Scrum For Hardware Case Studies

Introduction

The traditional hardware development paradigm—characterized by lengthy development cycles, expensive tooling, and resistance to change—stands in stark contrast to the agile, iterative world of software development. Many people believe that the physicality of hardware product development means it is not well-suited for an agile approach, and that agile should be left to software development alone.

The convergence of digital technologies with physical products has created new opportunities for applying software development methodologies to hardware. The automotive industry, in particular, continues to grapple with complex issues such as verification, validation, and the integration of software development with system engineering and safety. Team Wikispeed’s breakthrough demonstrates that these challenges can be addressed through systematic application of agile principles.

Genesis and Challenge

When Joe Justice embarked on the Wikispeed project for the Progressive Automotive X Prize, he faced constraints that would have seemed insurmountable under traditional development approaches. The automotive industry typically requires 2.5 years for fast-track projects, yet Justice had to develop a road-legal, 100 mpg car in a fraction of that time with limited resources.

His solution was radical: apply the same agile development principles that had proven successful in software to hardware development. This wasn’t merely an adaptation—it was a fundamental reimagining of how complex physical products could be conceived, developed, and manufactured.

The results were extraordinary and well-documented. Wikispeed achieved seven-day sprint cycles while maintaining compliance with road-legal safety specifications. When a B2B supplier in China created over 24 agile teams, it reduced its average time-to-market for new products by 20 percent within two years of implementing the change. This demonstrates the scalability of the approach beyond individual projects to enterprise-level transformation..

Modular Design Principles

The cornerstone of Wikispeed’s approach was modularization—a design principle that subdivides a system into smaller parts called modules, which can be independently created, modified, replaced, or exchanged with other modules or between different systems. This approach directly addresses the fundamental challenge of hardware development: managing complexity while enabling rapid iteration.

Modular design can be characterized by functional partitioning into discrete scalable and reusable modules, rigorous use of well-defined modular interfaces, and making use of industry standards for interfaces. Justice explained this philosophy: “This is the only way I know to think about a complex system, to break it into loosely coupled pieces. Object-oriented architecture, that is, splitting the car into separate modules that can be worked on independently, is a pre-requisite for reducing the costs of rapid iteration.”

The Nine-Module Architecture

Wikispeed established nine main modules, each with clearly defined interfaces:

1. Suspension – Eight fixed bolt positions to chassis, everything else modifiable
2. Side crush structure – Standardized connection points to frame
3. Front crush structure – Modular safety system
4. Rear crush structure – Independent safety module
5. Frame – Central connecting element
6. Interior – User experience module
7. Drive train – Power system module
8. Pedal plate – Control interface module
9. Body – Aesthetic and aerodynamic module

This modular approach enabled what Justice termed “loose coupling”—allowing rapid changes of entire sections without requiring modifications to other areas. Each module complied with defined interfaces, ensuring compatibility while enabling innovation.

Interface Design

Unlike software, hardware designs correspond to physical circuits and inherit constraints from them. Timing constraints—which cycle a signal arrives, when an input is read—and structural constraints—how often a multiplier accepts new inputs—are fundamental to hardware interfaces. Wikispeed addressed this by establishing clear mechanical and electrical interface standards.

Justice used the analogy of plastic building bricks: “Plastic building bricks come in all shapes, sizes and colors, but if you stay within the same building system you can swap them freely and build many different creations. Imagine how slow it would go if you had to make a mold and form a new type of brick each and every time you change your creation.”

Benefits

Modularity offers benefits such as reduction in cost (customization can be limited to a portion of the system, rather than needing an overhaul of the entire system), interoperability, shorter learning time, flexibility in design, non-generationally constrained augmentation or updating (adding new solution by merely plugging in a new module), and exclusion.

Research confirms these benefits in practice. Modular hardware design is everywhere and has been on the rise in the embedded world. The drastic rise of the Arduino and Raspberry Pi ecosystems has encouraged greater use of a modular architecture. This trend validates Wikispeed’s approach and demonstrates its broader applicability across hardware domains.

