This work presents Collaborative Routing Constructors (CoRC), a routing framework that enables independently solved subproblems to exchange customers and vehicles during optimization rather than relying solely on a fixed partition or a subsequent global re-optimization stage.
Abstract
Large-scale Capacitated Vehicle Routing Problems (CVRPs) are commonly solved by partitioning customers into smaller routing problems that can be optimized independently. While this substantially reduces computational complexity, independently constructed routing solutions may leave some customer demand unserved even when sufficient resources exist elsewhere in the fleet. We present Collaborative Routing Constructors (CoRC), a routing framework that enables independently solved subproblems to exchange customers and vehicles during optimization rather than relying solely on a fixed partition or a subsequent global re-optimization stage. Computational experiments on AGS benchmark instances and synthetic instances containing up to 200,000 customers compare CoRC against independent routing, post-routing global re-optimization, and state-of-the-art, end-to-end routing frameworks. Across all evaluated partitioning strategies, CoRC consistently constructs feasible routing solutions where competing partition-based methods do not. Furthermore, it remains effective on problem instances for which the evaluated end-to-end routing frameworks did not produce solutions under the same computational budget. These results demonstrate that collaboration between routing subproblems provides a robust and scalable approach for feasible large-scale route construction.
We introduce the Vehicle Routing Problem with Resource-Constrained Pickup and Delivery (VRP-RPD), where agents deploy finite identical resources at customer locations for processing before retrieval and redeployment. Applications include portable medical equipment, tool rental, and disaster relief. Unlike classical pickup-and-delivery variants, VRP-RPD permits different agents to perform dropoff and pickup for the same customer—creating inter-route dependencies absent from standard formulations. We provide a complete mixed-integer linear programming formulation and demonstrate that exact methods are intractable even for small instances. Problems with 16 customers cannot be solved to optimality within two hours of computational time. We develop a Biased Random-Key Genetic Algorithm (BRKGA) with a four-gene-per-customer encoding. Two genes assign dropoff and pickup agents independently, while two priority keys determine sequencing. A simulation decoder guarantees feasibility by deferring operations until resources become available. Experiments on 14 TSPlib-derived benchmarks across five variants of processing time (base, 2X, 5X, 1R10, 1R20) compare four configurations. Warm-start BRKGA achieves 13–67% makespan reduction over the heuristics, with larger gains on higher-resource instances. Ablation tests show warm-start initialization is the primary driver of performance. Friedman tests (p < 0.01) confirm warm-start BRKGA superiority across all instance variants.
M. Sodhi, Romesh Prasad· Annual Conference on Genetic...· 0 citations
The results suggest that quantum kernels may serve as complementary routing-aware decomposition modules within classical optimization pipelines, particularly for offline structural preprocessing.
This work proposes Double-Channel Graph Attention (DCGA), an end-to-end reinforcement learning framework that isolates network reachability and demand-service logic into separate graph channels and constructs valid routes using a simulator-coupled, constraint-informed decoder.
The Coordinated Truck-Drone Routing with Time Windows and Backhauls poses a complex challenge in collaborative logistics, requiring tight coordination between ground vehicles and drones under stringent operational constraints. Unlike traditional back-haul routing models that impose a strict priority between linehaul deliveries and backhaul pickups, this paper adopts an improved backhaul paradigm that allows pickups and deliveries to be interleaved along a route. Moreover, drones operate under limited endurance and can be launched and recovered by vehicles to serve selected customers, enabling flexible cooperation while preserving synchronization. We formulate this setting as a bi-objective problem that minimizes total cost and waiting time, and develop a Mixed-Integer Linear Programming model to optimally solve small-scale instances for benchmarking. For larger instances, this paper proposes the Constraint-Integrated Adaptive Grid-based Evolutionary Algorithm (CIAGEA) that integrates adaptive Pareto grid adjustment, adaptive local search, and a diversity mechanism to balance convergence and diversity while preserving feasibility. Extensive experiments on benchmark instances show that CIAGEA consistently outperforms state-of-the-art algorithms in terms of Hypervolume and Inverted Generational Distance, achieving particularly strong gains on large-scale problems and producing solutions close to optimal on small instances.
Bui Xuan Son, Nguyen Thi Ha, Nguyen Tien Dang et al.· Annual Conference on Genetic...· 0 citations
We study the Service-Aware Multi-Commodity Flow (SAMCF) problem, in which demand is elastic and governed by a logit choice model while routing is subject to hard capacity constraints. In a centralized, system-optimal setting, the network operator jointly determines how much demand to serve and how to route it. We formulate the SAMCF as a nonlinear program and propose an iterative fixed-point heuristic that alternates between solving an inelastic MCF via column generation and updating demand from the resulting service levels. Two linear approximations based on piecewise-linear demand functions and McCormick envelopes serve as benchmarks, while a piecewise-linear outer-approximation of the demand function is used to provide valid lower bounds. Computational experiments on public transport instances show that the heuristic finds near-optimal solutions in under two seconds - orders of magnitude faster than the benchmark methods - while matching their solution quality on all instances they can solve within a ten-minute time limit.