Inicio / Historico / Ciencia y Tecnologia de la Salud

Scientists design a computer system that helps to make decisions when several agents do not agree

Spanish version: sl.ugr.es/sistema_decisionesES

French version: sl.ugr.es/sistema_decisionesFR

Scientists from the University of Granada have designed a consensus model that helps making decisions: from choosing a wine in a restaurant to helping different departments of a bank involved in the configuration of the stock portfolios in the investment of their funds

This work has received a major award at the Annual Congress of the IEEE System, Man and Cybernetics Society, held in Budapest (Hungary)

Scientists from the universities of Granada and Cadiz have designed a new automatic system that allows to make decisions or solve daily problems when the different agents involved do not agree.

This consensus model, which could be applied through a cellphone ‘app’ in the next stage, has many different applications: from choosing a wine in a restaurant to helping different departments of a bank involved in the configuration of the stock portfolios to invest their funds.

The system is also useful for setting diets, for the selective distribution of information in libraries, or for establishing medical and prevention treatments of lower back pain in a personalized way through the recommendation of physiotherapy exercises.

The work has been published in the IEEE Transactions on Systems, Man, and Cybernetics: Systems journal, one of the most prestigious journals in the field of Artificial Intelligence, and it has received the award for the best scientific article of the year 2014 at the IEEE System, Man and Cybernetics Society Annual Congress, held in Budapest (Hungary). Its main author, professor Enrique Herrera-Viedma, from the Department of Computer Science and Artificial Intelligence at the University of Granada (UGR), has been in charge of receiving the award.

It is a model of consensus among heterogeneous experts with different levels of relevance in the group. The researchers propose a system that combines the prestige or specific weight of each expert along with a negotiation model with the experts, which suggests the modification of the preferences according to their importance in the group in order to reach consensus in few iterations (act of repeating a process with the intention of achieving a desired goal).

 

Complex decision-making

Not surprisingly, the UGR scientists have been working for years on the development of these type of consensus models, which allow the creation of automatic systems to aid in making complex decisions, in which reaching a solution to a problem may involve several decision-makers (or agents) who work and negotiate together in order to reach an agreement or consensus on what would be the best option among a set of possible solutions.

This problem can be complicated when, in addition, it is necessary to consider different criteria and even different degrees of importance of said criteria or decision makers. The latter aspect is precisely the one analyzed in the awarded work.

This article is part of the line of research in the development of consensus models for making decisions that the UGR Soft Computing and Intelligent Information Systems research group has been carrying out for years. Not surprisingly, the professors in charge of the group, Francisco Herrera Trigueros and Enrique Herrera-Viedma, have a long history and extensive international recognition, and they collaborate with international researchers of great prestige.

The results of this line of research have been published in the most relevant journals about Artificial Intelligence, and they have produced, among others, eight highly cited publications (in the Top 1% of their category in the period 2005-2015), as well as several applications and software tools.

1-Premio-IEEETSMC-A-EHV-01-600x509

2-Premio-IEEETSMC-A-EHV-02-600x600

1 and 2. UGR professor Enrique Herrera-Viedma receives the award for the best scientific article, in the SMC2016 Congress held in Budapest (Hungary)

2-Premio-IEEETSMC-A-EHV-03-600x397

3. The designed consensus model has many very different applications, such as choosing a wine in a restaurant

Bibliographic references:

A New Consensus Model for Group Decision Making Problems With Non-Homogeneous Experts. Ignacio Javier Pérez, Francisco Javier Cabrerizo, Sergio Alonso and Enrique Herrera-Viedma. IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS: SYSTEMS, VOL. 44, NO. 4, APRIL 2014

 

Contacto:

Enrique Herrera-Viedma

Departamento de Ciencia de la Computación e Inteligencia Artificial de la Universidad de Granada

Telephone: 958 244 258

E-mail: viedma@decsai.ugr.es

Web: http://decsai.ugr.es/~viedma

 

Francisco Herrera Triguero

Departamento de Ciencia de la Computación e Inteligencia Artificial de la Universidad de Granada

Telephone: 958 240 598

E-mail: herrera@decsai.ugr.es