We are pleased to announce that adago group's research and development work has been awarded the BSFZ Seal by the Certification Body for Research Grants (Bescheinigungsstelle Forschungszulage), acknowledging our contribution to advancing the field of artificial intelligence in supply chain optimisation.
At the core of our research lies the development of highly efficient simulation environments for supply chain systems, specifically designed to leverage Deep Reinforcement Learning algorithms. Our novel "self-learning digital twin" framework implements a custom-adapted Proximal Policy Optimization (PPO) algorithm, enabling rapid training through millions of simulation steps.
"The key innovation in our approach is the seamless integration of supply chain complexity into a gymnasium environment that maintains high computational efficiency," explains Prof. Leif Döring, Chief Scientist at adago. "By optimising our simulation architecture, we've achieved training speeds that make Deep RL practical for real-world supply chain applications."
Oren Neumann, AI Researcher at adago, provides technical insight: "Our implementation extends the established PPO framework from stable-baselines3, with specific adaptations for supply chain state spaces. The environment's object-oriented architecture allows for efficient parallel simulation, crucial for the extensive training iterations required in Deep RL."
This federal recognition validates our technical approach to solving complex optimisation challenges. Our research continues to advance, with several technical breakthroughs in development that will further enhance our Deep RL capabilities. These innovations focus on improving both the efficiency of training processes and the robustness of learned strategies in dynamic supply chain environments.
As a technology company specialising in AI solutions, we remain committed to pushing the boundaries of what's possible in supply chain optimisation through advanced machine learning techniques. We look forward to sharing more detailed technical insights about our ongoing research in the coming months.