Fraunhofer-GesellschaftThe individual institutes of the Fraunhofer-Gesellschaft are experts with deep domain knowledge that can be leveraged to translate industrial problems to quantum computers, going far beyond purely academic toy problems. In addition, their application expertise enables an assessment of which simplifications are feasible and where exact solutions are required. This is particularly valuable in optimization problems, where classical optimization often relies on heuristics that may not solve a problem in its full generality but are highly effective in practice. Examples from maritime logistics include network planning in liner shipping, maritime inventory routing, berth allocation, and quay crane optimization, all of which are being addressed within the QSH project at Fraunhofer CML.
Research is being conducted on simulation and optimization problems that scale more efficiently on quantum computers and may therefore enable future breakthroughs in fields such as chemistry, pharmaceuticals, and logistics. Efficient encoding techniques for fermionic systems, as well as the use of Clifford gates for pre-optimization, are being tested in quantum chemistry applications.
In particular, the topics of mathematical optimization in logistics using quantum algorithmic approaches, as well as the development and analysis of quantum-algorithmic solutions for computationally intensive imaging applications, are being investigated at Fraunhofer IIS within the projects QuaST (Quantum-enabling Services and Tools for Industrial Applications) and QACI (Quantum Algorithms for Application, Cloud & Industry).
At Fraunhofer IAIS, quantum machine learning and quantum optimization methods have been successfully applied in aerospace engineering, particularly for bundle adjustment and cloud removal from satellite imagery. In the area of reconfigurable circuit design (FPGAs), quantum optimization has been developed and successfully deployed for the placement problem. Furthermore, Fraunhofer IAIS methods define the state of the art in ESG-compliant portfolio optimization and the energy redispatch problem, which concerns the reallocation of configurable resources for energy generation.
Within the framework of the Baden-Württemberg Quantum Competence Center, the Fraunhofer Institutes IPA, IAO, and IAF have investigated a wide range of use cases in quantum optimization and quantum machine learning through the SEQUOIA and SEQUOIA End-to-End projects. The application areas include logistics, manufacturing, and automotive engineering and were explored in collaboration with numerous industrial partners.
Building on its broad portfolio of mathematical methods, Fraunhofer ITWM investigates use cases across diverse fields, including finance, energy systems, chemistry, materials simulation, production processes, anomaly detection, and image processing. To this end, the institute applies methods such as modeling, quantum machine learning, quantum optimization, and benchmarking to evaluate the potential benefits and progress of quantum computing technologies.
In financial mathematics, Quantum Amplitude Estimation (QAE) and Quantum Monte Carlo simulations are used for risk-return classification in retirement planning, pricing financial derivatives such as options, calculating financial risk measures, and determining the solvency capital requirements of banks and insurance companies. Quantum Architecture Search is also employed in this context. In addition, quantum optimization algorithms are applied to portfolio optimization problems.
A power plant unit commitment problem from the energy sector, including the integration of renewable energy supply, has been modeled and solved using hybrid approaches based on the D-Wave quantum annealer. In quantum chemistry, hybrid QC/HPC methods are provided for the electronic structure problem. Numerical simulation is an important tool for characterizing composite materials, and Fraunhofer ITWM focuses on the potential of quantum computers to solve complex material models rapidly and efficiently. This includes investigating both long-term approaches with significant future potential, such as the Quantum Fourier Transform (QFT), and heuristic methods that are applicable in the short to medium term, such as variational hybrid algorithms.
Drawing on its extensive expertise, Fraunhofer ITWM has recently expanded its activities to include quantum-inspired algorithms, particularly tensor-train methods. In the SQEDULE project, a use case from BASF laboratories involving the scheduling of laboratory robots was investigated. Multiple classical, quantum-inspired, and quantum-hybrid solvers were applied and benchmarked on problems up to industrial scale.
For various anomaly detection use cases, including production processes in the EniQmA project and credit card fraud detection, quantum machine learning methods are being applied. The goal of the production process anomaly detection use case is to identify anomalies in manufacturing processes automatically and at an early stage. This is intended to ensure high product quality while minimizing unplanned downtime, wear, and scrap.
The quality assurance approach presented here leverages quantum computing solutions to optimize both image processing algorithms and anomaly detection methods. Quantum machine learning offers the potential to significantly outperform conventional approaches through its high computational power and efficiency.