Illustrative sample. The plant, the weeks and the footage are synthetic.

Recover hidden press capacity
Operations

Recover hidden press capacity

Approved fixes get an update every week. Paused, Declined and Done do not. In this sample the weekly reports are already written, so changing a status here does not change them.

$120,000USD/ yr EBITDA

Extra profit a year once this is running, after what it costs to run. Profit, not revenue. This one is conditional, so it only lands if the check on the card passes.

~2,200freed press hours / yr

Press hours a year the machines sit idle or down that you could be running, counted only for the share sales can fill. An hour nobody buys is worth nothing.

medium confidence
Only pays if sales fill the freed press hours

Capex

$75,000

Payback

7 months

Difficulty

2 / 5

Time to implement

~13 weeks

The finding

  1. This plant molds plastic parts on about 26 injection presses, machines that melt plastic pellets and force the melt into a steel mold to make a part.

    • How much of its scheduled time a press actually spends making good parts is its uptime, or OEE, and before you can win back any of the time a press loses, you have to be able to see it.
    • Here that number is invisible: production counts live on a clipboard, nobody can state last week's uptime without digging, and a press that goes down at night can sit for an hour before anyone notices, so the lost hours stay hidden and the floor runs blind on the time it could win back.
    • Shop recordsProduction counts live on a clipboard and a whiteboard, and nobody can state last week's uptime without digging.
    • Walkthrough videoPresses sitting idle between jobs with the light stack showing green, and supervisors walking the floor just to see which machines are down.
    • Operator feedbackA press that goes down on the night shift can sit for an hour before anyone notices.
    • Read togetherThe clipboard counts only the parts that got made, the green lights say nothing about the gaps between them, and the night-shift stops are remembered rather than recorded; no record on this floor can say where the lost hours go.

  3. Clip a sensor onto every press so the floor can finally see which machines are running, idle, or down, and why.

    • The sensor reads each press's current draw and reports automatically what it is doing and how fast it is cycling; Guidewheel clip-on sensors run about $15K a year for the first ten machines and need no PLC wiring, which is exactly why they suit an older hydraulic fleet, with MachineMetrics and Amper as the alternatives.
    • They layer on top of the half-used ERP instead of replacing it.
    • Budget about $75K to cover the fleet the first year, plus a daily ten-minute huddle with a named owner on the top two losses, because the sensors only pay when someone acts on what they show.
    • The mechanismOnce the lost time on each press is visible and named, the floor can attack the biggest losses, and a recovered hour on a press already running sold work is output without a new machine.
      • Acting on the data recovers about five points of availability, roughly 5,200 hours, and only some of that sells in year one, so count about 42%, roughly 2,200 hours, as actually filled.
      • At a $55 machine-hour of contribution margin (the cash a sold hour leaves after the resin and the small extras) that is about $120K a year, for about $75K once, paid back in about seven months.
    • Industry evidenceShops like this run near a 71% runtime median today against a 97% top quartile that is mostly long-run molders, so a high-mix shop realistically tops out lower; five points is the de-overlapped, cautious share of that gap.
      • If sales cannot fill the hours, the same data still pays a cheaper way: cut overtime, consolidate the second shift on the presses that no longer need it, and trim the scrap the sensors also catch.

The math

Press hours that actually get filled2,184

26 presses × 4,000 hrs × 5% recovered × 42% filled

Value of the filled hours$120,120/ yr

2,184 hrs × $55/hr

Availability recovered
%
Scheduled hours per press
hrs
Share of freed hours sold
%
Sensors and onboarding
$
Active presses
Value of a freed press hour
$/hr

Deep dive

    • A recovered press hour becomes profit only when it is filled with sold work, used to bring outsourced parts back in-house, or used to run the same output in fewer scheduled hours and cut labor.
    • If none of those is real, the capacity stays on the dashboard and never reaches the P&L.
    • The whole $120K rests on this, and it is the most fragile assumption on the card.
    • What it looks like on the floor if this is wrong: uptime climbs quarter over quarter while revenue and EBITDA sit flat.
    • The check is to have the owner or sales lead name, in writing, where the first 2,000 hours go before the sensors are bought.
    • The value is not in the hardware; it is in the ten-minute daily huddle where a named person owns the top two losses and closes them.
    • These programs die when the novelty fades and the huddle slips.
    • What it looks like on the floor if this is wrong: the dashboards are still on the wall but nobody has looked at last week's losses, and the "Other" bucket has swallowed the reason codes.
    • The analytics are only as good as the reason codes.
    • If tagging a stop takes too long or the list is confusing, operators default to "Other" and the data goes dark.
    • What it looks like on the floor if this is wrong: more than 30% of downtime is tagged "Other," so you can see that a press is down but not why.
    • The fix is a short dropdown where a tap takes under ten seconds, plus a little coaching.
    • Current sensors clamp onto the power feed and infer run, idle, and down from the draw, which works on decades-old Van Dorns and new all-electrics alike, but the states still have to be verified against reality on the oldest machines.
    • What it looks like on the floor if this is wrong: the dashboard shows a press running while it is actually idling with the pump loaded, so the numbers lose the floor's trust.
    • A short calibration pass on the pilot presses settles it.
Load-bearing
The card breaks if this is wrong.
Supportive
The card weakens but survives if this is wrong.
  • Risk

    Every lever here creates press hours; none of them adds cash on its own. If there is no plan to sell, insource, or consolidate, the freed hours stay theoretical and the $120K never reaches the P&L.

