Kanban and the Theory of Constraints

[#1]

Have you ever heard of the Theory of Constraints? Kanban is based on it, but what is the Theory of Constraints and how does Kanban apply it?

• Kanban: Visualize your (unvisible) work
• How Kanban applies the principles of the Theory of Constraints
  • Identifying the bottlenecks
  • Utilizing the capacity of the bottleneck
  • Adjusting the pace of work
  • Increase capacity of the bottleneck
• Conclusion

Kanban: Visualize your (unvisible) work

Kanban is a work management system originally developed by Toyota as part of the Toyota Production System (TPS). The method has been adapted by, among others, David J. Anderson for application to knowledge work. It focuses on visualizing workflows using a physical or digital board, which contains columns and cards that represent the tasks and their status. Kanban is also about limiting Work In Progress (WIP) and promoting a continuous, steady flow (Flow) of work.

How Kanban applies the principles of the Theory of Constraints

The Kanban method applies multiple principles to improve productivity. Kanban borrows a number of those principles from the Theory of Constraints (Eliyahu Goldratt, 1984).

The Theory of Constraints (TOC) is a management philosophy that focuses on identifying and addressing the constraints that limit the performance of a system. It emphasizes the importance of identifying the weakest link in a process and improving it to optimize overall system performance.

The Theory of Constraints assumes that every system, be it a production process, a project, or an organization, is constrained by at least one constraint. By identifying and understanding the limitations, targeted actions can be taken to reduce or eliminate them. The TOC states that addressing the constraints leads to better throughput, efficiency and outcomes of the system as a whole.

As an example, take a production line for a factory where several machines are involved in producing a product. Let's assume that the packaging machine is the bottleneck since it has the longest lead time and limits the throughput of the entire production process.

Identifying the bottlenecks

In the factory it will be quite easy to see that the work (in the form of intermediate stock) is accumulating in front of the packaging machine and that this causes delays. This identifies the packaging machine as the throughput bottleneck.

However, if you are dealing with (invisible) knowledge work, it will not be so obvious.
Kanban visualizes the workflow and the work in it using different columns and cards on the Kanban board. This allows us to quickly identify bottlenecks. It becomes clear which process steps are slower or where work is accumulating. This allows us to focus specifically on this.

Utilizing the capacity of the bottleneck

The principle from the Theory of Constraints then says that with the packaging machine it must be ensured that we make full use of its capacity: Products are always being packaged and the machine does not stop unnecessarily. This can be done by ensuring that the machine continues to run at night, but also during breaks or when changing shifts.

If we translate this into knowledge work, then we also want to make maximum use of the resources of the bottleneck process. By focusing on utilizing the capacity of the bottleneck, we maximize productivity in the existing system and increase throughput. This can be done, for example, by ensuring that someone is always busy with the work in the bottleneck.

Adjusting the pace of work

We then adjust the production process to take into account the capacity of the packaging machine. We make sure that the preceding machines do not run faster than the rate at which the packaging machine can process, to prevent unfinished products from piling up. This ensures better coordination in the workflow and prevents bottleneck overload.

With knowledge work as is often the case with Kanban, we use this principle from the Theory of Constraints by adjusting the pace of work throughout the workflow to the capacity of the bottleneck. Here, too, there is no point in letting work pile up. This way we prevent overload and minimize waste of resources and time.

Increase capacity of the bottleneck

It makes no sense to improve other process steps to achieve further gains; that would only lead to a larger intermediate stock. The only solution to increasing overall throughput is to increase the capacity of the bottleneck.

In the factory, this means that we want to increase the capacity of the packaging machine. By making improvements to the packaging process, such as eliminating non-value-adding steps, deploying an additional machine and/or additional staff, or automating tasks, the capacity of the packaging step is increased and the overall productivity of the production process is increased. improved.

In knowledge work, the solution is similar: Optimize the critical process step by removing the steps that do not add value, deploy more people or automate parts of that step.

Conclusion

Kanban applies the principles of the Theory of Constraints regardless of domain. By identifying bottlenecks, making full use of them, aligning all operations to the bottleneck, and increasing the capacity of the bottleneck, Kanban optimizes system productivity and throughput. The production line example illustrates how improving the packaging stage increases the overall efficiency of the production process. By applying these approaches, organizations can optimize their workflows and achieve their goals more efficiently, regardless of the nature of their business activities.

Source: Bart Schrap, Agile Coach at the Agile Scrum Group

[Bill Sheboy]

Yes, and...to further these ideas and learn more about improving flow, consider some additional reading like The Principles of Product Development Flow: Second Generation Lean Product Development, by Donald Reinertsen.

[Omer Meshar]

Totally agree.

I would add that two methods in Kanban, done right, are implementations of the idea of subordinating and synchronizing the system to the constraint, coming from TOC (see https://www.leanproduction.com/theory-of-constraints/)

1) The method of "pull" in Kanban, instead of "push", guarantees that the system does not work harder than needed by the constraint.

2) The method of stopping everything and focusing on resolving an issue that is causing delays to the constraint, again guarantees we are focusing on the constraint and making sure it does not starve.

[Dave Mathijs]

Thanks for the additions @Bill Sheboy & @Omer Meshar !