More conveyor, or fix the jam?

Try it · 4 clicks · each number is the average of 200 simulated shifts

Click a step. The line below plays it.

More conveyor buys time. It doesn't stop the jams. Fix the cause.

Seed

0 of 450 min · running

Running Starved (nothing coming in) Blocked (downstream full) Down (failure) Line stop

Output

Good bottles out

Bottles made0

Rejects0

Cases packed0

Line efficiency

OEE so far this shift

–

Availability–

Performance–

Quality–

Downtime

Logged stop minutes

0 min

No stops yet

Stops0

Biggest loser–

Filler not producing0 min

Bottleneck

Right now (last 5 min)

–

This shift–

Its time starved or blocked–

Runner-up–

Where each machine's time went

Machine states this shift

Machine	Share of shift so far	Running	Starved	Blocked	Down	Stops

Minutes. A machine faster than the one feeding it shows a steady share of starved time: it is waiting for product, not broken.

Like the operator's downtime log

Stop log

What-if · full shifts

What is each option worth?

Shifts per scenario

ScenarioGood bottles per shift · average, with P10–P90 rangeAverageP10–P90OEEvs first row

Your line

Run the same test on your own line

5 minutes · no data needed1. Type in your line

Machines in order, their rated speeds, conveyor sizes, and roughly how often each one stops. Then compare a bigger buffer, a faster machine or a fix.

Build your line

10 minutes · your downtime log2. Upload your downtime CSV

Any export with one row per stop: machine or cause, and minutes. The stop rates are worked out in your browser. Nothing is uploaded anywhere.

Messy data · Claude3. Use the free Claude skill

Give Claude your CMMS or MES export, even a messy one. It maps the causes to machines, builds a simulator page for your line, and lists every number it had to guess.

Get the skill

Before you buy conveyor or a faster machine: see what it is worth in bottles first.
When people argue about the bottleneck: the simulator shows which machine everyone else is waiting on.
To justify a root-cause fix: turn the downtime log into bottles a shift, which is easy to put in dollars.
To train supervisors: starved, blocked and down are easy to understand once you watch them happen.

"Copy share link" puts your whole line in a link, so your team opens the same simulation. Code is MIT, free to use and change.

Your line

Build your line

2 to 10 machines in line order. Any change restarts the shift. Your line is kept in this browser and in the share link; nothing is uploaded.

Machines and conveyors i

Stops that hit the whole line i

Check the numbers

Data, method and assumptions

Stop source	Stops in data	Stops / shift data	Stops / shift simulated	Avg stop min set	Avg stop min simulated

Default line: Plant 2 dataset by Ivan Getov (CC BY 4.0), a fictional plant. A simulation is a way to think, not a forecast: check it against your own line before you spend money. Code: MIT. tools.getov.xyz →