    Early warning

    Uptime climbs quarter over quarter while revenue and EBITDA stay flat.

    Mitigation

    Tie the rollout to a written sales-and-absorption plan owned by the owner or sales lead, and pivot to cost capture if the hours are not selling.

    At stake

    Most of the $120K, which is why the card is marked conditional.

  • Risk

    The sensors go on, the dashboards light up, and then the daily huddle stops. Visibility without a decision loop changes nothing on the floor.

    Early warning

    Dashboards are open but no one can name last week's top loss, and the "Other" reason code dominates.

    Mitigation

    Owner-led weekly review with a named owner on the top two losses; treat the huddle as non-negotiable.

    At stake

    Essentially the entire $120K.

  • Risk

    If operators cannot tag a stop quickly, they tag it "Other," and you can see that a press is down but never why, so the losses cannot be attacked.

    Early warning

    More than 30% of downtime is tagged "Other."

    Mitigation

    A short ten-second dropdown and operator coaching, reviewed until the "Other" share drops below about 15%.

    At stake

    Roughly half the analytical value.

  • Risk

    The program works for a quarter on novelty, then the huddle slips, targets go unowned, and uptime drifts back toward where it started.

    Early warning

    Huddle attendance drops and the top-loss owners stop closing actions.

    Mitigation

    Bake the huddle and a weekly owner review into the standard routine, and hold a named champion accountable for it.

    At stake

    The gains erode back toward the starting uptime.

A. You cannot fix availability you cannot see

A press dialed in by the operator's hand and experience, not by measured data, and tracked on a clipboard hides its own downtime. The light stack is green, the operator is present, and yet the machine is idling between jobs, waiting on a mold, or stopped for a fault nobody logged. A clip-on current sensor clamps onto the power feed and reads what the press actually draws, so run, idle, and down become facts instead of guesses. Across shops like this, the presses run about 71% of scheduled time against a top quartile near 97%, though that top quartile is mostly long-run, high-volume molders and a high-mix shop with frequent changeovers realistically tops out lower. Even the gap to a realistic ceiling is capacity you already own, recoverable with no new press. The sensor is the live readout that finally makes it visible.

B. The four ways a freed hour becomes profit

A recovered hour is not money until one of these happens:

  • Sell new work into it. The highest value and the slowest, because new parts must be quoted, qualified, and tooled. And it has to be profitable work; giving machine time away just to keep the presses busy wrecks the margin you were trying to build.
  • Shrink the batch. Shorter, steadier runs let you make smaller batches, releasing tied-up work-in-process cash without needing a single new order.
  • Consolidate a shift. If the same output now fits in fewer scheduled hours, you can pull a second shift off selected presses and cut labor directly. This one does not wait on sales.
  • Insource outsourced parts. Bringing work back in-house fills the hours with demand you already control and avoids buying another press.

C. Why this is the softest number on the board

The paths above split cleanly. The cost-savings paths (overtime, shift consolidation, insourcing) land fast and with high certainty. The new-sales path lands only as fast as the pipeline fills. There is no clean published figure for how much recovered capacity converts to profit in year one at a high-mix custom molder, so the honest planning number is that roughly 40 to 60% of the theoretical gain becomes real EBITDA the first year, with cost savings first and new sales trailing. That is why the value here is discounted to about 42% filled and flagged conditional, while the maintenance and cycle cards, which cut real cost, are firmer.

D. Making it stick past the honeymoon

The hardware is the easy part. The value is the daily ten-minute huddle where a named person owns the top two losses and closes them, week after week. Pair the operational rollout with one management action above all: an explicit, written plan for who sells or absorbs the recovered hours. Without that plan the capacity stays theoretical, and the smart move if it stays theoretical for two quarters is to stop buying hardware and turn the same data toward cutting overtime and consolidating shifts instead.

The records we did not have for this read and would ask for on day one. Each one replaces an estimate on this card with your own number.

  • Owner
    Plant manager
    Why

    To baseline what the presses really run today versus scheduled, so the five-point recovery is measured against the plant's own numbers rather than an industry median.

  • Owner
    Production
    Why

    To pin down the true starting point. Runtime, utilization, and full OEE use different denominators, and picking one and holding it steady is what makes the trend trustworthy.

  • Owner
    Owner / sales lead
    Why

    This is the commercial gate. It decides how much of the recovered capacity can actually be filled in year one, which is the single biggest swing in the $120K.

  • Owner
    HR / plant manager
    Why

    To size the cost-capture Plan B: if sales cannot fill the hours, how much overtime and second-shift labor the freed uptime lets you take out instead.