Particulars of 'software-defined' transformations, courtesy of digital twins
What you’ll learn:
- A shift to software-defined development has widespread consequences for a product program.
- This shift introduces new complexities as a change in software functionality can trigger cascading effects across multiple interdependent domains.
- As product complexity grows, digital twins provide developers with a unified platform.
Multiple industries are undergoing a software-defined transformation, creating new opportunities for innovation. Modern cars offer a great test case. Software updates routinely pushed to vehicles remotely can improve performance, introduce new features, and even fix problems without a visit to a service center.
See also: How private 5G and IFPP are powering the next wave of industrial IoT
A shift to software-defined development has widespread consequences for a product program, affecting every engineering domain and product team. In my role at Siemens there’s been ample opportunity for collaboration with thoughtful and experienced professionals representing the spectrum of engineering capabilities needed to create tomorrow’s smart products.
The ongoing interaction with these experts provides new insights regarding technology, business, and more, especially in the face of today’s changing digital landscape.
I recently sat down with Mike Ellow, CEO of Siemens EDA, a segment of my company, Siemens Digital Industries Software, that provides electronic design automation software, hardware, and services. We discussed the changing landscape and how software has begun to define product functionality in multiple industries like automotive, aerospace, and consumer electronics.
Here’s key takeaways from that conversation, focusing on how software-defined development is changing how companies approach product and system development and how the comprehensive digital twin supports these new development approaches, particularly through digital thread connections between electronic design automation, software, and the rest of the system.
Software definition drives growing complexity
The shift toward software-defined products introduces new complexities as a change in software functionality can trigger cascading effects across multiple interdependent domains.
For example, changes to the behavior of the software onboard an upcoming electric vehicle can alter the power draw of the compute platform and impact overall drive range.
Depending on the severity of the range reduction, such a software change may lead to a re-evaluation of the vehicle’s battery size and even the physical packaging of the vehicle.
See also: This tech labor shortage that no one is talking about could complicate reshoring
The heightened attention paid to software development and differentiation also has ripple effects in the development program organization, specifically affecting the semiconductor devices that power software features and functionality.
While mechanical and electrical systems remain important to these systems, the locus of attention and investment is shifting to software and semiconductors.
Historically, companies used off-the-shelf chips that fit requirements and built software around the capabilities and limitations of the general-purpose hardware platform.
But, as software systems have grown in sophistication and importance, more companies are in search of custom silicon solutions that can be adapted to specific software workloads and development methods that support hardware-software co-design.
In the past, custom silicon development was impractical for most companies due to the length, cost, and risk associated with the creation of application specific silicon. Now, semiconductor manufacturers are driving down development timelines by pursuing key innovations like heterogenous integration.
Q&A: Could a software vendor be on the hook if your company's systems get hacked?
This enables them to combine affordable off-the-shelf technologies (such as memory) with smaller, more affordable, and less risky custom silicon to achieve the optimal performance and cost goals.
Digital twins facilitate holistic software-driven development
Today, makers of complex systems must contemplate the development and integration of several sub-systems, including:
- Software applications and base level functions
- Semiconductor devices and modules
- Electrical and electronic systems, including data networks
- Mechanical components and structures
All these require different engineering expertise and solutions, but must be seamlessly merged into a cohesive, desirable, and user-friendly product. Companies must accomplish this task despite tightening budgets and timelines just to maintain position in increasingly competitive markets.
These pressures demand connected, agile, and holistic development methodologies that ease the flow of information throughout interdependent teams; these pressures demand an investment in digitalization through the construction of comprehensive digital twins.
See also: ‘Operational’ twins should rise in their use
The comprehensive digital twin of any complex product or system connects all product models and data through digital threads that provide bi-directional flows of information among development teams and project managers. The result is constant feedback between development activities and design requirements.
As product complexity grows—such as in electric vehicles or aerospace systems—digital twins provide developers with a unified platform to visualize dependencies between software, electronics, and mechanical components.
This real-time synchronization mitigates costly delays, ensuring that every aspect of the product remains aligned with evolving requirements and regulatory standards.
Beyond individual teams, digitalization enhances cross-functional collaboration by providing stakeholders with accessible, transparent data. Supply chain managers, software engineers, and mechanical designers can interact within a shared digital ecosystem, streamlining validation processes and ensuring that modifications do not introduce unforeseen complications.
See also: Unleashing the potential of digital twin technology
Preparation for manufacturing can also be enhanced through early production planning, modeling, and validation using the Digital Twin of the product. This holistic approach reduces development risks and improves efficiency, enabling companies to bring high-quality, cutting-edge products to market faster.
Semiconductor virtualization enhances software development
For the software-defined product or system, the virtualization of silicon development and verification is crucial. Traditionally, software teams waited for physical hardware prototypes before testing software, slowing development cycles and increasing project risks.
Today, digitalization has paved the way for the virtualization of silicon architectures, decoupling software development from hardware development or selection. This enables active software development well before any physical semiconductor devices have been fabricated.
See also: Sustainability’s future in manufacturing: What to expect
Such virtualization enables a design paradigm in which software and hardware interdependencies can be addressed without total integration. Software teams can constantly update applications to reflect expected silicon performance as the fidelity of silicon models increases throughout development, reducing the likelihood of integration errors.
It also helps companies take full advantage of heterogeneously integrated silicon by facilitating the optimization of processor cores, memory configurations, and interfaces to match evolving computing demands while reducing costs.
Furthermore, a connected digital ecosystem ensures that evolving software and silicon specifications are immediately reflected in the comprehensive digital twin of the system.
As software and semiconductor configurations mature, electrical and mechanical teams can complete design work with the most up-to-date information, accounting for any impacts that a change in software may create in their respective systems.
Returning to the EV example, the digital twin ensures that the holistic impacts of a software change are understood throughout the product ecosystem.
For software developers, this offers the opportunity to optimize the change to achieve the desired behavior while minimizing external impact. Once changes to the software are finalized, new requirements can be pushed to affected teams to begin updating their designs as needed.
Embrace complexity to manage software-driven development
The growing influence of software-defined products continues to transform industries, unlocking new opportunities for innovation and efficiency. At the same time, the increasing reliance on software has introduced new complexities, necessitating a more integrated approach to development.
The convergence of software and semiconductor advancements has accelerated the demand for custom silicon solutions, optimizing computational power for specialized workloads. Digitalization, particularly through digital twins, has emerged as a critical tool in managing these interdependencies, allowing for real-time collaboration and validation across cross-functional teams.
See also: New report cites ‘agility emergency’ for manufacturers in tariff times
Ultimately, successful navigation of the software-defined age will rely on a commitment to digital infrastructure investment in the pursuit of a comprehensive digital twin.
By embracing digital twin technology, organizations can streamline innovation, mitigate risks, and enhance product differentiation. As industries continue to evolve, companies that deploy these technologies most effectively will shape the next generation of intelligent, interconnected products.
