Resource allocation pattern in infrastructure networks

TL;DR

This paper examines how decentralized infrastructure networks allocate resources, revealing that smaller capacity elements tend to operate with larger unoccupied capacities, which impacts their vulnerability to cascading failures.

Contribution

It uncovers the relationship between capacity and load in infrastructure networks, challenging previous assumptions and highlighting the evolution towards local failure minimization.

Findings

Smaller capacity network elements have larger unoccupied capacity portions.

Real systems show smaller load-to-capacity ratios on low-capacity elements.

Networks are optimized to prevent local failures, not large-scale cascades.

Abstract

Most infrastructure networks evolve and operate in a decentralized fashion, which may adversely impact the allocation of resources across the system. Here we investigate this question by focusing on the relation between capacity and load in various such networks. We find that, due to network traffic fluctuations, real systems tend to have larger unoccupied portions of the capacities--smaller load-to-capacity ratios--on network elements with smaller capacities, which contrasts with key assumptions involved in previous studies. This finding suggests that infrastructure networks have evolved to minimize local failures but not necessarily large-scale failures that can be caused by the cascading spread of local damage.