Corrugator Unplanned Stops: Where the Hours Go in an Indian Box Plant

A corrugator often loses its month not in one big breakdown but in stops that happen dozens or hundreds of times and last a few minutes each: order changes, waiting for the next reel, micro-stops nobody writes down. Price every cause by how often it happens and how long it lasts, then add the restart waste. In our worked example, about 67 lost hours a month cost roughly ₹16 lakh on a sold-out line, and the two cheapest fixes are planning fixes, not repairs.

By the Mama Editorial Team. We scope camera projects for Indian factories. Last verified: 7 October 2026, against the trade, standards and gazette sources linked inline. The stop log and every plant-level figure below are labelled assumptions. Replace them with your own.

This page is about the anatomy of stops on one line. Whether to buy a second line instead is covered in second corrugator or more uptime on the first, and board waiting between stages in reels and sheets waiting.

Key takeaways

• Every cause has two numbers: how often it happens and how long it lasts. Reliability engineers call them MTBF and MTTR. Frequent-short and rare-long stops need different fixes.
• A stop costs twice: lost tonnes and restart waste. Trade guidance puts waste from order changes, splices and start-up together at 3–8% of material input.
• In our example, order changes alone take 24 hours a month, across 160 stops of 9 minutes. Cutting each by 3 minutes is worth about ₹1.6 lakh a month on a sold-out line.
• A lost hour is worth about ₹20,000 in contribution only if you could have sold the board. If the printer-slotter or the order book is the real limit, the hour costs you the waste and the overtime to catch up, not the tonnes.
• Indian paper supply adds its own causes: kraft from several mills of varying quality, moisture and badly wound reels, as Indian corrugators say on the record.

The seven places a corrugator's hours go

We leave out "no orders" and shift handovers: they are scheduling losses, not stops of a running line. The rest falls into seven buckets.

1. Order changes that stop the line. A modern corrugator can make a "flying" change for orders on the same board, moving slitter-scorer and cut-off settings without stopping. A change of grade or flute needs a slowdown or a stop and produces transition waste. On short-run work this adds up fast. A US corrugator patent filed in 1984 already noted that, with the practices of the day, "a significant amount of time is required" to reconfigure the line between orders, and set out to cut both the downtime and the waste of each change. A line without flying changes stops every time.

2. Splice failures and missed splices. On an automatic line the splicer joins the new reel to the running one without stopping. Indian machinery suppliers list automatic zero-speed splicers for 5-ply lines. When a splice fails, the web breaks and the line stops with a tangle to clear and a re-thread. Normally the bridge between singlefacer and double-backer stores enough web to keep the double-backer running during a splice; a failed splice defeats that buffer, and board made while the line slows and restarts goes to waste. Splice failures often trace to web-tension problems from sensor drift, missed calibration or environmental conditions. TAPPI's Splicer Defects and Cures (TIP 0304-31) is a troubleshooting reference.

3. Paper breaks and bad reels. A Bengaluru box maker told PrintWeek India in 2021 that "not all mills have the same quality" and blamed poor waste raw material and mill water causing moisture retention. A Delhi converter in the same series said poor winding of rolls alone causes about 0.5–0.75% waste in handling and that using too heavy a gsm in the fluting, without balancing liner, medium and moisture, led to cracking. Weak joints, cracked liner and wet edges show up on the corrugator as breaks.

4. Waiting for reels. The line is ready, the next reel is not: wrong gsm brought up, forklift busy, reel buried in the godown. Nothing is broken, so it is rarely logged.

5. Steam, temperature and glue. When steam pressure sags as speed rises, operators compensate with wrap angles, brakes, starch application or speed, and those adjustments can increase waste, warp the board or weaken the bond. Hot plates run at about 125–175 psi of steam. These stops are rarer but long. One root-cause case traced persistent warp to a failed steam trap and a gap in the maintenance schedule.

6. Mechanical and electrical breakdowns. Bearings, belts, drives, the knife, sensors. Few and long, often longer than the repair itself because of reporting, walking and parts. That split is covered in why breakdowns take so long to fix.

7. Micro-stops. Stops under three minutes: a stacker jam, a quick clear at the cut-off, a pause to check board. Sheets filled from memory at shift end miss them, yet their total is real.

MTBF and MTTR, in shop-floor words

You don't need software for these two numbers. For each cause, over a month:

MTBF = running hours ÷ number of stops of that cause MTTR = minutes lost to that cause ÷ number of stops of that cause

Strictly, MTBF counts failures and the maintenance MTTR counts repair time; for costing, stretch both to every stop cause and measure from stop to good board again, waiting included.

The two numbers sort your causes into a 2×2:

The top-left box often holds the most hours and the cheapest fixes, yet no single event in it feels like a crisis.

