below 90%, at Week 12.
3
Stabilization
Weeks 13–18
Confirm the gains hold without the Owner pushing them — the PM cadence runs on its own, spares are tuned to real failure history, and the reactive share stays down through a normal hot-job week.
Milestone (Week 14): Reactive share holds at or below 50% across the five machines for two consecutive weeks, including at least one week with a rush job.
Milestone (Week 15): CMMS PM completion sustained at or above 90% without the Owner re-blocking windows.
Milestone (Week 16): Downtime reduction confirmed against the 22 hr/mo baseline and the cash-recovery rate validated by the Controller.
Milestone (Week 18): Spares min/max levels tuned from real failure history; no 4+ hour part-wait recorded during the period.
Gate question: Is the reactive share holding at or below 50% for two consecutive weeks with no Owner intervention required to keep PM on schedule?
→ YES: Advance to Phase 4.
→ NO: Extend 2 weeks and re-assess. Trigger: reactive share rising above 50%, or PM windows only held because the Owner manually enforced them.
4
Optimization
Weeks 19–26
Reach the 40% reactive target, extend the proven model to the next tier of machines, and lock the annualized run rate so the recovery is permanent, not a one-time dip.
Milestone (Week 22): Reactive maintenance reaches the 40% target across the five machines.
Milestone (Week 24): The CMMS PM model is rolled to the next tier of machines beyond the top five, using the now-proven playbook.
Final (Week 26 → Month 12): Full recovery confirmed — annualized run rate of $169K–$450K (expected case $310K), reconciled and routed to the line of credit.
Week-by-Week Schedule — Phases 1 and 2

Weeks 1–12

Week Tasks Owner Output Milestone
1 Pull 12-month downtime data and rank machines; document the baseline; open the CMMS account. Maintenance Lead, Controller, Tom Hartwell Ranked downtime list; signed baseline sheet; CMMS account live
2 Enter the five machines in the CMMS; identify 4+ hour failure modes and required spares; lock PM windows into the two-shift schedule. Maintenance Lead, Tom Hartwell, Shift Leads Five machines configured; spares list; PM blocks on the schedule Foundation complete — Phase 1 gate
3 Build PM schedules for machines 1–2; start work-order logging at the handoff. Maintenance Lead, Shift Leads PM schedules live for 2 machines; first work orders logged
4 Build PM schedules for machines 3–5; place purchase orders for critical spares. Maintenance Lead, Controller All five PM schedules live; spares POs placed
5 Receive and stage spares with min/max set in the CMMS; full PM cadence running on all five. Maintenance Lead Spares on the shelf; first full PM week logged
6 Capture failure history; produce the first reactive-vs-planned work-order split. Maintenance Lead Four weeks of work-order data; first reactive-share read CMMS fully operational on 5 machines
7 Attack the #1 downtime machine — root-cause from history, schedule and complete corrective PM. Maintenance Lead, Tom Hartwell Corrective action completed on the worst machine
8 Mid-phase review: check PM completion rate, tune PM intervals from early failure data. Maintenance Lead, Bruce Kaufmann PM completion ≥ 90% confirmed or corrective step set
9 Continue the cadence; corrective action on the second-worst machine. Maintenance Lead Corrective action completed on machine #2
10 Produce the first machine-uptime report against the baseline. Maintenance Lead, Controller Uptime report, Weeks 3–10 Downtime trending below baseline
11 Reconcile the downtime reduction; value the first cash at $2,024/hr. Controller, Bruce Kaufmann First-cash figure documented
12 Route the first recovered cash to the line of credit; hold the Phase 2 gate review. Tom Hartwell, Controller, Bruce Kaufmann First cash on the LOC; gate decision recorded FIRST CASH — Phase 2 complete

Phases 3 and 4 — milestone level

Phase 3 · Week 14: Reactive share ≤ 50% for two consecutive weeks, including a rush-job week.
Phase 3 · Week 16: Downtime reduction confirmed against baseline; cash-recovery rate validated by the Controller.
Phase 3 gate: Reactive ≤ 50% for two weeks with no Owner intervention to hold PM on schedule.
Phase 4 · Week 22: Reactive maintenance reaches the 40% target across the five machines.
Phase 4 · Week 24: CMMS PM model extended to the next machine tier.
Phase 4 final (Week 26 → Month 12): Full recovery confirmed — $169K–$450K annualized run rate (expected $310K).
How We Track Success

