Quantum Inspired Optimization for Industrial Scale Problems

TL;DR

This paper evaluates quantum-inspired tensor network models for large-scale industrial optimization, demonstrating their potential to improve solution quality when combined with traditional methods.

Contribution

It introduces the TN-GEO quantum-inspired optimization method and assesses its effectiveness on a realistic BMW assembly line scheduling problem.

Findings

Quantum-inspired models can find lower-cost solutions in some contexts.

Combining quantum-inspired and black-box methods enhances optimization performance.

The approach shows promise for industrial-scale combinatorial problems.

Abstract

Model-based optimization, in concert with conventional black-box methods, can quickly solve large-scale combinatorial problems. Recently, quantum-inspired modeling schemes based on tensor networks have been developed which have the potential to better identify and represent correlations in datasets. Here, we use a quantum-inspired model-based optimization method TN-GEO to assess the efficacy of these quantum-inspired methods when applied to realistic problems. In this case, the problem of interest is the optimization of a realistic assembly line based on BMW's currently utilized manufacturing schedule. Through a comparison of optimization techniques, we found that quantum-inspired model-based optimization, when combined with conventional black-box methods, can find lower-cost solutions in certain contexts.