BQPhy

BQPhy^® is a quantum optimization platform designed to accelerate complex engineering and scientific workloads in high-performance computing environments. By combining advanced computational techniques with quantum-inspired and hybrid quantum algorithms, BQPhy enables enterprises to solve mission-critical simulations faster, more accurately, and more efficiently on today's infrastructure, while maintaining seamless readiness for future quantum hardware.

Aimed at solving high-dimensional, non-convex, and compute-intensive optimization problems where conventional tools struggle, the BQPhy platform leverages quantum-inspired principles to maximize the computational potential of today's CPUs and GPUs. Powered by the QuantumNOW™ solver, the platform explores complex solution spaces to identify up to 10x performance gains today while providing a clear evolution path to much higher acceleration through the QuantumMAX solver for quantum-native acceleration.

Designed for usability as much as capability, BQPhy offers an intuitive interface that supports users with varying levels of expertise. Industries such as aerospace, defense, semiconductors, and energy rely on BQPhy to power next-generation digital twins, accelerate design cycles, uncover deeper design insights, and reduce costly experimentation. Its hybrid-ready design positions teams for the future of engineering computation.

BQPhy serves engineering teams and researchers working with complex simulations and computationally intensive optimization challenges. It is utilized by simulation and CAE engineers performing CFD, FEA, and thermal analyses, as well as optimization specialists addressing non-convex and multidisciplinary optimization problems. BQPhy's quantum-ready solver, including its Quantum-Inspired Optimization (QIO) capabilities, enables users to explore larger design spaces, reduce iteration cycles, and achieve faster, more robust convergence within familiar tools through seamless native integration with environments such as MATLAB^® and Simulink^®, without disrupting existing workflows.

BQPhy addresses a wide range of engineering applications that traditionally push the limits of classical algorithms. These include design optimization, cargo and passenger loading, trajectory and mission planning and scheduling, controller and parameter tuning of dynamic systems and machine learning models, and multidisciplinary optimization (MDO) tasks such as aerothermoelasticity optimization, aero-vibro-acoustic optimization, and electro-thermo-mechanical optimization.

BQPhy also supports emerging data-driven engineering workflows, blending machine learning models with quantum-inspired optimization to accelerate surrogate modeling, predictive analytics, and digital twin development. Across these applications, BQPhy provides immediate performance gains on existing CPU and GPU infrastructure, achieving up to order-of-magnitude speedups today while enabling a smooth transition toward future quantum-native solvers.

The quantum-ready QuantumNOW solver is available as a MATLAB toolbox that can be downloaded and installed by the user. While early versions included a hard-coded expiration date, current and future releases use a software license that can be installed through a simple licensing procedure provided by BQP.

Once installed, the QuantumNOW solver is invoked through a MATLAB command that links to the simulation model and passes solver parameters, including problem type, variable bounds, constraints, solver-specific configuration values, and stopping criteria. Users can also specify execution modes, including serial ("cpu") and multi-threaded ("openmp"), with GPU support planned for future releases.

A vectorized matrix of candidate design values, organized as population size by design variables, is passed to the simulation model. The model evaluates the performance of all candidate designs and returns a corresponding vector of fitness values to the optimization solver. This process is repeated iteratively until the specified stopping criteria are met.