Arc Routing Problem

The Capacitated Arc Routing Problem (CARP) focuses on finding the most efficient route for a vehicle to service a set of edges (e.g., streets needing cleaning) within capacity limits, minimizing total travel distance. Recent research emphasizes developing efficient algorithms, particularly deep reinforcement learning models and advanced metaheuristics like ant colony optimization and genetic algorithms, to solve CARP and its variations, such as those with stochastic demands or load-dependent costs. These improvements aim to provide high-quality solutions for large-scale instances, bridging the gap between the performance of neural network approaches and established metaheuristics. This work has significant implications for logistics, waste management, and other applications requiring optimized route planning.