Abstract:We give a probabilistic analysis of the unit-demand Euclidean capacitated vehicle routing problem in the random setting, where the input distribution consists of $n$ unit-demand customers modeled as independent, identically distributed uniform random points in the two-dimensional plane. The objective is to visit every customer using a set of routes of minimum total length, such that each route visits at most $k$ customers, where $k$ is the capacity of a vehicle. All of the following results are in the random setting and hold asymptotically almost surely. The best known polynomial-time approximation for this problem is the iterated tour partitioning (ITP) algorithm, introduced in 1985 by Haimovich and Rinnooy Kan. They showed that the ITP algorithm is near-optimal when $k$ is either $o(\sqrt{n})$ or $\omega(\sqrt{n})$, and they asked whether the ITP algorithm was also effective in the intermediate range. In this work, we show that when $k=\sqrt{n}$, the ITP algorithm is at best a $(1+c_0)$-approximation for some positive constant $c_0$. On the other hand, the approximation ratio of the ITP algorithm was known to be at most $0.995+\alpha$ due to Bompadre, Dror, and Orlin, where $\alpha$ is the approximation ratio of an algorithm for the traveling salesman problem. In this work, we improve the upper bound on the approximation ratio of the ITP algorithm to $0.915+\alpha$. Our analysis is based on a new lower bound on the optimal cost for the metric capacitated vehicle routing problem, which may be of independent interest.

Random generation of k-interactive capacities

An improvement of Random Node Generator for the uniform generation of capacities

Axiomatic structure of k-additive capacities

Pseudo-random Instance Generators in C++ for Deterministic and Stochastic Multi-commodity Network Design Problems

Bi-capacities -- Part I: definition, Möbius transform and interaction

Maximizing approximately k-submodular functions

Decision Rule-Based Method for Flexible Multi-Facility Capacity Expansion Problem

Weighting Models to Generate Weights and Capacities in Multicriteria Group Decision Making

Probabilistic Analysis of Euclidean Capacitated Vehicle Routing

Optimal Capacity Investment Decisions with Two-Sided Fixed-Capacity Adjustment Costs

The capacity constrained facility location problem

Capacity Planning in Stable Matching

Self-adaptive randomized constructive heuristics for the multi-item capacitated lot sizing problem

Capacity planning for integrated energy system based on reinforcement learning and multi-criteria evaluation

Availability Evaluation and Design Optimization of Multi-State k-out-of-n: G Systems With Random Performance Requirements

Computation of Riesz [math]-Capacity [math] of General Sets in [math] Using Stable Random Walks

Generation of discrete random variables in scalable frameworks

Lateral capacity exchange and its impact on capacity investment decisions

A strong law of large numbers for capacities

Supply Process Improvement Decisions for a Newsvendor with Random Capacity.

Fuzzy random programming problems with multiple criteria