Menu
Kanban in Manufacturing: Pull-Based Production Control

Kanban in Manufacturing: Pull-Based Production Control

A kanban system controls production with signals that pull work only when it is needed, capping inventory and exposing problems. How it works and where it fits.
Kanban in Manufacturing: Pull-Based Production Control

Key takeaways

  • Kanban is a signaling system that pulls production: a downstream step asks for more only when it has consumed what it has, so nothing is made before it is needed.
  • The signal (a card, a bin, a digital trigger) caps the amount of work-in-process. When all the signals are in use, production stops, which is a feature, not a fault.
  • That cap is what exposes problems. When inventory cannot hide a slow step or a quality issue, the bottleneck becomes visible and gets fixed.
  • Kanban suits repetitive, relatively stable demand. It struggles with highly variable or one-off production, where other scheduling fits better.

What kanban is

Kanban (Japanese for "signboard") is the practical mechanism behind pull production. Instead of pushing material forward on a schedule, each step makes or moves something only when a downstream signal says it is needed. The classic form is a card that travels with a bin: when the bin empties, the card returns upstream as authorisation to produce one more.

The deeper purpose is to limit work-in-process. By fixing the number of cards, you fix the maximum inventory between steps, which keeps lead times short and cash from piling up on the floor.

How the pull signal works

The loop is simple: a downstream process consumes a container of parts, the kanban signal goes back upstream, the upstream process produces exactly enough to refill it, and no more. With no signal, there is no production. The number of signals in circulation is a deliberate choice that sets the buffer between steps. Fewer cards means leaner flow but less cushion; more cards means more cushion but more inventory. The relationship to pull versus push is covered in pull versus push production.

Why the cap exposes problems

This is the part most people miss. A capped work-in-process is uncomfortable on purpose. When inventory is unlimited, a slow or unreliable step is hidden because the buffer in front of it absorbs the variation. Cap the buffer and the problem surfaces immediately: the line stops, everyone sees where, and the pressure goes onto fixing the real cause rather than papering over it with stock. Kanban does not solve problems; it makes them impossible to ignore.

Where kanban fits

  • Good fit: repetitive production with reasonably stable demand and standard parts.
  • Poor fit: highly variable demand, one-off or engineer-to-order work, or very long changeovers (fix those with SMED first).

Common mistakes

  • Too many cards. Setting the work-in-process cap high "to be safe" defeats the purpose; the inventory hides the problems kanban exists to expose.
  • Using it on unstable demand. Kanban assumes fairly even consumption; lumpy demand breaks the refill logic and causes stockouts or overstock.
  • Ignoring the stops. When a kanban loop halts production, that is the signal to investigate, not to add more cards.

How Fabrico fits

Kanban controls flow; Fabrico shows you whether the flow is actually healthy. When a pull loop keeps stalling at one step, Fabrico's live OEE and downtime by asset reveal whether that step is a genuine constraint, a reliability problem, or a quality issue, so you fix the right thing instead of adjusting cards blindly. 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 where your flow really stalls, book a demo.

Related reading

Many manufacturers pair these methods with the best inventory management systems.

For a practical next step, compare the leading options in our guide to the best shop floor management software.

Frequently asked questions

What is a kanban system?

A signaling system that pulls production: each step makes or moves work only when a downstream signal (a card, bin, or digital trigger) authorises it. This caps work-in-process and ensures nothing is produced before it is needed.

How does kanban limit inventory?

The number of signals in circulation is fixed, and production cannot happen without a free signal. So the maximum inventory between two steps is capped by the card count. When all signals are in use, upstream production simply stops.

Why is it good that kanban stops production?

Because the stop reveals a problem that unlimited inventory would have hidden. A capped buffer forces a slow or unreliable step into the open, directing effort at the real cause instead of masking it with stock.

When should I not use kanban?

When demand is highly variable or work is one-off, the refill logic breaks down and you get stockouts or overstock. Long changeovers also undermine it; reduce those with SMED before relying on small kanban loops.

Latest from our blog

Define Your Reliability Roadmap
Validate Your Potential ROI: Book a Live Demo
Define Your Reliability Roadmap
By clicking the Accept button, you are giving your consent to the use of cookies when accessing this website and utilizing our services. To learn more about how cookies are used and managed, please refer to our Privacy Policy and Cookies Declaration