Flow SimulatorOne-Piece Flow Simulator

What is one-piece flow?

One-piece flow (single-piece flow, continuous flow) means each part advances to the next process step the moment its current step finishes — it never waits for the rest of a batch. Batch production feels efficient because every machine stays busy, but busy machines are not the goal: flowing parts are. The simulator above makes the difference visible. Two identical four-station lines build the same 20 parts; the only difference is the transfer batch size. Same people, same machines, same work — the one-piece line delivers its first part 15× sooner and finishes 3.5× faster.

The waiting is easy to see once you know where to look: it is the pile of parts that builds up between the batch line’s stations. That pile is work-in-process (WIP) inventory, and every part in it is lead time — material you have paid to produce that nobody can use yet. This is the relationship captured by Little’s Law: average lead time equals WIP divided by throughput. Cut the WIP and lead time falls with it.

The penny game, rebuilt for your browser

Trainers have taught this lesson for decades with the penny game: 20 coins, a row of people, each person flips every coin and passes them on — first as one batch of 20, then in fives, then one at a time. It works brilliantly in a room and badly on a screen share. This simulator is that exercise with exact timing: the stations are your coin-flippers, the copper pucks are the pennies, and the scoreboard records what a stopwatch would. Batch size, cycle time, and playback speed are adjustable, so you can test intermediate batch sizes the tabletop version never has time for.

And because batching is not just a factory habit, the simulator ships with three scenarios: a factory machining 20 parts, a café building a catering order of 20 sandwiches (toast, fill, wrap, serve), and an office pushing 20 invoices through entry, coding, approval, and sending. The station names change; the physics do not. If you are demonstrating flow to an office or service team, run their scenario — the moment people recognize their own process, the lesson transfers.

Why batching feels efficient but isn’t

Batch logic optimizes each station in isolation: “while I’m set up, I’ll do the whole lot.” The cost appears between the stations. With a batch of 20, the first part cannot leave Stamp until all 20 are stamped, cannot leave Drill until all 20 are drilled — so the first delivery waits 5:05 while the flow line delivers at 0:20. Total completion suffers too (6:40 vs 1:55), because downstream stations stand idle while batches accumulate upstream, then get buried all at once. Watch the cumulative output chart: the flow line is a steady staircase from the first seconds; the batch line is a long flatline followed by a dump at the end. If your customers experience quoted lead times of weeks for hours of actual work, you are living on the flatline.

The remedy does not require jumping straight to a batch of one. Re-run the race at batch 10, then 5 — most of the benefit arrives long before the batches reach one piece. In practice the limit on batch size is changeover time, which is why SMED and EPEI are the practical companions to this simulation.

Round 2: batches hide defects

Speed is only half the lesson. Enable “hide a defect at Drill” and run the race again: Drill begins producing a defect that only final Test can detect. On the batch line, every part passes through Drill before the first one reaches Test — the defect is discovered after 20 of 20 parts already carry it. On the flow line, Test flags it after 3 parts, because the first piece reaches inspection in 20 seconds. Small batches shorten the feedback loop between making a problem and seeing it — the same principle behind jidoka and stopping the line at the first bad part instead of the thousandth.

Running it as a workshop exercise

The simulator is built to be projected. A 15-minute sequence that works well with production teams, office teams, and leadership alike:

1. Predict. Show the two idle lines and ask the room which will deliver the first part sooner, and by how much. Most people guess the right winner but underestimate the margin by 3–5×.
2. Race at batch 20. Run at 8× speed. Let the room feel the wait — the batch line delivers nothing while the flow line finishes its entire order.
3. Shrink the batch. Re-run at 10, then 5. Watch first-delivery time and peak WIP fall round by round, and note that most of the gain comes early.
4. Round 2. Turn on the hidden defect and run once more. Debrief the difference between scrapping 20 parts and scrapping 3.
5. Transfer.Ask: “Where in our process are we running Line A? What is our batch size — in parts, forms, tickets, or approvals?” Batching is not just a factory habit; month-end closes and approval queues are batches too.