Foundational Framework

Wikispeed’s success was built on adapting core Scrum principles to hardware development challenges. Unlike in agile software development, however, in agile hardware development complete product increments cannot be delivered after each sprint. This is because the features of a physical product usually only become usable at the end of the development cycle. Despite this constraint, the team successfully implemented all ten Scrum principles:

1. Organize work in short cycles – Seven-day sprints replaced traditional lengthy development phases
2. Management doesn’t interrupt the team during a work cycle – Teams were protected from external disruptions during sprints
3. The team reports to the client, not the manager – Direct customer feedback drove development decisions
4. The team estimates how much time the work will take – Bottom-up estimation rather than top-down mandates
5. The team decides how much work it can do in an iteration – Sustainable pace determined by the team
6. The team decides how to do the work in the iteration – Technical decisions made by those closest to the work
7. The team measures its own performance – Self-monitoring and continuous improvement
8. Define work goals before each cycle starts – Clear objectives for each sprint
9. Define work goals through user stories – Customer-focused requirements capture
10. Systematically remove impediments – Continuous improvement of the development process

Sprint Duration and Adaptation

Sprints for hardware developers are usually twice as long as those for software developers, i.e. 4 weeks instead of 2. However, Wikispeed achieved even more aggressive timelines with weekly sprints, demonstrating that with proper modularization, hardware development can match or exceed software development speeds.

The team’s approach to sprint planning incorporated the physical constraints of hardware development while maintaining agile responsiveness. Each sprint focused on specific modules or interfaces, allowing parallel development across teams while ensuring integration compatibility.

Extreme Manufacturing

Joe Justice coined the term “extreme manufacturing” (XM) to describe their approach, drawing inspiration from extreme programming (XP) principles developed by Kent Beck. XM represents a fundamental shift from traditional manufacturing paradigms that optimize for repeatability to systems that optimize for adaptability and responsiveness.

The core philosophy of XM is delaying final design decisions as long as possible. This can be achieved through two primary strategies:

1. Progressive requirement lock-down – Gradually fixing specifications as development progresses
2. Maintaining backup approaches – Keeping alternative design paths alive during decision-making

Test-Driven Development in Hardware

In development of hardware IP and FPGA base products where target technology is malleable, practices considered fundamental to agile development may be highly applicable. These practices would include iterative development and continuous deployment among others.

XM emphasized running tests from the very start of each sprint. Teams began by defining appropriate tests for any goal—whether road-legal safety, comfort, or efficiency. This approach ensured that development remained focused on measurable outcomes rather than theoretical specifications.

Simulation and Validation Strategy

Given the high cost of physical testing, Wikispeed developed a sophisticated simulation strategy:

• Crash tests: $10,000 per test plus $14,500 in material costs
• Delivery costs: $2,500 to transport to testing facility
• Recycling fees: Additional disposal costs

To manage these constraints, Wikispeed replaced physical testing with computer simulations wherever possible. The accuracy of simulations was validated against physical tests and continuously refined. This approach enabled rapid iteration while managing costs and maintaining safety standards.

XM Implementation Steps

The XM methodology follows a structured approach:

1. Define test criteria – Establish measurable success metrics for each sprint goal
2. Implement simulation models – Create virtual testing environments
3. Validate simulation accuracy – Compare virtual results with physical tests
4. Iterate design solutions – Use simulation feedback for rapid design cycles
5. Integrate module testing – Ensure individual modules work as a system
6. Conduct final validation – Perform physical tests on integrated systems

Stubs and Mock-Ups for Rapid Prototyping

Early prototypes and mock-ups allow teams to test concepts before committing to expensive tooling. The modular design tools in Upverter are unique in that they allow designers with any level of experience to reduce development time and create customized, cutting-edge products. This approach enables rapid iteration and customer feedback integration.

Continuous Integration in Hardware

Unlike software, hardware integration involves physical constraints and timing considerations. Timing constraints—which cycle a signal arrives, when an input is read—and structural constraints—how often a multiplier accepts new inputs—are fundamental to hardware interfaces. The framework addresses these through:

• Regular integration testing – Ensuring components work together throughout development
• Interface validation – Verifying electrical and mechanical compatibility
• System-level testing – Validating integrated functionality

Test and Data-Driven Development

With Internet of Things (IoT) and sensor technology data about almost every behavior of hardware parts can be measured and should be used to continuously improve. Modern hardware development can leverage extensive instrumentation for continuous improvement:

• Performance monitoring – Real-time data collection from prototypes
• Behavioral analysis – Understanding how components interact in real conditions
• Predictive maintenance – Using data to anticipate failure modes

Team Structure and Organization

The framework emphasizes specific organizational patterns that enable agile hardware development:

Cross-Functional Teams: Teams include mechanical engineers, electrical engineers, software developers, and domain experts working together throughout the project lifecycle.

Pair Programming Adaptation: Wikispeed organized all work in teams of two or more, enabling on-the-job training and knowledge transfer without extensive documentation.

Customer Integration: Direct customer feedback loops ensure development remains focused on value delivery rather than internal metrics.