Metrics architecture

Metric Baseline Phase 2 target Full target How measured Leading / Lagging Review cadence
Unplanned downtime (hrs/mo) 22 (264/yr) ≤ 17 ~ 10 CMMS downtime work-order logs, summed monthly across the five machines Lagging Weekly trend + monthly
Reactive maintenance share (%) 68% ≤ 55% 40% CMMS — reactive work orders ÷ total work orders for the five machines Lagging Monthly
PM completion rate (%) 0% (PM untracked, skipped "often") ≥ 90% ≥ 95% CMMS — completed scheduled PM work orders ÷ generated Leading Weekly
Repairs waiting 4+ hrs on parts (#/mo) measured in Phase 1 (needs verification) ≤ half of baseline 0 CMMS work-order wait-time field Leading Weekly
Critical spares in stock for 4+ hr failure modes (%) 0% 100% 100% CMMS min/max inventory against the identified failure-mode parts Leading Monthly
Reactive repair premium ($/yr) $28,754 trending down ~ $17,000 Controller — maintenance invoices coded reactive vs. planned Lagging Monthly
Cumulative cash recovered (production returned to revenue) $0 ~ $20K $169K–$450K / yr run rate (expected $310K)

Leading vs. lagging — what to watch and when

Leading indicators — PM completion rate, repairs waiting 4+ hours on parts, and critical-spares availability — predict future cash because they are the inputs to uptime: when PM is completed on schedule and the right spares are on the shelf, breakdowns shrink before the downtime number moves. These are the metrics the Owner and Maintenance Lead review at the weekly check-in, alongside the downtime trend, because a slip here is the earliest warning that next month's cash is at risk.

Lagging indicators — unplanned downtime hours, reactive share, the repair premium, and cumulative cash recovered — confirm what already happened. They are reviewed in the monthly summary, where the Controller reconciles the downtime reduction into dollars at $2,024/hr and routes the recovered cash to the line of credit.

Cash Impact Model

Three scenarios

Conservative (Floor)

Slower adoption, partial PM compliance, one major obstacle along the way.
Month-by-month: Months 1–2 $0 (setup) · Month 3 first cash ~$5K · ramping to ~$14K/mo by Month 6 · holding the run rate Months 6–12.
Year 1 total: $144K
$169K / yr
annualized run rate

Expected (Base Case)

Normal implementation pace, typical adoption, no major obstacles — this is the Option B target.
Month-by-month: Month 1 $0 · Month 2 ~$4K · Month 3 ~$12K · ramping to ~$28K/mo by Month 6 · holding through Month 12.
Year 1 total: $310K
$341K / yr
annualized run rate

Upside (Best Case)

Strong adoption, faster-than-expected behavioral change, favorable conditions.
Month-by-month: Month 1 ~$3K · Month 2 ~$12K · Month 3 ~$24K · reaching ~$45K/mo by Month 5 · holding through Month 12.
Year 1 total: $450K
$541K / yr
annualized run rate

Monthly milestone timeline

How We Work Together

Communication cadence

Meeting / Communication Format Frequency Duration Participants Purpose Output
Weekly progress check-in (Phase 1–2) Call / video Weekly 30 min Tom Hartwell, Maintenance Lead, Bruce PM completion, downtime trend, blockers, decisions Action items with owner and date, logged in the CMMS notes
Phase gate review Structured meeting At each gate (end of Weeks 2, 12, 18) 60 min Tom Hartwell, Maintenance Lead, Controller, Bruce Assess the gate question against data; go / no-go Written gate decision recorded in this plan
Bi-weekly metrics review (Phase 3–4) Status report + call Bi-weekly 30 min Tom Hartwell, Controller, Bruce Metrics vs. targets, reactive-share trend, cash routed to LOC Metrics snapshot sent before the call
Monthly executive summary Written report Monthly n/a Tom Hartwell (prepared with Bruce) Cash impact, downtime trajectory, overall trend One-page summary emailed to the Owner
Issue escalation Email / call As needed As needed Bruce + Tom Hartwell Obstacle resolution, scope clarification (e.g., a hot job threatening PM windows) Written decision or scope note
Final outcome presentation Meeting End of engagement (Week 26) 60–90 min Tom Hartwell, Maintenance Lead, Controller Results vs. plan, what was built, next steps Outcome report (separate document)
Roles and Responsibilities

RACI matrix

Activity Bruce (KC) Tom Hartwell (Owner) Maintenance Lead Controller Shift Leads
Pull and rank 12-month downtime dataCARCI
Document the maintenance baselineCARCI
Open and configure the CMMS accountCARC
Identify 4+ hr failure modes and required sparesCARIC
Lock PM windows into the two-shift scheduleCARC
Build PM schedules in the CMMS for five machinesCARC
Stand up work-order logging at shift handoffCARR
Pre-stage critical spares (approve spend)CARC
Attack the #1 downtime machine by nameCARIC
Produce machine-uptime report from CMMSCIRC
Verify downtime reduction and value the cashCACR
Route recovered cash to the line of creditCAIR
Facilitate weekly check-ins and phase gate reviewsRACCI
Tune spares and PM intervals from failure history (P3)CARIC
Extend the PM model to the next machine tier (P4)CARC
Document the final outcomeRACCI

Escalation. Any obstacle that threatens a PM window or the schedule — most often a rush job competing with a scheduled PM block — is raised by the Maintenance Lead or Shift Lead to Tom Hartwell, who makes the call on the floor. Anything that changes scope, spend, or the engagement timeline is raised to Bruce, who responds within 24 hours on any flagged obstacle. A scope conversation is triggered when a fix requires capital, headcount, or work beyond the five target machines and the CMMS — none of which is in this engagement without a written change to the brief.

