A QUBO-Based Approach to Capacitated VRPTW Using Real Data

Shuhan Yang

Co-Presenters: Individual Presentation

College: College of Business and Public Management

Major: BS.ACCOUNTING

Faculty Research Mentor: Liu, Dan  

Abstract:

This study investigates the application of quantum computing to the Vehicle Routing Problem (VRP), a fundamental NP-hard combinatorial optimization problem with broad relevance in logistics and operations research. Building on the D-Wave quantum annealing framework, the research examines several hybrid quantum–classical solution strategies, including direct QUBO-based formulations, clustering-based decomposition methods, and solution partitioning approaches. Rather than focusing on raw computational performance, this work emphasizes how different modeling choices, input representations, and problem decomposition strategies influence scalability, solution interpretability, and the practical role of quantum components.By analyzing multiple solver architectures and input formats, the study highlights that quantum annealing is primarily used as a subroutine for solving reduced subproblems after substantial classical preprocessing. The results suggest that the effectiveness of quantum-assisted optimization in this context depends less on the quantum hardware itself and more on how the original VRP is restructured into tractable components. This research contributes to a clearer understanding of the boundaries and assumptions underlying current hybrid quantum–classical optimization frameworks and provides insight into how emerging quantum methods can be meaningfully integrated into complex real-world optimization systems.