Quantum Annealing Hyperparameter Analysis for Optimal Sensor Placement in Production Environments

2025-07-22

AI Summary

This work addresses the sensor placement optimization problem for autonomous vehicle transportation in automotive manufacturing—specifically, minimizing the number of sensors while ensuring full coverage—where conventional heuristic methods suffer from local optima and poor scalability. Method: We propose a Quantum Approximate Optimization Algorithm (QAOA)-inspired QUBO modeling and solving framework tailored for quantum annealing, implemented on D-Wave hardware. Our approach introduces a novel hyperparameter co-optimization strategy—including penalty coefficients, annealing time, and a hybrid one-hot/binary encoding scheme—combined with problem decomposition to overcome hardware size limitations. Results: On real-world production-line instances, our method significantly outperforms default parameter settings: coverage increases by 12.7% while sensor count decreases by 18.3%. This constitutes the first systematic empirical validation of quantum annealing’s feasibility and practical potential for large-scale industrial combinatorial optimization deployment.

Technology Category

• Search and Optimization: Combinatorial Optimization
• Constraint Satisfaction and Optimization: Constraint Optimization
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

To increase efficiency in automotive manufacturing, newly produced vehicles can move autonomously from the production line to the distribution area. This requires an optimal placement of sensors to ensure full coverage while minimizing the number of sensors used. The underlying optimization problem poses a computational challenge due to its large-scale nature. Currently, classical solvers rely on heuristics, often yielding non-optimal solutions for large instances, resulting in suboptimal sensor distributions and increased operational costs. We explore quantum computing methods that may outperform classical heuristics in the future. We implemented quantum annealing with D-Wave, transforming the problem into a quadratic unconstrained binary optimization formulation with one-hot and binary encoding. Hyperparameters like the penalty terms and the annealing time are optimized and the results are compared with default parameter settings. Our results demonstrate that quantum annealing is capable of solving instances derived from real-world scenarios. Through the use of decomposition techniques, we are able to scale the problem size further, bringing it closer to practical, industrial applicability. Through this work, we provide key insights into the importance of quantum annealing parametrization, demonstrating how quantum computing could contribute to cost-efficient, large-scale optimization problems once the hardware matures.

Problem

• Optimal sensor placement for autonomous vehicle movement coverage
• Overcoming computational challenges in large-scale optimization problems
• Quantum annealing for cost-efficient industrial optimization solutions

Innovation

• Quantum annealing for optimal sensor placement
• D-Wave quadratic binary optimization formulation
• Decomposition techniques for industrial scalability