New Equipment vs. Old: When Does It Make Sense to Stop Repairing and Replace?
A practical look at the real cost of keeping aging equipment running versus investing in new — and why the math rarely stays the same year to year.
Ken deAlmeida
8/11/20265 min read
Short answer: when the true cost of keeping it running — repairs, downtime, and the labor spent babysitting it — starts closing in on what a replacement would cost. That point usually arrives faster than people expect, because aging equipment doesn't get more expensive at a steady pace. It spikes. A machine that cost $6,000 in repairs last year can turn into an $18,000 problem this year, not because anything dramatic happened, but because that's how wear compounds (FieldFix, 2025).
Why does repair get more expensive the older equipment gets?
A few things stack on top of each other as equipment ages, and none of them show up on the invoice for the first repair:
Parts get harder to find and more expensive. Once a manufacturer discontinues a component, aftermarket suppliers can charge 200 to 400% above the original cost for what's left in inventory (MT Unirepair, 2026).
Repairs take longer and need specialized help. Fewer technicians know legacy systems well, and the ones who do charge a premium for that knowledge — and every extra day of diagnostic or repair time is another day of lost production (MT Unirepair, 2026).
The bigger picture: older equipment costs 40 to 70% more to repair overall than newer equipment, once parts scarcity and complexity are factored in (MT Unirepair, 2026).
This isn't a fringe problem, either. The average age of industrial fixed assets in the U.S. is now around 24 years — the oldest it's been in nearly 70 years (The True Cost of Downtime, 2024). A lot of plants are running equipment well past the point where "keep patching it" was the obvious right call.
What's the "50% rule," and is it enough on its own?
The most commonly used starting benchmark in maintenance planning: if a repair costs more than 50% of what a replacement would cost, replacing is usually the better move (WorkTrek, 2026). It's a decent gut check, but it's a floor, not the whole analysis — it only looks at one repair in isolation, not the pattern building around it.
The more useful question is what the repair history looks like over the past 12 to 24 months. One expensive repair on an otherwise reliable machine isn't the same situation as three mid-size repairs in six months. The second pattern is usually the earlier, cheaper warning sign — the 50% rule just catches it later, after the number gets big enough to notice on its own.
What does downtime actually cost while you're deciding?
More than most people budget for, and the number keeps climbing. Unplanned downtime costs have roughly doubled per hour industry-wide between 2019 and 2024, and the average manufacturing facility now sees around 25 unplanned downtime incidents a month, adding up to over 300 hours of lost production a year (The True Cost of Downtime, 2024). Aging equipment isn't a side factor in that number — it's consistently ranked the single leading cause of unplanned downtime, responsible for something like 44 to 50% of unscheduled stoppages depending on the survey (WorkTrek, 2026). To add to the problem, the average time to actually fix an unplanned failure has climbed from 49 minutes to 81 minutes, largely because of skills gaps and parts supply delays (The True Cost of Downtime, 2024) — exactly the kind of delay that shows up more on older, harder-to-source equipment.
The exact dollar cost of an hour of downtime varies enormously by plant size and what's actually stopped — figures cited across the industry range from tens of thousands to hundreds of thousands per hour for larger operations. The number that matters isn't an industry average, it's yours. If you don't already know your own cost of an hour down, that's worth pinning down before this decision, not after.
What am I not counting if I only look at the repair invoice?
This is where most repair-vs-replace decisions go wrong — the visible number (the repair bill) is rarely the real number. A fuller picture includes:
The downtime around the repair itself — not just the fix, but the diagnostic time, parts lead time, and the production lost while it's all happening.
Labor spent babysitting the problem — operators working around a quirky machine, maintenance techs doing extra checks on something they don't fully trust anymore.
Efficiency loss that's easy to miss because it's gradual. Machines older than 10 years can run at roughly 15% lower efficiency than newer equivalents, even when they're technically "working fine" (Amindus Consulting, 2025) — that's capacity you're not getting credit for anywhere on a repair invoice.
Tax treatment of a new purchase. Current bonus depreciation and Section 179 rules can let you deduct a large share of a new machine's cost in the year you buy it — a real offset to the sticker price that a repair bill doesn't come with. Worth running by your accountant when you're comparing the two numbers side by side.
A real comparison stacks the honest five-year cost of continued repairs, downtime, and inefficiency against the honest cost of replacement. In one documented example, a plant's aging CNC machine was projected to cost over $50,000 in repairs over five years — against $45,000 for a replacement (Amindus Consulting, 2025). On paper, the repair option looks free until you actually add it up.
Are there cases where keeping old equipment is still the right call?
Yes, and it's a legitimate answer more often than automation vendors like to admit. Repair makes sense when:
The machine has real life left in it — a solid asset with a few good years remaining doesn't need to go just because something broke.
A qualified repair can genuinely restore it to original spec, not a patch that buys a few months before the next failure.
Replacement lead times are long enough that repair is the only realistic option in the near term — capital equipment lead times have stretched in recent years, and that timeline belongs in the decision too (Lee Contracting, 2026).
The goal isn't "always replace." It's making the call with the full cost picture in front of you instead of just the number on this month's repair invoice.
How does AutoIC help make this call?
This is a natural extension of our troubleshooting and repair work. When we're called in for emergency repair or ongoing maintenance support, we're not just fixing what's broken — we're seeing the pattern of what keeps breaking. That's the same information that should be driving a repair-vs-replace decision, and it ties directly into the kind of 5- and 10-year equipment planning we do with clients: knowing which pieces of equipment are approaching that tipping point lets you plan the replacement before it becomes an emergency, instead of after.
What's AutoIC's take on this?
Keep It Super Simple doesn't mean "keep the old machine running forever to avoid a purchase." It means matching your solution to your real situation — and sometimes the simplest, most cost-effective path actually is a well-timed replacement, not one more repair. The plants that get this right aren't the ones that never repair anything; they're the ones that know, honestly, when the math has flipped.
If you're staring at a repair quote and wondering whether it's worth it, that's exactly the kind of question worth a second opinion. Check out our FAQ for more on our troubleshooting and maintenance support, or get in touch and we'll help you look at the real numbers.
Solutions
Expert automation for manufacturing process improvement.
Consulting
Support
ken@autoicllc.com
603-748-6216
© 2025. All rights reserved.