Economic Benefits

The modular architecture and agile practices dramatically reduced the cost of implementing changes. If pre designed modules are already readily available through libraries, the expense of design is lower. Nevertheless if modular libraries are being established as a part of the development effort the development expense and time are higher.

However, the long-term benefits justify initial investments:

• Reduced development cycles – Faster time-to-market generates revenue earlier
• Lower tooling costs – Modular approaches reduce expensive custom tooling
• Improved quality – Continuous testing and integration reduces defect costs
• Enhanced flexibility – Ability to respond to market changes provides competitive advantage

Automotive Sector Transformation

The automotive industry, in particular, continues to grapple with complex issues such as verification, validation, and the integration of software development with system engineering and safety. Several companies have successfully implemented agile approaches:

Volvo Trucks Case Study: A case study of volvo trucks demonstrates practical implementation challenges and solutions. The company faced traditional automotive development constraints but achieved significant improvements through systematic agile adoption.

Chinese B2B Supplier: When a B2B supplier in China created over 24 agile teams, it reduced its average time-to-market for new products by 20 percent within two years. This demonstrates scalability across cultural and organizational contexts.

Aerospace and Defense Applications

In the 1960s, Lockheed Martin built its “skunk works,” in which small development teams were removed from the normal working environment and freed from managerial constraints, making them autonomous and empowered. This historical precedent validates agile principles in highly regulated industries.

The agile development approach enables approval of the safety of already developed parts to be done by the manufacture and operator in parallel with other development. Modern aerospace applications demonstrate that safety-critical systems can benefit from agile approaches when properly implemented.

Electronics and Embedded Systems

The push for AI on edge is fostering greater use of modular hardware design approaches, both at the hardware and software level. The electronics industry has embraced modular approaches extensively:

• Arduino and Raspberry Pi ecosystems demonstrate mass-market modular hardware success
• System-on-Chip (SoC) development applies agile principles to semiconductor design
• IoT product development leverages modular architectures for rapid prototyping

Research and Academic Validation

AHA researchers pledge to use and develop open-source hardware and software, and it is the intention of all AHA researchers that any hardware and software will be released under an open source model. Academic institutions are developing systematic approaches to agile hardware development:

Stanford AHA Center: While advances in software tools and frameworks have enabled individuals to create interesting new products in reasonable time frames, hardware designs take large teams multiple years. This disparity in required effort decreases hardware innovation and interest.

Research Findings: Seven hundred thirty papers were first studied, and eventually, after a rigorous filtering process, 25 papers were thoroughly analyzed. Systematic literature reviews confirm growing academic interest and validation of agile hardware approaches.

Assessment and Preparation

Organizations considering agile hardware development should begin with comprehensive assessment:

Organizational Readiness: Evaluate cultural readiness, leadership commitment, and change management capabilities.

Technical Infrastructure: Assess current tools, processes, and technical capabilities for supporting agile workflows.

Regulatory Context: Understand compliance requirements and identify opportunities for agile-compatible approaches.

Conclusion

The transformation of hardware development through Scrum principles represents more than a process improvement—it constitutes a fundamental reimagining of how physical products can be created, validated, and brought to market. Two decades later, agile-for-hardware product development is making a strong comeback, driven by digital transformation, competitive pressures, and successful implementations like Wikispeed.

References

Primary Sources

Justice, J. (Founder, Team Wikispeed). Personal communications and documented practices from Team Wikispeed operations across 23 countries. Scrum Inc. Scrum in Hardware Guide, 2017.

Academic Research

Atzberger, A., & Paetzold, K. (2019). Evolution of the Hype around Agile Hardware Development. ResearchGate. Retrieved from https://www.researchgate.net/publication/334260520_Evolution_of_the_Hype_around_Agile_Hardware_Development

Cao, H., & Choudhary, P. (2023). Agile transformation in the automotive industry: A case study of volvo trucks. Master Thesis at Uppsala University.

Drutchas, G., & Eppinger, S. D. (2022). Current Challenges of Agile Hardware Development: What are Still the Pain Points Nowadays? ResearchGate.

Kasauli, R., et al. (2017). Requirements engineering challenges in large-scale agile system development. Proceedings of the IEEE 25th International Requirements Engineering Conference, 352-361.

Koren, Y., et al. (2023). Agile Safety Case and DevOps for the Automotive Industry. Research publication focusing on automotive safety in agile environments.

