[Paper Review] Optimal Resource Allocation for Network Protection: A Geometric Programming Approach.

Víctor M. Preciado, Michael Zargham+3 more|arXiv (Cornell University)|Sep 24, 2013

Complex Network Analysis Techniques24 references31 citations

TL;DR

This paper proposes a geometric programming approach to optimally allocate preventive and corrective resources in directed networks to contain spreading processes like epidemics. It proves both budget-constrained and minimum-budget resource allocation problems are solvable in polynomial time, even with node-dependent costs, and validates the method on an air transportation network model.

ABSTRACT

We study the problem of containing spreading processes in arbitrary directed networks by distributing protection resources throughout the nodes of the network. We consider two types of protection resources are available: (i) Preventive resources able to defend nodes against the spreading (such as vaccines in a viral infection process), and (ii) corrective resources able to neutralize the spreading after it has reached a node (such as antidotes). We assume that both preventive and corrective resources have an associated cost and study the problem of finding the cost-optimal distribution of resources throughout the nodes of the network. We analyze these questions in the context of viral spreading processes in directed networks. We study the following two problems: (i) Given a fixed budget, find the optimal allocation of preventive and corrective resources in the network to achieve the highest level of containment, and (ii) when a budget is not specified, find the minimum budget required to control the spreading process. We show that both resource allocation problems can be solved in polynomial time using Geometric Programming (GP) for arbitrary directed graphs of nonidentical nodes and a wide class of cost functions. Furthermore, our approach allows to optimize simultaneously over both preventive and corrective resources, even in the case of cost functions being nodedependent. We illustrate our approach by designing optimal protection strategies to contain an epidemic outbreak that propagates through an air transportation network.

Motivation & Objective

• To address the challenge of efficiently containing spreading processes—such as epidemics—in arbitrary directed networks using limited protection resources.
• To simultaneously optimize both preventive (e.g., vaccines) and corrective (e.g., antidotes) resource distributions under general, possibly node-dependent cost functions.
• To determine the minimum budget required to fully contain a spreading process in a directed network, regardless of node heterogeneity.
• To solve the budget-constrained problem of maximizing containment for a fixed resource budget, ensuring cost-optimality across all nodes.
• To provide a scalable, polynomial-time solution applicable to real-world networks such as air transportation systems with complex topologies.

Proposed method

• Formulates the network protection problem as a geometric program (GP), leveraging posynomial inequalities to model resource allocation constraints and objectives.
• Models both preventive and corrective resources as decision variables per node, with individual node-specific cost functions for each type of resource.
• Uses duality theory and convex optimization techniques inherent to geometric programming to ensure global optimality and polynomial-time solvability.
• Transforms the original non-convex, combinatorial problem into a convex optimization framework via logarithmic changes of variables.
• Applies Karush-Kuhn-Tucker (KKT) conditions to derive optimality conditions for the GP formulation, enabling efficient computation.
• Validates the approach on a realistic air transportation network model, simulating epidemic spread and optimal protection deployment.

Experimental results

Research questions

• RQ1 Can we find the optimal distribution of preventive and corrective resources in a directed network to maximize containment under a fixed budget?
• RQ2 What is the minimum total cost required to fully contain a spreading process in a directed network with heterogeneous nodes?
• RQ3 How can we simultaneously optimize over both preventive and corrective resources when their costs vary per node?
• RQ4 Does the proposed method scale efficiently to large, arbitrary directed networks with non-identical nodes and complex cost structures?
• RQ5 Can the geometric programming framework handle realistic, real-world network topologies such as air traffic networks with non-uniform spreading dynamics?

Key findings

• The proposed geometric programming approach solves both the budget-constrained and minimum-budget containment problems in polynomial time for arbitrary directed graphs.
• The method supports node-dependent cost functions for both preventive and corrective resources, enabling realistic modeling of heterogeneous network environments.
• Optimal allocation of both resource types can be computed simultaneously, avoiding suboptimal trade-offs between prevention and correction.
• The framework is scalable and applicable to real-world networks, as demonstrated by its successful application to an air transportation network model.
• The solution achieves full containment of the spreading process with minimal cost, and the optimal resource distribution is explicitly computable via standard GP solvers.