The following statements and questions come up very often when introducing agility and Scrum in hardware development:

Every work package is different in length and that cannot be predicted!
We have so many dependencies on suppliers and cannot predict delivery times.
Our work packages are much larger than 4 weeks. If anything, our iterations would have to be 6 months long!
What is the point of fixed, maximum 4-week iterations in hardware development? Kanban is more suitable for us. Then work packages can be any size.

In this article, we want to discuss the last question in more depth, which will also clarify some of the other questions.

The following advantages result from the fixed short iterations in Scrum:

1. Better predictability and rhythm

Fixed iterations create a regular cycle (every 2, 3, or 4 weeks) in which results are expected and the team’s own approach is reviewed in retrospect.
This cycle creates planning security for everyone involved, allowing review appointments with stakeholders to be planned months in advance, for example.
Teams get used to realistic planning, e.g., by formulating acceptance criteria and estimating the effort required for the work. This teaches them what is actually feasible in an iteration.

2. More frequent review of results and processes (inspect & adapt)

Short iterations force you to review results regularly – not just at the end of the phase. This forces developers to think and plan in small steps.
Early reviews with stakeholders show whether you are on the right track.
Problems and risks (technical, organizational, supplier-related) are identified early on before they become costly.
Your own work process is reviewed and adjusted. This is a massive advantage, especially when introducing the new way of working.

3. Faster value delivery

In hardware development, too, incremental results can be delivered – such as concepts, functional prototypes, tested modules, or subsystems.
Customers or users see progress earlier, which builds trust and enables feedback.
Early visible results help to dynamically adjust priorities.

4. Increased learning rate

Each iteration is an experimental cycle: hypotheses (e.g., regarding functions, materials, manufacturing processes) are tested.
Errors are quickly discovered and used to learn from—instead of avoiding them and bundling them at the end.
Risks are reduced iteratively instead of waiting for a large integration step.

5. Improved integration of disciplines

Joint iteration planning forces mechanics, electronics, and software to work in sync instead of in silos.
In interdisciplinary teams, subject matter experts learn from each other.
The team develops a common understanding of the product and its progress.
Dependencies and interface issues become apparent earlier.

6. Transparency and focus

Iterations make progress visible – for the team, management, and stakeholders.
The focus is on the most important goals of the current iteration, not on long-term speculation.
Clear timeboxes reduce “gold plating.”

7. Motivation and team dynamics

Short cycles generate regular successes.
Teams experience their effectiveness directly and remain more engaged.
Joint retrospectives promote continuous improvement and self-organization.

8. Improved capacity planning

Fixed iteration lengths allow team velocity (i.e., average delivery capability) to be observed.
Overloading and unrealistic plans become visible and addressable.
This also allows deliveries to external teams and suppliers, as well as hardware resources, to be optimized.

9. Basis for incremental construction

Even though physical products are more difficult to change, iterative work enables:
- Virtual prototypes, simulations, and 3D printing in early phases,
- Modularization of development,
- Early integration tests.

This makes product development more agile without sacrificing technical maturity.

Comparison with classic planning and Kanban

Here is a structured comparison and discussion of the differences between

plan-driven approach (classical),
Kanban (without fixed iterations),
Scrum (with fixed iterations):

A. Basic principle and conceptual model

Approach	Basic idea	Planning logic
Plan-driven	The future is predictable. Detailed planning at the outset can help avoid deviations.	A large, sequential plan with milestones and phases (e.g., planning → design → test → production).
Kanban	Work should flow continuously. Bottlenecks and overloads are made visible in order to improve the flow.	Tasks are pulled when capacity becomes available. No fixed iteration limits.
Scrum	The future is unpredictable. Learning happens through regular experimentation and adaptation.	Planning, implementation, and reflection take place in fixed-time iterations (sprints).

B. Time frame and planning horizon

Approach	Timely structure	Planning
Plan-driven	Project phases often last months or even years.	Long-term, detailed in advance. Changes are considered disruptions.
Kanban	No fixed time frame. Continuous processing.	Planning is dynamic and happens “on demand.”
Scrum	Fixed short iterations (max. 4 weeks).	Within the sprint: stable plan; after each sprint: adjustment based on new knowledge.

Advantage of Scrum: Iterations provide a deliberate rhythm for learning, similar to a test cycle in hardware development.
With Kanban, learning tends to be continuous and unsystematic, and with the classic approach, it usually comes too late (at the end of a long phase).

C. Dealing with uncertainty and learning

Approach	Dealing with uncertainty	Learning happens …
Plan-driven	Minimization through early analysis and control.	Learning usually at the end (e.g., in tests or field trials).
Kanban	Adjustment ongoing, but without fixed “stop-and-think” moments.	Learning happens ad hoc when problems arise.
Scrum	Uncertainty is factored in: each iteration is a learning cycle.	Learning is ritualized (review & retrospective).

D. Team Organization and Responsibility

Approach	Who plans and controls?	Role of the team
Plan-driven	Project management controls and monitors.	Teams carry out tasks.
Kanban	Teams control flow of work (via Work-in-Progress Limits).	Self-organization, but less common goal orientation.
Scrum	Teams jointly manage iterations and goals.	High level of personal responsibility; clearly defined roles (Scrum Master, Product Owner, development team).

E. Measuring progress

Approach	Measured variable	Significance
Plan-driven	Meeting planned or milestone deadlines.	Progress can be illusory when late integration causes problems.
Kanban	Throughput, cycle time, work in progress.	Good focus on efficiency, but less on goal achievement or value.
Scrum	Completed, tested increments per sprint.	Focus on truly useful results, not just activity.

F. Applicability to physical product development

Approach	Strengths	Weaknesses
Plan-driven	Good for known, stable requirements, technologies and processes.	Cumbersome, risky in terms of innovation or uncertain markets.
Kanban	Easy to get started, especially for mature processes (e.g., change management, series maintenance).	Little structure for learning loops in early stages of development.
Scrum	Ideal for iterative learning and prototyping (e.g., in the concept or sample construction phase).	Requires discipline and adaptation to deal with long hardware lead times.

Conclusion for hardware development

Plan-driven: Good when everything is known and stable – bad when you have to explore new territory.
Kanban: Good for stabilizing existing processes – less suitable for exploratory learning.
Scrum: Good for developing new solutions iteratively.

Title pictured by pikisuperstar onFreepik