Enhanced Reinforcement Learning-based Process Synthesis via Quantum Computing

2026-05-20

AI Summary

This work addresses the scalability bottleneck in traditional quantum reinforcement learning for process synthesis, where the required number of qubits grows prohibitively with problem size. By formulating the task as a Markov decision process, the authors introduce a novel state encoding strategy that effectively decouples problem scale from qubit count. The resulting quantum-enhanced reinforcement learning framework demonstrates markedly improved scalability: while all methods converge to optimal solutions on small-scale instances, the proposed approach matches classical baselines in single-run performance on medium-scale problems and exhibits superior parameter efficiency.

Abstract

In this work, we present quantum reinforcement learning (RL) as a solution strategy for process synthesis problems. Building on our prior work, we develop a generalized framework that formally poses process synthesis as a Markov decision process and introduces quantum-enhanced RL algorithms to solve it with improved scalability. Earlier implementations of quantum-based RL for process synthesis were limited by qubit requirements, which scaled poorly with problem complexity. This work overcomes this challenge by introducing state encoding algorithms to decouple qubit requirements from problem size. A classical RL-based solution strategy is used as a baseline to benchmark the quantum algorithms under identical training conditions. All algorithms are evaluated across a flowsheet synthesis problem of increasing unit counts to analyze their performance and scalability. Results show that all approaches are capable of identifying the optimal flowsheet designs in small design spaces. For moderate-scale unit counts, quantum approaches demonstrate competitive performance on a per-episode basis and improved efficiency on a per-parameter basis versus the classical RL benchmark. This work provides a foundation for future quantum computing applications within process systems engineering, establishes a controlled benchmark for comparing classical and quantum algorithms, and shows that the proposed quantum variants remain competitive for the process synthesis problem examined in this work.

Problem

• process synthesis
• quantum computing
• reinforcement learning
• scalability
• Markov decision process

Innovation

• quantum reinforcement learning
• process synthesis
• state encoding
• scalability
• Markov decision process