Hardware-Efficient Quantum Optimization for Transportation Networks via Compressed Adiabatic Evolution

Talha Azfar, Ruimin Ke, Sean He, Cara Wang, José Holguín-Veras · 2026

Read Hardware-Efficient Quantum Optimization for Transportation Networks via Compressed Adiabatic Evolution by Talha Azfar, Ruimin Ke, Sean He, Cara Wang, José Holguín-Veras (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-26175.txt.

Introduction

Azfar and four colleagues develop a hardware-grounded hybrid quantum optimization framework for transport problems such as vehicle routing, sales tours, and facility location. Approximate quantum compilation compresses early segments of digitized adiabatic steps into shallow circuits. With variational layers, it shows how start, depth, and expressivity interact on near-term hardware. Run on an IBM gate-based machine, moderate compression cuts two-qubit gate depth while keeping feasible-solution discovery steady or better for routing. Gains depend on fit between the compressed prefix and the variational ansatz.

Catalog record

CREATORTalha Azfar, Ruimin Ke, Sean He, Cara Wang, José Holguín-Veras
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE92101
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.26175

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Hardware-Efficient Quantum Optimization for Transportation Networks via Compressed Adiabatic Evolution Talha Azfar, Ruimin Ke, Sean He, Cara Wang, José Holguín-Veras Abstract Transportation systems such as urban logistics, vehicle routing, and infrastructure planning require solving large-scale combinatorial optimization problems under complex constraints. Problems such as the vehicle routing problem (VRP), traveling salesman problem (TSP), and facility location problem (FLP) involve large discrete search spaces and the need to generate multiple feasible solutions in real time. In this work, we develop a hardware-grounded hybrid quantum optimization framework that uses Approximate Quantum Compilation (AQC) to compress early segments of digitized adiabatic evolution into shallow circuits. The compressed prefix is combined with variational layers, enabling a systematic study of how initialization, circuit depth, and expressivity interact on near-term quantum hardware. All experiments are performed on an IBM gate-based quantum computer, and circuits are evaluated as stochastic generators of candidate transportation plans. Results show that moderate prefix compression reduces two-qubit gate depth while maintaining or improving feasible solution discovery, particularly for routing problems. These benefits depend on compatibility between the compressed prefix and the variational ansatz: while standard QAOA effectively leverages AQC initialization, linear-chain QAOA shows limited improvement. Overall, this work demonstrates that hybrid AQC-QAOA methods provide a practical pathway for hardware-efficient quantum optimization, positioning quantum algorithms as candidate generators within transportation decision-making workflows. [type=editor, auid=000,bioid=1, orcid=0000-0002-1293-5036] [type=editor, auid=000,bioid=1, orcid=0000-0001-9139-6765] [type=editor, auid=000,bioid=1, orcid=0000-0002-1633-4748] [type=editor, auid=000,bioid=1, orcid=0000-0001-5819-0212] [type=editor, auid=000,bioid=1, orcid=0000-0001-8118-9383] keywords Quantum Computing ,Facility Location ,Vehicle Routing ,Traveling Salesman Problem ,Optimization ,QAOA † † credit: Conceptualization of this study, Methodology, Software † † credit: Data curation, Writing - Original draft preparation † † corresponding: Corresponding author 1 Introduction Modern transportation and logistics systems depend critically on the ability to solve large-scale combinatorial optimization problems that arise across multiple levels of planning and operations. Many of the key decisions required to manage these systems involve selecting discrete actions over complex network structures, where the number of feasible configurations grows combinatorially with system size. Examples of such decisions include the strategic placement of infrastructure such as warehouses or distribution centers, the tactical design of vehicle routes serving geographically distributed demand, and the real-time dispatch of vehicles and resources in response to dynamic conditions. These decisions must typically be made under resource constraints, network connectivity limitations, and operational requirements that significantly increase the complexity of the resulting optimization models. A number of well-established combinatorial optimization formulations capture these transportation decision problems. The Vehicle Routing Problem (VRP) models the task of determining efficient routes for a fleet of vehicles that must serve a set of customers while minimizing travel cost or time ( 11 ) . The Traveling Salesman Problem (TSP) represents a simplified single-vehicle routing problem in which a tour must visit each location exactly once before returning to the starting point ( 25 ) . The Facility Location Problem (FLP) addresses strategic infrastructure planning by determining which candidate facilities should be opened and how customers should be assigned to them in order to minimize combined facility and service costs ( 1 ) . Although these models provide useful abstractions of