Towards Carbon-Efficient Urban Logistics: A Constructive Routing Framework for Heterogeneous Courier Fleets

An Intelligent and Sustainable Green Supply Chain Optimization of Transportation Systems Using Classical and Quantum-Powered Computing

Transportation strategists increasingly handle heterogeneous fleets under tight service levels, fluctuating costs, and stated environmental targets. This paper addresses the operational Heterogeneous Vehicle Routing Problem (HVRP) with a practical, mathematically light decision framework that treats sustainability as the main goal of the design. We introduce a Quantum-Inspired Supply Chain and Logistics Optimization Algorithm (QISCLOA), solver-independent HVRP approach for transportation system. Our proposed framework builds a curated, capacity-feasible, and diversified route pool using multi-strategy generation. In our $16\times 16$ HVRP benchmarks, our proposed algorithm reduces elapsed time by up to 98.6% compared with strong classical formulations, while matching classical distance quality and delivering higher load utilization (95–99%) under the same feasibility constraints. Compared with existing classical system, our proposed system preserves full feasibility and comparable distance, shows clear distance-per-demand gains in symmetric unit-demand cases, and attains near-zero solve time on identical Binary Quadratic Models (BQM), evidencing orders-of-magnitude speedups. The proposed model further introduces a load–carbon emission correlation with speed, road-condition, and congestion factors, and the proposed Quantum Annealing (QA) and existing Quantum Approximate Optimization Algorithm (QAOA) comparison demonstrates competitive solution quality, solver-dependent efficiency, and full logical hardware compatibility. Additionally, we extend the formulation to a multi-depot setting with depot availability, dispatch-capacity, time-window, and Electric Vehicle (EV) charging constraints, and validate it on 21 Multi-Depot Vehicle Routing Problem (MDVRP) benchmark instances to demonstrate scalable multi-depot route selection, while a reduced comparison confirms competitive solution quality and solver-dependent efficiency, with existing QAOA improving overall route distance by 0.63% and proposed QA achieving a 9.14% overall runtime advantage. In multitasking Capacitated Vehicle Routing Problem (CVRP) benchmarks used in explicit evolutionary multitasking studies, our approach achieves lower routing cost for both best and average results with a win rate of 81.48%, indicating a lower emission of CO2. Empirically, our proposed method outperforms competitive state-of-the-art approaches in our evaluation metrics while maintaining core feasibility.

An Optimization of the Two-Echelon Vehicle Routing Problem with Drones Considering Truck Travel Time and Carbon Emissions

The Two-Echelon Vehicle Routing Problem with Drones (2E VRP-D) model can initiate flights from the truck, complete several deliveries to different customer locations, and then rendezvous with the truck again. In addition to economic benefits, logistics providers must consider the environmental impacts of the order-fulfillment process. A novel multi-objective optimization framework is established in this study to simultaneously minimize the total time required for truck travel while also reducing total carbon emissions. Due to restrictions in payload capacity and battery energy limits, drones need to work alongside trucks to deliver services effectively. A dynamic energy consumption model is applied for the drone, where energy use changes based on the loading rate, enabling a more realistic representation of actual operations. This complex problem is addressed using the Non-Dominated Sorting Genetic Algorithm (NSGA-II) with two approaches: the Giant Chromosome (GC) and K-means methods. Routing plans for both trucks and drones are then constructed using a novel heuristic algorithm. Overall, the K-means method delivers better average objective values, reflecting enhanced exploitation performance. Conversely, the GC method produces a higher Hypervolume (HV), indicating superior convergence and coverage of the Pareto front, supported by a lower spacing value, while K-means achieves a slightly better spread. These outcomes contribute to improving logistics operations and informing government policy decisions.

Genetic algorithm for agricultural logistics optimization: solving the capacitated vehicle routing problem in rural distribution

With the rapid advancement of agricultural modernization and the increasing demand for agricultural products, inefficient logistics distribution has become a major bottleneck in rural supply chains. This study addresses the capacitated vehicle routing problem (CVRP) in agricultural logistics. A genetic algorithm (GA)-based optimization model was proposed to enhance distribution efficiency. The model integrates critical agricultural characteristics, including multidistribution center networks, seasonal delivery schedules, and regional road infrastructure constraints, to minimize both transportation distance and operational costs. Experimental results show that the GA outperforms traditional metaheuristic methods (e.g., particle swarm optimization and simulated annealing), achieving a >5 km reduction in total delivery distance, an 11% decrease in delivery time, and a 5% reduction in path distance compared to conventional planning approaches. Notably, the hybrid GA-CVRP framework converges faster and achieves higher cost efficiency, with empirical tests validating its ability to optimize route planning under complex rural conditions. This research provides a robust, data-driven solution for agricultural enterprises to enhance supply chain resilience, reduce carbon footprints, and support sustainable rural development. By bridging AI-driven optimization and agricultural logistics practices, the study offers practical insights for deploying intelligent routing systems in global rural contexts.

A constraint-aware adaptive pigeon-inspired optimization algorithm for the green pickup and delivery problem with time windows

A Multi-Strategy Improved Ant Colony Optimization Algorithm for the Electric Vehicle Routing Problem with Time Windows and En-Route Recharging

With the continued electrification and digitalization of urban logistics, electric freight routing increasingly requires the coordinated consideration of customer time windows, vehicle capacity, limited battery range, and en-route charging. This study formulates an electric vehicle routing problem with time windows (EVRPTW) for smart-city electric freight and develops a multi-strategy improved ant colony optimization algorithm (IACO). The proposed model integrates customer service, route continuity, time windows, vehicle capacity, battery-energy propagation, and en-route charging. IACO combines a route–charging-state representation with feasibility-guided sweep-insertion initialization, max–min pheromone control, multi-representative guidance, reachable charging-station insertion, greedy feasibility repair, and 2-opt local search, forming a multi-stage search process that integrates global exploration, feasibility restoration, and local intensification. Computational experiments on an R-C benchmark scenario with 51 customers and 9 charging stations compare IACO with ACO, GA, TS, LNS, SA, PSO, and WOA over 100 independent runs under a common 300-iteration limit. Under the current experimental protocol, IACO records a representative generalized cost of 570.88, with reductions of 7.75–44.65% relative to the seven comparison methods, while its median CPU time is 28.42 s. These results demonstrate a clear solution-quality–computation trade-off and indicate the potential of IACO for plan-level electric freight routing and en-route charging coordination.

Truck-drone collaborative delivery route optimization with carbon cost under normal-emergency dual-use

Urban freight systems in megacities must simultaneously address carbon reduction targets and emergency preparedness requirements. This study develops a systems-oriented optimization framework for truck-drone collaborative delivery that integrates both objectives. Using Chengdu as a case study, we propose a dual-mode routing model that explicitly incorporates carbon costs as a carbon tax into the objective function. The model is solved using an Adaptive Large Neighborhood Search algorithm enhanced with Q-learning (ALNS-RL), featuring carbon-sensitive destruction and battery-constrained repair operators. We calibrate the model using Chengdu’s emission inventory across nine transport modes (2020–2025) and simulate it under carbon tax scenarios of 80, 150, and 300 CNY/ton. Results show that the drone usage ratio increases from 10.3% to 25.0% as the carbon tax rises, with a marginal incentive threshold near 150 CNY/ton. Under emergency mode, drone usage increases by 20%, while total cost and carbon emissions decrease by 4.3% and 4.2%, respectively. Overall, the proposed method reduces total cost by 31.0% and carbon emissions by 29.2% relative to a pure truck baseline. This study provides a practical decision-support framework for designing low-carbon and resilient urban logistics systems in megacities.