Quantum Bayesian Optimization for Quality Improvement in Fuselage Assembly

TL;DR

This work tackles the challenge of sample-inefficient optimization in fuselage assembly by introducing Quantum Bayesian Optimization (QBO), which uses a quantum Monte Carlo-based mean estimator and a UCB acquisition to dramatically reduce the number of costly simulations or measurements. The framework couples a quantum oracle with mean estimation and a weighted Gaussian Process surrogate to guide actuator-placement and force-selection for minimizing dimensional gaps between joined fuselage sections. Across simulated experiments with noise, QBO demonstrates superior sample efficiency and lower MAE compared to classical BO, highlighting practical potential for faster, more reliable quality improvement in aerospace manufacturing. The study also discusses limitations and future directions, including incorporating physical constraints and moving towards real quantum hardware implementations.

Abstract

Recent efforts in smart manufacturing have enhanced aerospace fuselage assembly processes, particularly by innovating shape adjustment techniques to minimize dimensional gaps between assembled sections. Existing approaches have shown promising results but face the issue of low sample efficiency from the manufacturing systems. It arises from the limitation of the classical Monte Carlo method when uncovering the mean response from a distribution. In contrast, recent work has shown that quantum algorithms can achieve the same level of estimation accuracy with significantly fewer samples than the classical Monte Carlo method from distributions. Therefore, we can adopt the estimation of the quantum algorithm to obtain the estimation from real physical systems (distributions). Motivated by this advantage, we propose a Quantum Bayesian Optimization (QBO) framework for precise shape control during assembly to improve the sample efficiency in manufacturing practice. Specifically, this approach utilizes a quantum oracle, based on finite element analysis (FEA)-based models or surrogate models, to acquire a more accurate estimation of the environment response with fewer queries for a certain input. QBO employs an Upper Confidence Bound (UCB) as the acquisition function to strategically select input values that are most likely to maximize the objective function. It has been theoretically proven to require much fewer samples while maintaining comparable optimization results. In the case study, force-controlled actuators are applied to one fuselage section to adjust its shape and reduce the gap to the adjoining section. Experimental results demonstrate that QBO achieves significantly lower dimensional error and uncertainty compared to classical methods, particularly using the same queries from the simulation.

Figures (7)

• Figure 1: Schemes of (a) force-controlled actuators system and (b) fixtures in fuselage assembly WEN2018272.
• Figure 2: The pipeline of our proposed method. Data from FEA simulations are first collected and used to train a Gaussian Process surrogate model. Based on the GP posterior, the acquisition function selects the next force combinations. The quantum oracle encodes deviations into a distribution, and mean estimation evaluates the candidate force combinations through repeated queries. The new observation and candidate force combinations update the GP posterior.
• Figure 3: Scheme of 2 actuators used in the fuselage shape control.
• Figure 4: (a) Comparison of cumulative regrets between classic BO and QBO in discrete cases. (b) Compare the best shape control performance between classic BO and QBO with 2 actuators. The black dashed line marks the incumbent after 2,000 iterations.
• Figure 5: (a) Comparison of Cumulative regrets between classic BO and QBO using 8 actuators in fuselage assembly. (b) Compare the best shape control performance between classic BO and QBO with 8 actuators. The black dashed line marks the incumbent after 2,500 iterations.