real-world transportation planning problems, they remain computationally challenging due to their combinatorial structure, and their difficulty increases rapidly as the number of locations, vehicles, and operational constraints grows. In practice, transportation analytics therefore blends exact optimization (e.g., mixed-integer programming) with heuristics and metaheuristics that trade optimality guarantees for computational tractability ( 26 ) . The increasing scale and complexity of modern transportation systems further amplify these computational challenges. Urban freight systems involve interactions among logistics providers, infrastructure constraints, regulatory policies, and stochastic demand patterns. Evaluating alternative system configurations frequently requires exploring large combinatorial search spaces while simultaneously accounting for operational feasibility constraints. These characteristics motivate the exploration of new computational paradigms capable of complementing classical optimization methods. Quantum computing has recently emerged as a potential framework for solving certain classes of combinatorial optimization problems. Many discrete optimization models—including routing, scheduling, and assignment problems—can be expressed as Quadratic Unconstrained Binary Optimization (QUBO) formulations or equivalently as Ising Hamiltonians ( 29 ) , enabling them to be represented within quantum optimization algorithms. Two prominent approaches are adiabatic quantum computing ( 17 ) and the Quantum Approximate Optimization Algorithm (QAOA) ( 18 ) . Both frameworks aim to prepare quantum states whose measurement yields high-quality solutions to the underlying optimization problem. Despite these promising theoretical developments, practical deployment of quantum optimization for transportation systems faces significant challenges. Current quantum devices operate in the Noisy Intermediate-Scale Quantum (NISQ) regime, where circuit depth, qubit connectivity, and noise levels impose strict limits on executable algorithms ( 33 ) . In particular, digitized implementations of adiabatic evolution often require deep circuits containing many sequential two-qubit gates, while variational algorithms such as QAOA introduce high-dimensional parameter optimization problems that can be difficult to train. Consequently, designing hardware-efficient quantum algorithms remains a central research challenge. A key observation motivating this work is that the computational difficulty of adiabatic evolution is not uniform throughout the evolution process. The spectral structure of the system Hamiltonian typically exhibits distinct regimes characterized by different energy-gap behaviors. Early and late portions of the evolution often possess relatively large spectral gaps (difference between the lowest and second lowest eigenvalues), making the dynamics more robust to perturbations, while intermediate regions may contain narrow spectral gaps associated with avoided crossings that govern the difficulty of reaching the ground state. In fact, the running time of the adiabatic computation is directly determined by the minimal spectral gap ( 2 ) . Although directly measuring spectral gaps on current quantum hardware is generally infeasible, the presence of such gap-critical regions often manifests indirectly through increased state mixing and broader cost distributions in sampled measurement outcomes. Unlike quantum annealers, which perform analog optimization via continuous-time evolution, our approach targets gate-based quantum computers, where annealing dynamics are discretized into circuits that must operate under depth and connectivity constraints. In this work, we therefore adopt a hardware-accessible diagnostic based on the statistical properties of measured solution distributions. In particular, moments of the sampled objective function, such as energy variance of the cost distribution and susceptibility of the Hamiltonian, serve as empirical indicators of state mixing during the evolution. These quantities can be computed directly from measured bitstring samples without requiring full state tomography or explicit Hamiltonian diagonalization. Peaks in these quantities can indicate regions where the evolving quantum state spreads across multiple nearby energy levels, providing a practical proxy for gap-sensitive regions of the optimization landscape. Building on this observation, this paper proposes hardware-efficient quantum architectures for transportation optimization by comparing circuit families that are implementable today. The proposed approach decomposes digitized adiabatic evolution into two components. First, the early portion of the evolution is compressed using Approximate Quantum Compilation (AQC) ( 35 ; 21 ) , which replaces multiple Trotterized time-evolution steps with a shallower circuit approximation. Second, the remaining portion of the evolution is replaced with structured variational circuits based on either standard QAOA or a Linear-Chain QAOA (LC-QAOA) architecture designed to reduce two-qubit gate depth under hardware connectivity constraints. The concept is shown in Figure 1 . The goal of this study is to analyze the computational tradeoffs associated with different quantum algorithm architectures rather than computational

Read the complete text