Customer Story
How a global automaker mirrored its factories asset-by-asset — cars, robots, operators, and spare parts — on Akka, closing the gaps between five to six legacy production systems.
Renault Group, with reflek.io, built a SaaS Digital Twin Execution Platform on Akka that mirrors its distributed factories in real time. Each industrial object — car, robot, operator, spare part, building — became a live, event-driven twin, collapsing five to six disconnected production systems into a single execution layer and ending the digital discontinuity that made every change risky.
Akka ROI Scorecard
Five to six systems that ran the design-to-delivery cycle — each with its own data model — were unified into one real-time twin, so a change in one system no longer forces changes across all of them.
lower parts inventory. Accurate real-time views of component consumption let Renault Group hold less buffer stock, translating into millions in annual savings.
plants mirrored across the continuum from on-premises to cloud, with millions of physical assets reflected as live twins that move from location to location.
Business outcome: a real-time digital replica of a globally distributed factory network and extended supply chain, with granular cost and carbon data per operation.
Renault Group operates one of the largest factory networks of any automaker. Like most large manufacturers, it managed the vehicle production cycle from design to delivery across five or six major applications, each with its own data model. Legacy systems exchanged data in opaque, inflexible ways, and integration with newer digital tools was poor. Several systems were no longer covered by vendor support, and the employees who maintained them were nearing retirement.
This is digital discontinuity: changing one system forces changes across all of them. It left Renault Group unable to respond in real time to the demands modern manufacturing places on the production line — adjusting output to shifting demand, avoiding supply-chain ruptures, and controlling energy use and carbon footprint. The company wanted to transform operations without the risk or cost of ripping out and replacing its existing systems.
Renault Group worked with reflek.io to build a SaaS platform on Akka, sitting between cloud and edge. The platform provides digital execution twins — real-time, accurate mirrors of physical objects and processes that interact with their counterparts on the factory floor. Every asset is modeled in natural language, producing a full, live picture of the entire factory and its operations: what was supposed to happen, what happens next, and the status, location, energy, and CO2 of each machine.
The twins had to be available in any location and moveable from place to place, running across the full continuum from on-premises to cloud. Akka handles the distributed clustering, durable state, and resilience that make this possible, so Renault Group recreates a layer of digital continuity on top of its legacy systems and decommissions critical shopfloor systems step by step.
Scale. Renault Group gained a real-time digital replica of its distributed factories and extended supply chain. By populating that replica with production data, it closed the information and execution gaps between its legacy applications, mirroring assets across 35+ plants from edge to cloud.
Cost to operate. A next-generation development model with no operations overhead and generative AI made development costs marginal. Real-time visibility into part consumption enabled reduced inventory, which Renault Group expects to translate into millions in annual savings.
Speed to production. Modeling processes and assets in natural language drastically simplified the application landscape, letting Renault Group identify bottlenecks, recombine processes, and optimize operations without disrupting the running line — a low-risk, low-cost migration off legacy systems.
Business impact. All real-world activity is reliably recorded in the twins, so Renault Group can enrich each operation with its cost and carbon footprint, roll that up across a factory, and meet manufacturing and sustainability requirements while planning better and adapting faster to supply-chain disruption.
The digital twin Renault Group built is the substrate agentic AI runs on. On the Akka Agentic AI Platform, AI agents keep the twin synchronized across the vehicle production chain, detect anomalies against the live mirror of every asset, and orchestrate real-time production and supply-chain adjustments — reacting to demand shifts, heading off supply ruptures, and controlling energy and carbon at the operation level. The same distributed, durable execution that mirrors the factory today is what governs those agents at plant scale.
The platform that gave Renault Group a real-time twin of its factories is the same platform that builds and runs agentic AI. Enterprises build on the Akka Agentic AI Platform directly, or have a system delivered and operated through Akka Specify. Both provide the production guarantees a global manufacturing network requires.
