Key takeaways
Most production instinctively batches: do a hundred at station one, move all hundred to station two, and so on. One-piece flow does the opposite, passing each unit forward as soon as its step is done. Work moves like a steady trickle rather than in big lots that surge and stall.
The idea seems inefficient at first glance, which is why it is so often resisted, but the math of lead time favours it strongly.
Consider three steps of one minute each. With a batch of ten, station two cannot start until all ten clear station one, so the first finished unit takes about 21 minutes. With one-piece flow, the first unit moves the moment step one is done and finishes in about 3 minutes. The whole batch still takes similar total time, but lead time, the wait for the first good unit and for any given order, collapses.
Short lead time is not a luxury. It is what lets you respond to demand, hold less inventory, and find out faster whether the process is even working.
In batch production, a defect introduced at step one is not discovered until step two processes the batch, by which point a hundred bad units exist. In one-piece flow, the next unit reaches step two almost immediately, so the problem surfaces after one bad part, not a hundred. The feedback loop that statistical process control formalises is built into the flow itself.
One-piece flow struggles where steps have wildly different cycle times, where changeovers stay long despite effort, or where processes are physically batch by nature (heat treatment, curing). In those cases a small controlled batch or a buffer at the constraint is more realistic than literal single-piece movement.
One-piece flow only holds if every step is reliable and balanced, because there is no inventory buffer to hide a stop. Fabrico shows the OEE and downtime of each step in real time, so you can see which station is unbalanced or unreliable and stabilise it before it breaks the flow. The constraint view is covered in bottleneck analysis. Fabrico is built and hosted in the EU with data residency in mind and is ISO 27001 certified. To see whether your line can sustain flow, book a demo.
To turn this into a tool decision, see our overview of the best production monitoring systems.
A method where each unit is passed to the next step as soon as its current step finishes, rather than building a whole batch before moving the lot. Work flows one piece at a time instead of surging through in batches.
Because lead time is dominated by queuing, not processing. In a batch, a unit waits for all its batch-mates at every step. In one-piece flow it moves immediately, so the time for any given unit or order to finish drops sharply even though total throughput is similar.
A defect is caught at the next step on the very next unit, instead of being discovered after a full batch is already produced. That shrinks the quantity of bad product made before a problem is noticed from a batch down to roughly one.
Short changeovers (so frequent small batches are viable), balanced step times (so no station starves the others), and reliable equipment (since a stop halts the whole flow). Fixing changeover and reliability usually comes before flow.
Iniciar sesión
Blog
