On the Hardness of Network Design for Bottleneck Routing Games

TL;DR

This paper studies the complexity of improving network performance in bottleneck routing games by edge removal, revealing NP-hardness results and proposing methods for near-optimal subnetworks under certain conditions.

Contribution

It establishes the NP-hardness of recognizing and approximating network design improvements for bottleneck routing games, and provides algorithms for near-optimal subnetworks in specific cases.

Findings

Bottleneck routing games do not suffer from Braess's paradox in series-parallel networks.

Recognizing instances with improved Price of Anarchy is NP-hard.

An algorithm for computing near-optimal subnetworks with quasipolynomial time complexity.

Abstract

In routing games, the network performance at equilibrium can be significantly improved if we remove some edges from the network. This counterintuitive fact, widely known as Braess's paradox, gives rise to the (selfish) network design problem, where we seek to recognize routing games suffering from the paradox, and to improve the equilibrium performance by edge removal. In this work, we investigate the computational complexity and the approximability of the network design problem for non-atomic bottleneck routing games, where the individual cost of each player is the bottleneck cost of her path, and the social cost is the bottleneck cost of the network. We first show that bottleneck routing games do not suffer from Braess's paradox either if the network is series-parallel, or if we consider only subpath-optimal Nash flows. On the negative side, we prove that even for games with strictly increasing linear latencies, it is NP-hard not only to recognize instances suffering from the paradox, but also to distinguish between instances for which the Price of Anarchy (PoA) can decrease to 1 and instances for which the PoA is as large as \Omega(n^{0.121}) and cannot improve by edge removal. Thus, the network design problem for such games is NP-hard to approximate within a factor of O(n^{0.121-\eps}), for any constant \eps > 0. On the positive side, we show how to compute an almost optimal subnetwork w.r.t. the bottleneck cost of its worst Nash flow, when the worst Nash flow in the best subnetwork routes a non-negligible amount of flow on all used edges. The running time is determined by the total number of paths, and is quasipolynomial when the number of paths is quasipolynomial.