Agile Hardware Development: What do SpaceX and dormakaba have in common?

12 Oct, 2021
Xebia Background Header Wave

It might be unusual to compare a high-tech, leading-edge program like SpaceX with dormakaba. However, there are clear parallels in the (new) agile hardware development approach for both companies. Let's take a look at the two protagonists.

(Agile) Hardware development: leading in different areas

SpaceX, founded by serial entrepreneur Elon Musk, became the first privately funded group to put a payload in Earth orbit, in 2008. Since then, the company has continued to impress. Launching unmanned and manned cargo vehicles to the International Space Station (ISS) and winning a contract from NASA to fly astronauts as well, as early as 2017. In December 2020, for the first time in history, a rocket was lauched which later landed upright, setting itself down under the power of its own engines.

Dormakaba on the other hand is one of the top three companies for access control and security solutions in the global market. They provide everything you need for secure access to buildings and rooms. This with over 150 years' experience and millions of products installed worldwide.

Different impressions of dormakaba solutions: (Agile) Hardware development
Dormakaba is one of the top three companies for access control and security solutions in the global market. (Bild: dormakaba)

Agile learning and design principles

The parallels between the two companies can be explained along two major principles from agile development: «fast learning cycles and set-based versus point-based solutions».

Scrum Sprints are based on the Plan-Do-Check-Act/Adapt cycle (W. Deming) which enables fast learning and fast adaptation. Fast feedback cycles identify errors early, thereby reducing risk, and also open doors to exploit unforseen opportunities. The shorter the cycle, the faster the learning. This process helps the team to make small changes, get feedback, adapt and move ahead. 

The second principle is set-based versus point-based solution creation. In a traditional project, the milestones might force the design team to work towards «the final solution design» relatively early in the project. If integration and tests reveal weaknesses in the design in a later stage, it might be very hard to countersteer.

The traditional project set-up does not cope well with multiple iterations. Changes at the end of the project are most often costly and problematic. With a set-based design, specifications are kept flexible. This enables the team to evaluate multiple solution designs during the course of the project. Over time, the final design will emerge from a set of solution designs, created with e.g, rapid prototyping, 3D printing and/or extensive computer models and simulations. Because there is no «End of Design» Milestone in the Agile Framework, teams are not forced into one specific solution design too soon. This flexibility allows them to have a continuous solution design method, permitting changes throughout development.

Agile principles in action

The SpaceX rocket development is a perfect example of set-based design. SpaceX has astounded the world by its rapid pace, but also by its failures (read: learning moments!). SpaceX deployed new rocket models in a very rapid pace. They were trying out different solutions for e.g. re-entry and landing the rockets on floating platforms. The designs evolved based on the fast feedback from actual flights. 

There are a few important differences compared to the traditional approach followed by the NASA Space Agency. NASA worked with lengthy specification, build, intregration and test phases. Extensive testing was undertake to ensure the solutions created matched the highest safety standards. So nothing would go wrong that could cost human lives. On the downside, this approach was very time-consuming. SpaceX on the other hand has no qualms with things going wrong. As long as they can collect enough test data for future improvements and no one gets hurt in the process.

Dormakaba introduced agile hardware development in 2020. With the aid of CAD-Computed Aided Design, 3D printing and extensive computer simulation models, the Hardware development speed in the (pilot-)project has significanty increased. The new way of work allowed the team to explore multiple options prior to decisions about the final design, e.g. the final module lay-out. This enabled the team to evaluate options in a very early stage of the project, accelerating the development significantly.

Fast-feedback loops with the stakeholders ensured that fast learning was directly used in the next iterations of the Hardware development. By using the new Scrum Team set-up in combination with the set-based design approach, a newly developed advanced door mechanism MVP was ready for production within a year. This is a big step forward compared to similar projects from the past in terms of development speed and customer centric design.

The combination of a set-based design approach with a Scrum Team set-up created a new methodology for Hardware development.

By bringing all the necessary skills together in one team, the first integrated agile hardware development team was born. Hardware, mechanics, electronics, firmware and software disciplines worked closely together on the new design. Production preparation was also part of the team to ensure that the solution could indeed be manufactured.

To sum it all up

While there is a big difference in the nature, culture and history of both companies, SpaceX and dormakaba are both adopting the same principles in agile hardware development. The goal is the same: to increase the development speed through faster learning. 

Although not all agile practices originating from software development can be applied to agile hardware development one-to-one, important principles of Plan-Do-Check-Act/Adapt cycle and the set-based solution design can be applied succesfully. The biggest challenge in moving to shorter hardware development/solution cycles and creating an atmosphere of learning is not only to apply the agile principles to Hardware development, but also to change employees’ mindset towards this new approach. Innovation is not possible without failure.

Interested in finding out more? Get in touch with me to discuss a possible cooperation around the theme of Agile@Hardware Development.


Get in touch with us to learn more about the subject and related solutions