Review Industrial DevOps: Developing better products faster

Many people already know that I am working on a new book. In Product Velocity is about the development of physical products at software speed. There is surprisingly little literature on the subject. Most of the books I have found on the subject are quite old.

One book that stands out is Industrial DevOps by Suzette Johnson and Robin Yeman. The book was published in 2023, so it is also quite up-to-date.

Below is the review of Industrial DevOps.

Target group: Program managers, system architects, decision-makers in product development and coaches or consultants

Rating: ★★★★★ Practical, well-structured, extremely relevant.

Acquire: The paper version costs approx. €25 at amazon.co.ukthe e-book version is a bargain for just $6.

Authors: Dr. Suzette Johnson is an author and expert on Lean/Agile transformations in the aerospace industry, particularly at Northrop Grumman. She has helped over a hundred companies adopt agile principles and founded several initiatives to promote women and agile practices.

Robin Yeman worked at Lockheed Martin for over 26 years, where she introduced agile methods in large-scale projects ranging from submarines to satellites. She now drives agile and DevOps practices in safety-critical systems as Space Domain Lead at Carnegie Mellon's Software Engineering Institute and is pursuing a PhD in Systems Engineering.

Foreword by Mik Kersten and Dean Leffingwell: Shorten iterations, utilize value streams

The first is by Mik Kersten, whom I know as the founder of the company Tasktopor as the author of the book Project to Product. Tasktop has developed an integration hub that synchronizes data between the various development tools. But many years ago, Tasktop changed direction and focused on value stream management, i.e. optimizing the value chain in development. Tasktop has since been acquired by Planview.

Dean Leffingwell is the originator of SAFe. I personally know few people who are really convinced of SAFe. Nevertheless, many see SAFe as a practicable way to introduce agile principles in large companies on a large scale.

Away from SpaceX and Tesla

I often cite SpaceX as a case study for product development in short iterations: These companies have done impressive things, but I clearly distance myself from company founder Elon Musk. However, I often lack alternative case studies (although I now refer more often to Wagner and GM Cruise fall back)

Of course, SpaceX and Tesla also occupy a prominent position in Industrial DevOps. At the same time, however, many other companies are also used for case studies, such as Bosch, Chevron, Apple, NASA, Planet Labs, Amazon and many more.

By applying Agile practices, we created a product that has been a smash hit in the market. Without innovative physical heating technology, we wouldn't make any progress with IoT software.

Soraia Ferreira, Software Engineer supporting Bosch in the area of smart heating and heating systems

Book structure: To the point

The book can be read sequentially from beginning to end. The many case studies break up the text. But the summaries and instructions at the end of each chapter are particularly useful. Depending on the content, there is a short list of points on

• Getting Started: How do I start implementing the topic?
• Key takeaways: the most important from the chapter
• Questions for the team: What information do I need from my employees?
• Coaching tips: When I act as a coach, what do I need to consider?

In addition to a detailed bibliography, the book also contains an appendix on tools. However, this refers to conceptual tools such as value analysis, Kanban and the like.

Part 1: Transferring the success of agility to product development

The book consists of three parts. The first part deals with the following three topics:

• Chapter 1: Towards a practice of industrial DevOps
  This chapter describes the fundamentals of Industrial DevOps as a further development of existing Lean, Agile and DevOps principles. The authors show how these concepts can be combined for complex systems consisting of hardware, software and production. The aim is to deliver better products faster and more efficiently.
• Chapter 2: Advantages of Industrial DevOps
  The specific advantages of DevOps implementation in industrial contexts are explained here, particularly for cyber-physical systems. In addition to shorter development cycles and higher quality, the importance of measurably defining and consistently pursuing the desired improvements is emphasized.
• Chapter 3: Misconceptions about Industrial DevOps
  The third chapter dispels common misconceptions, for example that agility makes planning or documentation superfluous (although in my experience, word has already gotten around. Instead, it is about planning, integrated quality and continuous learning. The goal should be to continuously capture and utilize knowledge.

