Digital twin simulation is driving next-gen manufacturing
Digital twins are more than just digital models; they are the foundation of a dynamic learning system. During the Honeywell Users Group 2025 conference in San Antonio, Manzoor Tiwana, lead product manager for digital twins at Ansys, highlighted how high-fidelity simulations are pushing the boundaries of design and manufacturing. His presentation, “Enhancing Smart Manufacturing with Simulation: Honeywell’s Collaboration with Ansys to Optimize Lithium-Ion Battery Production” explored how cutting-edge digital twin technology is changing electric vehicle (EV) battery performance, predictive maintenance and real-time process control.
Understanding digital twins
Tiwana explained that digital twins are sophisticated digital representations of physical assets, systems or processes. He said that while many in industry misconstrue digital twins as mere digital models, true digital twins are much more complex. “For a lot of people,” he said, “a digital twin is just any digital model they create,” but this oversimplification misses the essence of what a digital twin encapsulates.
A digital twin integrates real-time data from sensors placed on physical assets with predictive models that simulate their behavior under a range of conditions. This allows companies to gain insights into performance and operational efficiency, which in turn help pave the way for predictive maintenance, optimized production and extended asset lifetimes.
Tiwana noted that for more than 50 years, Ansys has worked in the simulation market, primarily known for computational fluid dynamics (CFD) and finite element method (FEM) simulations. These high-fidelity simulations mimic real-world scenarios with remarkable accuracy, he said, providing designers and engineers with insights that can improve product outcomes.
“The goal is to bring all those physics together,” he said, adding that these simulations are critical during the design phase. Before actual products, such as lithium-ion battery packs, are manufactured and deployed, it is important to understand how they will perform in the real world.
A collaborative approach to EV battery management
One of the most significant projects Tiwana talked about involved a collaboration between Honeywell and Ansys, focused on EV battery management. The initiative, which also involved Hyundai, Bosch and AWS, aimed to optimize battery performance in electric vehicles, addressing a common challenge faced by manufacturers: limited range and efficiency.
He explained how the collaboration leverages both Honeywell’s expertise in industrial automation and control systems and Ansys’s advanced simulation capabilities. The objective of the project was to create an adaptive digital twin that responds to various driving conditions in real time. The digital twin not only simulates the battery’s behavior but also integrates sensor data from the vehicle to make dynamic adjustments.
In practice, Tiwana says this means that as an EV approaches its range limits, the digital twin can automatically adjust power consumption in various vehicle systems, similar to the “power save” on smartphones. This collaborative effort aims to provide drivers with better insights into their vehicle’s performance, helping them manage their power usage effectively and improve overall driving experiences.
He noted the broader implications of such technologies, too: “You can optimize your production…control your operations and gain insights into the future performance of your assets. All these things together come together in this platform.” By combining efforts with Honeywell, the project seeks to not only improve battery efficiency but also create potential revenue streams through advanced service contracts tied to predictive insights from digital twins.
Hybrid digital twin bridges gap between data and physics
A unique digital twin aspect that Tiwana covered is the integration of machine-learning data with traditional physics-based models. This hybrid method allows for more nuanced insights into manufacturing processes, addressing some of the shortcomings associated with relying solely on either data-driven or physics-based approaches.
“We combine that data and physics, and you get good insights from your model,” he said. For example, sensor data such as heater power and the temperature from a thermometer on the electrode coating line is combined on an IoT platform (Azure, PTC, SAP, etc.) with system model data such as a transient temperature prediction to make the hybrid model. This methodology enhances the fidelity of the simulations and offers a more comprehensive understanding of the factors influencing production outcomes, from raw material properties to environmental conditions.
By combining physics, telemetry and artificial intelligence (AI), these hybrids enable adaptive inline optimization.
Looking ahead, Tiwana explained that while predictive maintenance is vital, real-time insights are necessary for effective process control. And the session seemed to resonate with attendees, who sparked discussions about the diverse applications of digital twins across sectors, including battery manufacturing, integrated process systems and supply chain systems.
Tiwana also directed the audience to the full case study of how Honeywell and Ansys are playing a transformative role in battery manufacturing.