Industry Reports and Case Studies

Berteletti, E., Frank, S., Haazen, P., Rocha, A., & Wu, Y. (2023). It’s coming home: The return of agile hardware product development. McKinsey & Company Operations Practice. Retrieved from https://www.mckinsey.com/capabilities/operations/our-insights/its-coming-home-the-return-of-agile-hardware-product-development

Eppinger, S. D. (2023). Embracing Agile Product Development. Keynote presentation at PTC’s LiveWorx 2023 event. MIT Sloan School of Management.

Smart and Agile Manufacturing Framework, A Case Study for Automotive Industry. (2020). Energies, 13(21), 5766. DOI: https://doi.org/10.3390/en13215766

Technical Standards and Frameworks

Scrum in Hardware Guide: Purpose and Technical Practices. (2017). Scrum Inc. Available at: https://www.scruminc.com/scrum-in-hardware-guide/

ISO 22737:2021. Low-speed automated driving systems. International Organization for Standardization.

BSI PAS 1881:2020. Assuring the safety of automated vehicle trials and testing-specification. British Standards Institution.

BSI PAS 1883:2020. Operational design domain (ODD) taxonomy for an automated driving system (ADS)-Specification. British Standards Institution.

Design and Engineering Research

Modular Design Structure and High-Level Prototyping for Novel Embedded Processor Core. (2005). ResearchGate. DOI: 10.1145/1102253.1102275

TTL: a modular language for hardware/software systems design. ScienceDirect. Available at: https://www.sciencedirect.com/science/article/pii/S0022000003000023

Modular Hardware Design with Timeline Types. (2023). Proceedings of the ACM on Programming Languages. DOI: https://doi.org/10.1145/3591234

Software Engineering Foundations

Beck, K. (1999). Extreme Programming Explained: Embrace Change. Addison-Wesley Professional.

Schwaber, K., & Sutherland, J. (2020). The Scrum Guide: The Definitive Guide to Scrum. Available at: http://www.scrumguides.org

Takeuchi, H., & Nonaka, I. (1986). The new new product development game. Harvard Business Review, 64(1), 137-146.

Rigby, D., Sutherland, J., & Takeuchi, H. (2016). The secret history of agile innovation. Harvard Business Review, 94(5), 40-50.

Hardware Development and Manufacturing

Product Design Trends in 2020: Modular Hardware vs. Modular Software. (2020). Gumstix/Altium Resources. Available at: https://resources.altium.com/p/product-design-trends-in-2020-modular-hardware-vs-modular-software

Reduce Development Time for Electronics with Modular Design Tools. (2021). Gumstix/Altium Resources. Available at: https://resources.altium.com/p/reduce-development-time-for-electronics-with-modular-design-tools

Agile Hardware Development. (2019). Semiconductor Engineering. Available at: https://semiengineering.com/knowledge_centers/eda-design/methodologies-and-flows/agile/agile-hardware-development/

Implementation Guides and Best Practices

Agile for Hardware Development: Modified Agile Approaches for Hardware Teams. CoLab Software. Available at: https://www.colabsoftware.com/agile-for-hardware-development

Salimi, S. Agile Hardware Development. Agile Academy. Available at: https://www.agile-academy.com/en/agile-dictionary/agile-hardware-development/

Sammichelli, P. Scrum Hardware. Leanpub. Available at: https://leanpub.com/Scrum-for-Hardware

Open Source Hardware Research

AHA Agile Hardware Project. Stanford University. Available at: https://aha.stanford.edu/

Does SoC Hardware Development Become Agile by Saying So: A Literature Review and Mapping Study. (2023). ACM Transactions on Economics and Computation. DOI: https://doi.org/10.1145/3578554

Automotive Industry Case Studies

An Exploration of Agile Methods in the Automotive Industry: Benefits, Challenges and Opportunities. (2024). arXiv preprint arXiv:2409.12676. Available at: https://arxiv.org/abs/2409.12676

Agile Development in Automotive Software Development: Challenges and Opportunities. (2014). ResearchGate. DOI: 10.1007/978-3-319-33515-5_1

Agile Predevelopment of Production Technologies for Electric Energy Storage Systems – A Case Study in the Automotive Industry. ScienceDirect. DOI: https://doi.org/10.1016/j.procir.2016.11.195

Quality and Testing Frameworks

A Case Study on Implementing Agile Techniques and Practices: Rationale, Benefits, Barriers and Business Implications for Hardware Development. (2022). Applied Sciences, 12(17), 8457. DOI: https://doi.org/10.3390/app12178457

Implications of agile manufacturing in the automotive industry for order management in the factories-evidence from the practitioner’s perspective. ScienceDirect. DOI: https://doi.org/10.1016/j.procir.2018.03.051