Preventive maintenance on the presses
Maintenance

Preventive maintenance on the presses

Financial impact

$62,000USD/ yr EBITDA

Operational impact

~700fewer breakdown hours / yr
medium confidence

The math

Uptime you keep on booked work+$38,500

700 hrs × $55/hr

Plus the repair bill you avoid$62,000/ yr

$38,500 + $23,500

Capex

$40,000

Payback

8 months

Difficulty

3 / 5

Time to implement

~14 weeks

Breakdown hours avoided
hrs
Repair and overtime saved
$
CMMS, spares, and sensors
$

Also uses value of a freed press hour $55/hr. Edit on the dashboard.

Description

The shop fixes machines when they break, usually at 2 AM in the middle of a run. There is oil pooling under the presses, heater bands fail mid-job, and nobody keeps a maintenance log. In shops like this, mechanical breakdowns are the highest-frequency downtime event, about an hour each and hundreds a year, and an emergency repair costs three to five times what the same job costs planned. When a hydraulic pump lets go mid-run, it also scraps the parts in process.

Evidence

  • Oil pools under the presses, and heater bands fail in the middle of a run
  • Maintenance is reactive, the culture is run it till it breaks, usually at 2 AM
  • There is no maintenance log, so nobody knows a given press's repair history
  • A breakdown waits days on a part because the critical spare was not on the shelf

Deep dive

Expand for detail

Load-bearing, card breaks if wrong

Maintenance discipline holds under production pressure.

A PM schedule is only worth what actually gets done. The failure mode is universal: the floor gets busy, the PM window is the easiest thing to skip, and the plant slides back to reacting. This is the most fragile assumption on the card. What it looks like on the floor if this is wrong: the overdue-PM count climbs month over month, cleaning only happens at changeover, and the breakdowns quietly return. The only real fix is the owner treating the PM window as protected time, scheduled into the plan.

A meaningful share of the downtime is mechanical and preventable.

PM pays when the breakdowns follow patterns you can catch on an interval or a sensor: worn seals, dirty filters, tired pumps, failing heater bands. If the real downtime is one-off electrical faults or operator errors that no schedule catches, the program prevents little. What it looks like on the floor if this is wrong: the breakdown log is full of random, unrelated faults rather than the same handful of wear items coming due. The check is to sort the breakdown history before spending.

Supportive, card weakens but survives

There is enough maintenance labor to actually execute the PMs.

A schedule with no one to run it is paperwork. The plant needs enough maintenance hours to do the planned work without starving the breakdowns that still happen during the ramp. What it looks like on the floor if this is wrong: the schedule is built but the one maintenance tech is always chasing the next breakdown, so the PMs never get touched.

The failures are predictable enough for condition sensors to warn in time.

Vibration and temperature sensors buy you lead time only when the failure builds gradually, like a bearing or a pump degrading over weeks. A part that fails instantly with no warning gives the sensor nothing to catch. What it looks like on the floor if this is wrong: a pump lets go with no rising trend beforehand, so the alarm and the failure arrive together.

PM gets skipped when the floor is busy

Risk

The PM window is the easiest thing to cut when the plant is behind, so it gets cut, and the breakdowns return right when capacity is tightest.

Early Warning

The overdue-PM count climbs and cleaning only happens at changeover.

Mitigation

The owner protects the PM window and schedules it into the finite plan as fixed, not flexible, time.

Stake

The full $62K.

Sensors installed but alarms ignored

Risk

Condition sensors only prevent downtime if their warnings turn into action. Muted or distrusted alarms are just noise on a screen.

Early Warning

Alarms accumulate with no work orders behind them.

Mitigation

Auto-generate a work order from every alarm and tune the thresholds so the alarms stay trustworthy.

Stake

The sensor half of the program.

Spares not on the shelf

Risk

A known wear part that is out of stock turns a two-hour planned swap into a multi-day wait on a supplier.

Early Warning

Emergency parts orders and jobs bumped while a press waits on a component.

Mitigation

Min/max stock on the critical wear parts for the top presses.

Stake

Downtime plus late shipments.

Intervals set by the calendar, so PM comes too late

Risk

A press on two shifts ages twice as fast as a date-based schedule assumes, so calendar PM is chronically late and misses the failure it was meant to prevent.

Early Warning

Failures happen on machines that were "up to date" on their calendar PM.

Mitigation

Trigger PM off run-hours and shot-count, not dates.

Stake

The prevention value, since late PM does not prevent much.

A. Run-hours, not the calendar

A calendar PM assumes every press ages at the same rate. It does not. A machine running two shifts racks up its "annual" hours in five or six months, so an annual calendar PM on it is half a year late by the time it comes due. Tie the trigger to run-hours or shot-count and the PM arrives when the machine has actually earned the wear, not when a date on a wall rolls around. This one change is most of the difference between a program that prevents breakdowns and one that documents them.

B. The five zones that cause most press downtime

You do not need to PM everything; you need to PM the systems that actually stop the press. Hydraulics (pumps, seals, filters, oil condition) and heaters (bands, thermocouples) are the usual high-frequency offenders; the barrel and screw, the clamp mechanism, and the safety interlocks round out the five. Building the checklist around these zones keeps the program focused on the failures that cost real uptime rather than spreading effort thin.

C. Condition sensors warn you, they do not fix anything

Vibration and temperature sensors give you lead time on the failures that build slowly, a degrading bearing or a tiring pump. They earn their keep on the worst-breakdown presses, where the breakdown history says a failure is coming, not scattered across the whole fleet. And they only work if the alarm becomes a scheduled work order. An alarm that gets muted because it gave a false alarm last week is worse than no alarm, so tune the thresholds and wire the alarm straight into the work-order queue.

D. Why this money is firmer than found-capacity

The OEE card finds new press hours that only pay if sales fill them. This card is different: the 700 hours it saves are hours you would otherwise lose on work already booked, so keeping them does not depend on the sales pipeline, and the repair-bill half is firm cost the day you avoid it. That is why the card is not marked conditional. The whole risk sits on discipline: protect the PM window when the floor is slammed, because that is exactly when a breakdown hurts most and exactly when the temptation to skip the PM is highest.

What we did not have in the corpus and would request from the client on day one. The aggregated gaps across all 15 cards become the engagement-letter ask sheet.

  1. 01

    Maintenance history and work orders

    Owner

    Maintenance lead

    To see which presses break, how often, and how long each event runs, so the 700-hour avoided-downtime figure is anchored to this plant rather than an industry frequency.

  2. 02

    Breakdown logs by press

    Owner

    Production

    To sort the preventable, pattern-following mechanical failures from the random one-offs, which decides where the condition sensors go and how much the program can really prevent.

  3. 03

    Hydraulic oil-analysis records

    Owner

    Maintenance

    To catch a contaminating hydraulic system before it needs a full flush, the kind of catastrophic failure whose avoidance sits inside the maintenance-bill line.

  4. 04

    Spare-parts inventory

    Owner

    Maintenance / purchasing

    To set min/max levels on the wear parts whose absence idles a press for days, which is what turns a planned PM program into actual avoided downtime.