Principles of Industrial DevOps

Part 2 of the book is based on a total of 9 principles, which are repeatedly referred to at various points in the book. These are

• Principle 1: Organization along the value stream enables regular demos, feedback and value-oriented results across multiple product teams.
• Principle 2: Multi-level planning helps to deal with scale and complexity and to promote continuous learning.
• Principle 3: Data-based decisions are based on current metrics to control the workflow and improve it across systems.
• Principle 4: Architecture for change and speed reduces dependencies and accelerates adaptability.
• Principle 5: Iterations and queues promote rapid feedback, experimentation and continuous learning.
• Principle 6: Clocking and synchronization reduce variance and increase predictability in the development process.
• Principle 7: Early and frequent integration accelerates the creation of results in complex system landscapes.
• Principle 8: "Shift Left" stands for a test-driven way of thinking across all test levels.
• Principle 9: A growth mentality promotes constant learning, innovation and adaptability in the face of change.

Part 2: The principles and practices of Industrial DevOps

The second part takes up most of the book. It deals with concrete principles and practices that are required to successfully establish Industrial DevOps in organizations:

• Chapter 4: Aligning the organization with the value stream
  This chapter is strongly influenced by Mik Kersten's ideas and shows how companies can design their structures in such a way that they optimize value flows. Teams should be able to act as autonomously as possible in order to minimize dependencies and decouple activities.
• Chapter 5: Using multiple planning horizons
  It explains how companies can apply a product strategy to multiple levels, making it more realistic and leading to adaptable products Planning is empirical rather than purely predictive, and large systems are sensibly broken down into smaller units.
• Chapter 6: Implementing data-based decisions
  This chapter is about making progress measurable and visible, for example through demonstrations. Digital tools and regular stakeholder feedback are particularly important for data-driven work in complex systems with software and hardware.
• Chapter 7: Architecture geared to change and speed
  A modular, adaptable architecture is essential for rapid development and integration. This chapter highlights architectural decisions, standardization and the importance of evolving system designs for industrial DevOps.
• Chapter 8: Shortening iterations and optimizing value streams
  Short development cycles with small work packages enable faster results and faster learning. Queues and iterative work increase innovation and reduce risk, for example with the help of simulations.
• Chapter 9: Establishing timing and synchronization
  "Cadence" is the term I've come across quite often recently for this phase. This is about coordination through coordinated integration points, similar to the Release Trainsthat I know from the Eclipse Foundation.
• Chapter 10: Integrate early and often
  Continuous integration has long been practiced in software development and can also be applied to hardware. This reduces technical risks and improves the flow of development.
• Chapter 11: Shift Left
  Testing and quality assurance are integrated into the development process at an early stage, i.e. via Behavior Driven Development (BDD). The triangular relationship of the V-model is an enabler for automation.
• Chapter 12: Creating a growth mindset
  The human component is one of the most important, and the shift to DevOps requires cultural change. Managers must be prepared to rethink and create a safe environment for the team. This requires shared thinking models, clear objectives and acceptance of cultural change.

Part 3: Shaping the future

The third part of the book is about how companies can not only introduce Industrial DevOps, but also develop it further. The focus is on implementation, change and leadership.

Chapter 14: Hurdles and acceptance
Typical obstacles to the introduction of Industrial DevOps are identified here, such as a lack of communication, unclear responsibilities or a lack of support from decision-makers. The chapter emphasizes the importance of structured change management, a learning-oriented attitude and a clear training approach. Change will only succeed if the focus is on quality, security and people.

Chapter 13: Bringing everything together
The task now is to firmly anchor Industrial DevOps throughout the company. This is a management task in which a clear target vision is defined, the initial situation is assessed and a roadmap is developed and implemented.

Conclusion

Precisely because I am currently working on a Book on the same topic I can recommend this book to my readers with a clear conscience. Not only is it one of the most up-to-date books on Industrial DevOps, it is also excellently structured and written in a practical way.

The book is clearly aimed at managers and larger organizations. That is not to say that engineers or smaller organizations cannot benefit from it; however, those looking for an in-depth analysis of CI/CD tool chains or the technical details of simulation integration will not find it here. This is fine and does not reflect the book's ambition; but it is precisely this gap that I will address more specifically in my own book.

Overall, this book provides a strong foundation for strategic change in complex organizations. This makes it a valuable tool for accelerating product development.