Risk Register

Eight risks specific to this engagement

Risk Category Likelihood Impact Score Mitigation Contingency
PM stops get skipped the first time a hot job comes through — the shop already skips PM "often" to keep machines up — and reactive creeps back to 68%. Organizational High Medium High PM blocks are locked into the two-shift schedule as fixed named events; completion is reviewed at the shift handoff, not in a new meeting. If PM completion drops below 90% for two weeks, Tom Hartwell personally re-blocks the windows and pauses the lowest-priority job to protect the PM slot.
The CMMS becomes a data-entry burden no one keeps current, so the failure history is useless within a quarter. Technical Medium High High Track only the five worst machines at first, and log at the existing shift handoff — a small dataset that is actually maintained beats a complete one that isn't. If logging compliance is below 90% by Week 8, cut scope to the top three machines until the habit holds, then re-expand.
Over-buying spares trades a downtime problem for an inventory problem — the shop already carries $1.085M of inventory under LOC pressure. Data Medium Medium Medium Stage spares only for the failure modes with a documented 4+ hour wait; the Controller approves each part against carrying cost before purchase. If spares spend exceeds the approved cap, freeze further purchases and require CMMS failure history to justify any added part.
The Controller is stretched thin and the parallel collections priority competes for the same bandwidth, so cash reconciliation slips. Resource High Medium High Limit the Controller's role here to invoice coding and the monthly cash reconciliation (≤ 2 hrs/week); the Maintenance Lead owns all CMMS data entry. If the Controller cannot sustain the monthly reconciliation, Bruce performs the cash valuation directly from CMMS exports.
The single Maintenance Lead is a one-person point of failure — if they are out, PM and the CMMS stall. Resource Medium High High Assign each top-machine PM checklist to the operator who runs it most, so the knowledge isn't only in the Lead's head; the CMMS holds the failure history. If the Maintenance Lead is unavailable, Shift Leads run the posted PM checklists and log to the CMMS until they return.
Standing up the CMMS pulls the Maintenance Lead off actual maintenance, so reactive repairs spike during the rollout — on a floor running two shifts hard. Timeline Medium Medium Medium Phase the CMMS build one machine at a time and keep the existing reactive response running underneath, so coverage never drops during the transition. If reactive downtime rises above the baseline for two weeks, pause new-machine onboarding until the cadence stabilizes.
Restoring uptime exposes a different constraint — changeovers and scrap also steal capacity — so shipped output, and therefore cash, lags the downtime gain. Technical Medium Medium Medium Track parts out the door alongside downtime hours, so the next bottleneck shows up as soon as it appears. If throughput doesn't rise with uptime, flag the changeover/scheduling finding (FN-03) as the next engagement rather than over-investing here.
Recovered cash gets absorbed by day-to-day operations instead of paying down the line of credit (drawn $480K, 64%, 1.9 weeks of headroom). External Medium High High The Controller routes the valued downtime-recovery cash to the LOC monthly, tracked on the cash report — not left in the operating account where it gets spent. If the LOC draw doesn't fall as cash is freed, Tom Hartwell and Bruce review operating outflows at the monthly summary and adjust.
Starting This Week

Five actions to begin

  1. Pull the last 12 months of downtime and repair records and rank the machines.
    Owner: Maintenance Lead (with Controller). By: Day 3. Output: a ranked downtime list — hours, failure category, repair-wait time — in the shared folder, top five flagged.
  2. Document and sign off the maintenance baseline.
    Owner: Tom Hartwell. By: Day 7. Output: a one-page baseline (68% reactive, 22 downtime hrs/mo, $148K spend) signed and shared with Bruce.
  3. Open the CMMS account and enter the five worst machines.
    Owner: Maintenance Lead. By: Day 7. Output: a live CMMS account with five machines configured and the work-order and PM modules switched on.
  4. List the failure modes that caused 4+ hour repair waits and the spares each needs.
    Owner: Maintenance Lead. By: Day 7. Output: a failure-mode list tagged with the exact spare parts to stage — and nothing beyond them.
  5. Place the first PM windows on the two-shift schedule as fixed blocks.
    Owner: Tom Hartwell (with Shift Leads). By: Day 7. Output: recurring PM blocks on the schedule for the top five machines, visible to both shifts.