Optimization of the p-Hub Median Problem via Artificial Immune Systems

10th International Conference, ICCL 2019,Barranquilla, Colombia, September 30 – October 2, 2019 Proceedings, Sept. 21, 2019

Recent advances in logistics, transportation and in telecommunications offer great opportunities to citizens and organizations in a globally-connected world, but they also arise a vast number of complex challenges that decision makers must face. In this context, a popular optimization problem with practical applications to the design of hub-and-spoke networks is analyzed: the Uncapacitated Single Allocation p-Hub Median Problem (USApHMP) where a fixed number of hubs have unlimited capacity, each non-hub node is allocated to a single hub and the number of hubs is known in advance. An immune inspired metaheuristic is proposed to solve the problem in deterministic scenarios. In order to show its efficiency, a series of computational tests are carried out using small and large size instances from the Australian Post dataset with node sizes up to 200. The results contribute to a deeper understanding of the effectiveness of the employed metaheuristic for solving the USApHMP in small and large networks.

Download the article

Metaheuristics in Telecommunication Systems: Network Design, Routing, and Allocation Problems

IEEE Systems Journal, Dec. 8, 2018

Recent advances in the telecommunication industry offer great opportunities to citizens and organizations in a globally connected world, but they also arise a vast number of complex challenges that decision makers must face. Some of these challenges can be modeled as combinatorial optimization problems (COPs). Frequently, these COPs are large-size, NP-hard, and must be solved in “real time,” which makes necessary the use of metaheuristics. The first goal of this paper is to provide a review on how metaheuristics have been used so far to deal with COPs associated with telecommunication systems, detecting the main trends and challenges. Particularly, the analysis focuses on the network design, routing, and allocation problems. In addition, due to the nature of these challenges, the paper discusses how the hybridization of metaheuristics with methodologies such as simulation and machine learning can be employed to extend the capabilities of metaheuristics when solving stochastic and dynamic COPs in the telecommunication industry.

Download the article

Artificial Immune Systems for Solving the Uncapacitated Single-Allocation p-Hub Median Problem

Proceedings of XV Encontro Nacional de Inteligência Artificial e Computacional (ENIAC), Oct. 22, 2018

Optimization problems such as the Uncapacitated Single-Allocation p-Hub Median Problem represent good models for real network design issues, hence an increasing research interest has emerged. A good hub location reduces costs and improves the quality of delivered services on network-based systems. In this work, two artificial immune systems are employed in order to address the problem, where the numerical results indicate good quality of solutions.

Download the article

Immune-Inspired Optimization with Autocorrentropy Function for Blind Inversion of Wiener Systems

Proceedings of the IEEE Congress on Evolutionary Computation, July 8, 2018

Blind inversion of nonlinear systems is a complex task that requires some sort of prior information about the source e.g. whether it is composed of independent samples or, particularly in this work, a dependence “signature” which is assumed to be known via the autocorrentropy function. Furthermore, it involves the solution of a nonlinear, multimodal optimization problem to determine the parameters of the inverse model. Thus, we propose a blind method for Wiener systems inversion, which is composed of a correntropy-based criterion in association to the well-known CLONALG immune-inspired optimization metaheuristic. The empirical results validate the methodology for continuous and discrete signals.

Download the article

A Simheuristic Algorithm for the Uncapacitated and Stochastic Hub Location Problem

Proceedings of 12th Metaheuristic International Conference, July 4, 2018

Telecommunication network-design problems have gained attention over the last decades. Among them, the problem looking for a network design that minimizes total costs, while satisfying users’ demands is well-known. This is the particular case of the Hub Location Problem (HLP). In data communications, a hub is a place of convergence where data arrive from one or more directions and are forwarded in one or more other directions. In this work-in-progress we proposed the application of a simheuristic algorithm for solving the stochastic uncapacitated HLP in which uncertainty is associated to transportation costs.

Download the article

A 2-Stage Biased-Randomized Iterated Local Search for the Uncapacitated Single Allocation p-Hub Median Problem

Transactions on Emerging Telecommunications Technologies, March 11, 2018

The hub location problem has gained attention over the last decades. In telecommunications network, a hub is a place of concurrence in where the work of the network is centralized with the purpose of delivering out the data that arrives from one or more directions to other destinations. There are different versions of the hub location problem depending upon (1) the existence or not of restrictions on the capacity related to the volume of flow a hub is allowed to support, (2) the existence or not of a set-up cost associated with selecting any node as a hub, etc. In these types of configurations, the hubs serve as connection point between 2 installations, allowing, in this way, to replace a large amount of direct connections between all pair of the nodes, therefore, minimizing the total transportation cost of the network. Thus, this work proposes a 2-stage metaheuristic based on the combination of biased-randomized techniquewith an iterated local search framework for solving the uncapacitated single allocation p-hub median problem, with computational results that validate the methodology for large-size instances from the literature.

Download the article

An Immune-Inspired, Dependence-Based Approach to Blind Inversion of Wiener Systems

Proceedings of the European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning., April 27, 2016

In this work, we present a comparative analysis of two methods — based on the autocorrelation and autocorrentropy functions — for representing the time structure of a given signal in the context of the unsupervised inversion of Wiener systems by Hammerstein systems. Linear stages with and without feedback are considered and an immune-inspired algorithm is used to allow parameter optimization without the need for manipulating the cost function, and also with a significant probability of global convergence. The results indicate that both functions provide effective means for system inversion and also illustrate the effect of linear feedback on the overall system performance.

Download the article