A Blister Packaging Line Loses Its Capacity at the Cartoner, Not on the Blister Machine

Rate matching: where the 35 % goes

Nobody buys a 320 blisters/min thermoformer and receives 320 cartons per minute of packed, coded, cased product. Every stage has its own sustained rate, and the losses multiply rather than add. The example below is a real configuration: two blisters per carton, ten cartons per bundle, six bundles per case.

Now multiply four stage efficiencies of 92 %, 90 %, 88 % and 90 % and you get 65 % of nameplate before a single unplanned stop. That is the arithmetic behind most "the line never reaches its quoted speed" conversations, and it is why I insist the URS states a sustained output at the case, measured over a full shift, rather than a rate at the blister machine.

Accumulation: how many seconds you actually need

The question to ask is simple: how long can the cartoner stop before the blister machine has to stop with it? Take 90 seconds, which is a realistic leaflet jam plus restart. At 240 blisters/min that is 360 blisters in flight that need somewhere to go.

A slat bank or accumulation table typically holds 60–120 blisters, which is 15–30 seconds. It is not a buffer, it is a shock absorber. A true first-in-first-out buffer of four to eight minutes keeps the line running through a 90-second stop and, critically, keeps product in order and separated instead of butted edge to edge.

Without it, two things happen. The blister machine stops with a heated web sitting in the forming station, which scorches two to three metres of film and forces a re-thread. Or the bank packs solid and blisters contact each other, which scuffs the print and can press pockets together. Both cost more than the buffer did.

Rejection logic and GMP reconciliation

Decide early whether a fault rejects a single cavity or a whole blister. Cavity-level rejection sounds precise, but on a ten-cavity blister a single bad cavity means destroying nine good ones, and every destroyed cavity has to be reconciled. Many sites write a rule instead: reject the blister when two or more cavities fail, and accept a documented, capped level of single-cavity defects.

Reconciliation is where inspectors spend their time. The count that has to close is tablets issued against tablets in good blisters, tablets in rejects, samples, line losses and returns to stores. A yield of 99.2 % with a signed investigation for the difference reads far better than a clean 100 %, because a perfect number usually means the counting is not being done.

On rework, be careful. Reintroducing blisters into cartons is rework, and rework of packaged product needs a documented, QA-approved procedure with its own checks. Many sites simply forbid it. Write your position into the URS so the supplier prices the right reject handling, including a reject bin that is lockable and countable.

Serialisation and aggregation: your regulatory affairs function owns this

I can build the coding and aggregation hardware. I cannot tell you what your market requires, because the rules differ by destination and they change. The table is a starting map, not a specification.

The hardware architecture is stable regardless of market: print the code, verify it with a camera that grades rather than merely reads, reject on failure and confirm the reject happened, then aggregate item to bundle, bundle to case, case to pallet. What changes is the data content and where it is uploaded, which is a software and regulatory question. Price the software licence separately and ask what happens to the line when the network drops.

A two-shift downtime Pareto from a real line

Below is the distribution I recorded on a two-shift European line over eleven weeks, alongside what actually moved each number.

Four of the top five causes on this blister packaging line sat downstream of the blister machine. After the changes above, availability went from 71 % to 82 %, performance from 88 % to 91 % and quality from 99.1 % to 99.3 %, giving OEE of roughly 62 % to 74 % over a fourteen-week programme. No new machine was bought.

Two sites, two different bottlenecks

Western European generic packer. Two shifts, five days, thermoformer at 280 blisters/min feeding a 140 cartons/min cartoner through an 18-second bank. The bank was the problem: any leaflet jam over about twenty seconds stopped both machines. We installed a six-minute FIFO buffer, rebuilt the leaflet pick-off and rewrote the leaflet specification with the converter. The same hardware then ran at 132 cartons/min sustained instead of 104, with OEE from 61 % to 73 %.

Southeast Asian packer. Ambient 32 °C with 75–85 % RH in the monsoon. Carton blanks absorbed moisture overnight and the erecting station jammed constantly. The fix was not mechanical: a conditioned carton store at 24 °C and 55 % RH with 48 hours of conditioning before use, an air receiver sized up to ride out compressor dips, and drives derated for 32 °C ambient. Carton-related downtime fell from 18 % to 6 % and sustained output went from 96 to 132 cartons/min.

From URS to PQ: the schedule and the scope nobody quotes

An indicative 16–28 weeks from URS sign-off to PQ completion is realistic only when the platform is a proven build with no special tooling and the buyer does its part: freeze the format list, ship your own materials to FAT, and book validation resource before the machine lands. Most projects reach 26–38 weeks because one of those three slips.

Scope items suppliers habitually leave out of a blister packaging line quotation, and I have watched every one of them become a variation order: interconnecting conveyors and the FIFO buffer, vision systems quoted as optional, a second set of change parts, validation documents in English, the software licence for line management and serialisation, training (including a return visit after three months), and a two-year spares kit. Write them into the first quotation request, not the second.

What Line Buyers Ask Us During the URS Review

How fast should we specify the cartoner relative to the blister machine?
Rate it on cartons, not blisters, and leave 15–20 % headroom over the blister machine's sustained output. Two blisters per carton at 240 blisters/min is 120 cartons/min, so a 140 cartons/min machine with realistic efficiency is comfortable and a 130 is tight.

Is a bank enough, or do we need a true buffer?
A bank is fine if your longest credible cartoner stop is under about twenty seconds. Above that, buy a FIFO buffer of four to eight minutes. The cost is small against the scorched web and re-threads you avoid.

What output number should go into the contract?
Sustained cases per shift, measured over a full shift on your materials, not nameplate blisters per minute. Attach the acceptance test method and the reject allowance to the same clause.

Do we need aggregation, or is item-level coding enough?
That depends entirely on the destination market's requirements, which your regulatory affairs team must specify. From a hardware view, aggregation is far cheaper to design in than to retrofit, so if there is any chance you will need it, price it now.

Who owns the serialisation specification?
Your regulatory affairs function owns the data content, the upload destination and the reporting format. The machinery supplier owns printing, grading, rejection and the parent-child data structure. Get that boundary in writing at URS stage.

What should we insist on at FAT?
Run your own film, foil, carton blanks, leaflets and product, at your target speed, for a defined number of hours. Insist the vision and coding systems are commissioned at FAT rather than at site, and take the code grading reports home with you.

Written by Helen Xu | Chief Industrial Application Engineer

Helen Xu is a Chief Industrial Application Engineer with 9 years of specialized experience in packaging machinery and liquid filling machine design, equipment model selection, and full production line process optimization. He focuses on delivering customized packaging & filling solutions for pharmaceutical, food, and chemical manufacturing industries, with mature practical expertise in GMP compliance, ISO 9001 quality management standards, and turnkey large-scale filling & packaging production line integration.