Prefer coins on a table? The free facilitator worksheet above has round-by-round timing sheets and the same debrief questions, so you can run the physical penny game and compare your room’s times against the simulator’s.

What the simulation assumes

The model keeps every variable identical between the two lines so batch size is the only difference: four stations in series, the same fixed cycle time at each, no changeover time, no variability, and instant transfers. Real lines add setup times, uneven station loads, and variation — all of which make batching look better at first and cost more in the end. Balancing uneven station loads is its own discipline; see the Yamazumi chart for that analysis, and takt time to set the target pace demand actually requires.

Frequently asked questions

What is one-piece flow?

One-piece flow (also called single-piece flow or continuous flow) means each part moves to the next process step as soon as its current step is finished, instead of waiting for a whole batch to be completed. It is one of the core principles of the Toyota Production System because it minimizes work-in-process inventory, exposes problems immediately, and delivers the first finished unit dramatically sooner — in this simulator, 15× sooner than a batch of 20 on the same line.

What is the penny game?

The penny game (also called pass the pennies or the coin game) is a classic lean training exercise. Teams of 4–6 people "process" 20 coins by flipping them and passing them along — first in one batch of 20, then in smaller batches, then one at a time. Timing each round shows that one-piece flow delivers the first coin and the full order far faster with the same people doing the same work. This simulator runs the same experiment in your browser, with exact timings, live WIP counts, and a results scoreboard.

Why is one-piece flow faster than batch production?

In a batch system, every part waits while the rest of its batch is processed — at every step. With 4 stations, 5 seconds per operation, and a batch of 20, the first part is not delivered until 5:05, because it must wait for 19 other parts at each of the first three stations. With one-piece flow the first part arrives at 0:20 — it only ever waits for itself. Total completion improves too (1:55 vs 6:40) because the stations work in parallel on different parts instead of idling while a batch accumulates upstream.

How do I demonstrate one-piece flow to my team?

Run this simulator on a projector: let the team predict the winner, run the race at batch size 20, then re-run at 10, 5, and finally discuss the one-piece line. Follow up with the hidden-defect round to show the quality effect. Pick the scenario that matches your audience — factory (machined parts), café (a catering order of sandwiches), or office (invoices through entry, coding, approval, and sending) — so people recognize their own process. For a hands-on version, download the free facilitator worksheet and run the physical penny game with real coins; the simulator and the tabletop exercise use the same rounds and debrief questions.

Do smaller batches really improve quality?

Yes — smaller batches shorten the feedback loop between making a defect and discovering it. In the Round 2 scenario, the Drill station starts producing a defect that only final Test can catch. On the batch line, all 20 parts pass through Drill before the first one reaches Test, so 20 of 20 parts carry the defect by the time it is found. On the one-piece flow line, Test catches it after 3 parts. That is the difference between scrapping an entire order and a five-minute correction.

Is one-piece flow always the right answer?

Not always — it is the direction, not a dogma. True one-piece flow requires capable, reliable processes and quick changeovers; where changeover times are long, the practical path is to reduce them (SMED) and shrink batches progressively, using EPEI to decide how small your intervals can economically get. Even moving from batches of 20 to batches of 5 captures most of the lead-time benefit, which you can verify directly in this simulator.

Is this lean simulation free?

Yes. It runs entirely in your browser with no signup, no install, and no data collection. You can adjust batch size, cycle time, and playback speed, run the hidden-defect round, export a print-ready results summary, and download a free facilitator worksheet for running the physical penny game with a team.

Related tools

• Takt Time Calculator — set the pace one-piece flow should run at, from real customer demand
• Yamazumi Chart Builder — balance station workloads so flow does not break at the slowest step
• Kanban Card Calculator — size the pull system that connects processes where flow must break
• SMED Changeover Calculator — cut the setup times that force large batches in the first place
• Value Stream Map — find where WIP and waiting hide across your whole value stream