Improved Approximation Algorithms for Multidepot Capacitated Vehicle Routing

The Multidepot Capacitated Vehicle Routing Problem (MCVRP) is a well-known variant of the classic Capacitated Vehicle Routing Problem (CVRP), where we need to route capacitated vehicles located in multiple depots to serve customers' demand such that each vehicle must return to the depot it starts, and the total traveling distance is minimized. There are three variants of MCVRP according to the property of the demand: unit-demand, splittable and unsplittable. We study approximation algorithms for $k$-MCVRP in metric graphs where $k$ is the capacity of each vehicle, and all three versions are APX-hard for any constant $k\geq 3$. Previously, Li and Simchi-Levi proposed a $(2\alpha+1-\alpha/k)$-approximation algorithm for splittable and unit-demand $k$-MCVRP and a $(2\alpha+2-2\alpha/k)$-approximation algorithm for unsplittable $k$-MCVRP, where $\alpha=3/2-10^{-36}$ is the current best approximation ratio for metric TSP. Harks et al. further improved the ratio to 4 for the unsplittable case. We give a $(4-1/1500)$-approximation algorithm for unit-demand and splittable $k$-MCVRP, and a $(4-1/50000)$-approximation algorithm for unsplittable $k$-MCVRP. Furthermore, we give a $(3+\ln2-\max\{\Theta(1/\sqrt{k}),1/9000\})$-approximation algorithm for splittable and unit-demand $k$-MCVRP, and a $(3+\ln2-\Theta(1/\sqrt{k}))$-approximation algorithm for unsplittable $k$-MCVRP under the assumption that the capacity $k$ is a fixed constant. Our results are based on recent progress in approximating CVRP.