Real-Time Scrap Detection Software for FMCG Plants Guide

By James Smith on October 19, 2026

real-time-scrap-detection-software-for-fmcg-plants-guide

Overproduction is the waste that looks like success. A line that runs a few percent past its order shows high output and happy operators, until the extra packs reach a warehouse that did not ask for them. In FMCG the cost is real: material and labour spent on stock nobody ordered, shelf life used up on the shelf and write-offs when it expires. Most overruns are not careless. They come from buffers, long changeovers and the lack of a clear stop signal at the line. Plants that want to see how far their lines run past orders can review order-by-order overrun measured from their own production counts in a short session.

FMCG Real-Time Loss Detection · Overproduction

Stop the Line at the Order, Not a Few Percent After It

iFactory AI compares live counts with the order, warns before the target is reached and triggers the changeover on time.

Order 4417, produced against orderedAdherence 92%


Order
0%50%100%120%
Illustrative. The line ran 8% past the order before it was stopped.

Why Lines Run Past the Order

Overruns have causes, and most of them are reasonable on the day. Each cause below has a matching fix.

A yield buffer built into the order
Set the buffer from measured yield, not habit.
Fear of shipping short
Show real stock and demand, so extra output is not a safety net.
Long changeovers reward long runs
Cut changeover time, so a short run costs less.
Batch sizes larger than orders
Align batch size with order patterns where the process allows.
No clear stop signal at the line
Count against the order live and warn the crew.

What an Extra Thousand Packs Costs

The cost of overproduction is spread across several lines of the budget, which is why it is rarely added up. The bars show an illustrative split for one extra thousand packs.

Materials and packaging
45%
Labour and line time
25%
Storage and handling
15%
Write-off risk
15%
For short-life products the write-off share is the largest in practice. A pack that expires has cost everything and earned nothing.

Find Out How Far Past the Order Your Lines Really Run

Share a quarter of orders and production counts. iFactory AI will show overrun by line, product and shift, with the cost of the extra output.

Planned Against Actual, Order by Order

Schedule adherence is easy to say and hard to see without data. The columns show eight illustrative orders, with planned quantity and actual output side by side.

















Order 1Order 4: large overrunOrder 8
Planned
Actual
Most orders land close to plan. Two overruns and one shortfall account for nearly all of the variance, and each has a cause that can be found.

A Changeover Trigger That Warns Before the Target

A crew cannot stop at the right pack if the line gives no warning. Four states, tied to the live count, prepare the changeover in time.

90%
Heads-up
Supervisor and next-order materials are alerted.
98%
Prepare
Changeover tasks that can run in parallel begin.
100%
Switch
The line stops or changes over at the order quantity.
After
Review
Any overrun is recorded with its reason.

Pull Signals and Allowed Overrun

A pull signal tells the line what is wanted, and when, instead of letting it guess. Allowed overrun then depends on the product, as the table shows, with illustrative policy values. Planners can see overrun limits applied by product on their own orders in a short walkthrough.

Customer order
Demand is confirmed.
Warehouse pull
Stock falls below a set level.
Line schedule
The line makes what was pulled.
Product TypeAllowed OverrunReason
Short shelf life0 to 1%Write-off risk is high
Promotional packs0%Pack has no later use
Stable ambient range1 to 3%Extra stock can be sold later
Private labelBy contractTolerance agreed with the customer

A Composite Scenario: Overrun Cut by Half

A plant added live counts against orders and a changeover warning. The figures below are illustrative.

3.4%
average overrun before
1.6%
average overrun after six months
Fewer
write-offs on short-life lines
Most of the gain came from a few products that had always run long. Naming them was the first step, and the warning did the rest.

Where iFactory AI Fits

Orders, counters and the line schedule live in separate systems. iFactory AI joins them, so overrun is seen as it happens.

Live Count Against Order

Output is tracked against the ordered quantity in real time.

Changeover Warnings

Crews get a heads-up before the target and a prepared handover.

Overrun Reporting

Overrun and shortfall are reported by order, line, product and shift.

Cost of Extra Output

Each overrun is costed in material, time and write-off risk.

Delivered turnkey, live in 6–12 weeks
iFactory AI arrives pre-configured on an NVIDIA server that ships racked and ready with software pre-loaded. Rack it, connect power and Ethernet, and overproduction tracking begins building. Scope covers cabling, network, ERP and MES integration, team training and 24×7 remote monitoring.
Weeks 1–4
Ship, network and connect order and counter data
Weeks 5–8
Measure overrun and agree limits by product
Weeks 9–12
Go live with warnings and train planners and crews
Planner: which products overran most last month?
iFactory AI: two short-life yoghurt lines, averaging 4 percent over, and one promotional pack that ran 7 percent past its order.

Expert Review: Preventing Overproduction Without Missing Orders

Preventing overproduction must not create shortages. This review sets out how experienced planners judge the trade-off.

Expert reviewOverproduction control
What holds up
Counting live against orders, warning before the target and setting limits by product are practical controls that directly reduce excess stock.
What to verify
Real yield, customer fill-rate targets, shelf-life rules and who may authorise an overrun. Check that limits do not cause missed orders.
What to avoid
One overrun limit for every product, hiding shortfalls and blaming crews for buffers that planning created.
Bottom line: control the overrun and the shortfall together, and set limits by what each product can bear.
Written from established planning practice. Policy and customer terms are set by your organisation.

Frequently Asked Questions

Is some overproduction always necessary?

A small allowance is often sensible, to cover yield loss and quality rejects. The aim is to set it deliberately, from measured yield, and not to add it by habit. The right level differs by product. Our team can estimate your allowance from a sample of orders, so buffers are based on evidence.

Will stopping at the order risk short shipments?

Not if yield is understood. Orders should be released with an allowance for expected loss, and the live count should track good packs only. Where a short shipment would be costly, planners can set a small tolerance. Visibility of stock and demand reduces the temptation to run long as a precaution.

How does this relate to changeover time?

Closely. When changeovers are long, supervisors run extra to avoid them, so shortening changeovers reduces overproduction pressure. Starting preparation before the target is reached also saves time. See a changeover warning flow in a short walkthrough using your own line states.

Can the line stop itself at the order quantity?

That depends on your controls and policy. Many plants begin with warnings and a clear display, and some link the count to a line stop or an automatic recipe change later. Any link to line control should be agreed with your controls engineers and operations leaders, with a clear manual override.

How is the benefit measured?

By tracking average overrun, the share of orders outside limits and write-offs linked to overproduction, before and after. All three come from order and production records. Ask for a sample overrun report that planning and finance can both use.

Make the Order the Finish Line

See live counts, warnings and overrun costs on your own lines, so extra packs stop being a habit. Book a working session with an FMCG planning specialist.


Share This Story, Choose Your Platform!