The worked ₹ example: one 5-ply line, one month

Assumptions (all ours, for illustration; use your own log):

• One automatic 5-ply line, two 8-hour shifts, 26 days: 416 scheduled hours.
• 230 hours of board actually running in the month, after no-order time, warm-up and handovers.
• Output of 2.9 tonnes per running hour (45 m/min × 1.65 m usable deckle × 0.65 kg/m²), the same conservative assumption used in our second-corrugator comparison.
• Contribution of ₹7 per kg after paper and variable cost, so a running hour is worth ₹20,300 if the board would have sold.
• Restart waste valued at ₹23 per kg: kraft at about USD 0.44/kg in Q1 2026 (about ₹38/kg at an assumed ₹88 to the dollar), less an assumed ₹15/kg for selling the waste as scrap.
• Stop counts, minutes and waste kg per stop are invented for illustration, not benchmarks.

Three things stand out.

First, the three frequent-short causes (order changes, waiting for reels, micro-stops) account for 42 of the 67 hours and about ₹9.9 lakh of the ₹16.3 lakh. None of them involves a broken machine.

Second, the rare-long causes carry more waste per stop. A steam or glue problem here wastes 150 kg each time, because bad board keeps coming until someone notices warp or a weak bond. Splice failures behave the same way. The wider wastage picture is in corrugated box wastage in India.

Third, 508 stops is one every 27 running minutes, far more than a memory-filled sheet catches.

What each lever is worth

Same assumptions, one change at a time:

The two planning levers are worth more than the two maintenance levers combined and need almost no capital. Grouping orders by flute and paper also cuts the number of stopping changes, so the first row is conservative.

When a lost hour is not worth ₹20,300

The ₹20,300 assumes the corrugator is your constraint and every extra tonne would have sold.

• If the line is sold out, a lost hour is lost sales. The full ₹16 lakh applies.
• If the printer-slotters are the bottleneck, board already waits in stacks. A corrugator stop then costs only the restart waste (₹2.7 lakh here) plus some congestion. Fix the conversion end first.
• If you catch up on overtime, the cost is the overtime premium plus fuel and supervision. Under the OSH Code, 2020 (in force since 21 November 2025, PIB), overtime needs the worker's consent and is paid at twice the ordinary rate of wages, meaning basic plus DA. Your state sets the quarterly ceiling, so recurring catch-up overtime has a limit as well as a price.

Stops also have a crew cost. Helpers sized for reel and order changes stand idle between them, which is the subject of corrugator crew size and helper ratios.

How to build your own stop log in ten days

• Define a stop as board not coming off the double-backer at production speed, and log every one over three minutes with start, end and one cause from the seven above.
• Count micro-stops separately for two shifts with a tally sheet at the line: count and typical length, not every event.
• Weigh the baler per shift and note which shifts had steam, glue or splice trouble. Restart waste shows up there.
• Compute MTBF and MTTR per cause, sort by hours, then by ₹.
• Repeat the count after the first fix. If MTTR for order changes doesn't move within a month, the fix didn't stick.

Run the log as fact-finding about the line, not a judgement of the crew.

What cameras can and cannot tell you here

A camera with a clear view of the double-backer exit, the stacker and the reel stands can timestamp each stop and restart, which is what MTBF and MTTR need. It can count the sub-three-minute stops a shift sheet drops, split each stop's duration into waiting for someone and being worked on, and show whether the reels at the stand matched the next order.

It cannot read steam pressure, hot-plate temperature, glue viscosity or gel point, paper moisture, gsm or splice tension. It cannot tell a weak reel joint from a threading error, or judge bond or warp: that needs instruments and the lab, and assigning the cause of a steam or glue stop stays a human job. Footage of the line also shows people, so give workers a clear notice of what the cameras are for; see the DPDP worker CCTV notice.

Where Mama fits

Mama reads the cameras a plant already has, or ones we add, and every morning sends the owner a short note on yesterday's corrugator: how many hours it ran, how many times it stopped, the longest stops and how long order changes took. It watches the machine and the material, not individual workers, and no one is named or ranked. It gives you the stop counts and durations your MTBF and MTTR rest on, not why the glue failed. To see where your line's hours appear to go, leave an email or phone in the form below and Mama will get back to you with a camera plan.

FAQ

What is a good MTBF and MTTR for a corrugator? We found no published Indian benchmark, and the numbers depend on how you define a stop. One trade guide's capacity formula uses 80% uptime as a worked example, not a benchmark. Track your own MTBF and MTTR by cause, month on month.

Why does my automatic splicer still stop the line? A splice should cause only a brief interruption that the bridge absorbs without stopping the double-backer. Failures can trace to web tension, worn parts, sensor calibration or bad reel ends. Log each one with the reel's mill and gsm. If failures cluster on one supplier's paper, the fix may be in purchasing, not maintenance.

Should order changes count as downtime? Yes, for costing. You chose the job, but standing minutes are lost tonnes all the same. Track them as their own cause.

How much board is wasted each time the corrugator stops? It varies widely by cause and line. Trade guidance puts order-change, splice and start-up waste together at 3–8% of material input. The most reliable figure for your plant is the baler weight per shift set against the stop log.