Juan M. Alberola

Performance evaluation of open-source multiagent platforms

Nowadays, most multiagent platforms are internally designed as middleware and are usually implemented in Java and run on top of an operating system. This kind of design maximizes portability and reduces the development cost; however, it may lead to low performance and scalability. In this context, our research has the long-term goal of integrating into the operating system some key services which are currently supported by middleware platforms. The first step in achieving this goal is to study some well-known, open-source platforms in order to understand to what extent the internal design of a platform influences its performance.

A performance evaluation of three multiagent platforms

In the last few years, many researchers have focused on testing the performance of Multiagent Platforms. Results obtained show a lack of performance and scalability on current Multiagent Platforms, but the existing research does not tackle poor efficiency causes. This article is aimed not only at testing the performance of Multiagent Platforms but also the discovery of Multiagent Platform design decisions that can lead to these deficiencies. Therefore, we are able to understand to what extent the internal design of a Multiagent Platform affects its performance. The experiments performed are focused on the features involved in agent communication.

A Scalable Multiagent Platform for Large Systems

A new generation of open and dynamic systems requires execution frameworks that are capable of being efficient and scalable when large populations of agents are launched. These frameworks must provide efficient support for systems of this kind, by means of an efficient messaging service, agent group management, security issues, etc. To cope with these requirements, in this paper, we present a novel Multiagent Platform that has been developed at the Operating System level. This feature provides high efficiency rates and scalability compared to other high-performance middleware-based Multiagent Platforms.

A group-oriented secure multiagent platform

Security is becoming a major concern in multiagent systems, since an agents incorrect or inappropriate behaviour may cause non‐desired effects, such as money and data loss. Some multiagent platforms (MAP) are now providing baseline security features, such as authentication, authorization, integrity and confidentiality. However, they fail to support other features related to the sociability skills of agents such as agent groups. What is more, none of the listed MAPs provide a mechanism for preserving the privacy of the users (regarding their identities) that run their agents on such MAPs. In this paper, we present the security infrastructure (SI) of the Magentix MAP, which supports agent groups and preserves user identity privacy. The SI is based on identities that are assigned to all the different entities found in Magentix (users, agents and agent groups). We also provide an evaluation of the SI describing an example application built on top of Magentix and a performance evaluation of it. Copyright

An artificial intelligence tool for heterogeneous team formation in the classroom

Abstract Nowadays, there is increasing interest in the development of teamwork skills in the educational context. This growing interest is motivated by its pedagogical effectiveness and the fact that, in labour contexts, enterprises organise their employees in teams to carry out complex projects. Despite its crucial importance in the classroom and industry, there is a lack of support for the team formation process. Not only do many factors influence team performance, but the problem becomes exponentially costly if teams are to be optimised. In this article, we propose a tool whose aim it is to cover such a gap. It combines artificial intelligence techniques such as coalition structure generation, Bayesian learning, and Belbin’s role theory to facilitate the generation of working groups in an educational context. This tool improves current state of the art proposals in three ways: i) it takes into account the feedback of other teammates in order to establish the most predominant role of a student instead of self-perception questionnaires; ii) it handles uncertainty with regard to each student’s predominant team role; iii) it is iterative since it considers information from several interactions in order to improve the estimation of role assignments. We tested the performance of the proposed tool in an experiment involving students that took part in three different team activities. The experiments suggest that the proposed tool is able to improve different teamwork aspects such as team dynamics and student satisfaction.

Multidimensional Adaptation in MAS Organizations

Organization adaptation requires determining the consequences of applying changes not only in terms of the benefits provided but also measuring the adaptation costs as well as the impact that these changes have on all of the components of the organization. In this paper, we provide an approach for adaptation in multiagent systems based on a multidimensional transition deliberation mechanism (MTDM). This approach considers transitions in multiple dimensions and is aimed at obtaining the adaptation with the highest potential for improvement in utility based on the costs of adaptation. The approach provides an accurate measurement of the impact of the adaptation since it determines the organization that is to be transitioned to as well as the changes required to carry out this transition. We show an example of adaptation in a service provider network environment in order to demonstrate that the measurement of the adaptation consequences taken by the MTDM improves the organization performance more than the other approaches.

Cost-Aware reorganization service for multiagent systems

Reorganization in Multigent Systems is aimed at providing support to dynamically adapt the structure and the behaviour of organizations. Current reorganization approaches are mainly focused on providing reorganization solutions that take the benefits of the future organization into account but that do not include the impact of the reorganization costs in the process. Therefore, the costs for achieving future instances of an organization cannot be computed until the reorganization process ends. Organization transition provides a paradigm for relating two different instances of the same organization at different moments. In this paper, we provide a Reorganization Facilitator Service that implements a cost-aware reorganization mechanism that is based on organization transitions. This service provides the associated costs for transition from a current organization to a future organization and the sequence of steps required for this transition. The paper also presents two different examples of organization transition in order to illustrate the use of the proposed service.

A Self-configurable Agent-Based System for Intelligent Storage in Smart Grid

Next generation of smart grid technologies demand intelligent capabilities for communication, interaction, monitoring, storage, and energy transmission. Multiagent systems are envisioned to provide autonomic and adaptability features to these systems in order to gain advantage in their current environments. In this paper we present a mechanism for providing distributed energy storage systems (DESSs) with intelligent capabilities. In more detail, we propose a self-configurable mechanism which allows a DESS to adapt itself according to the future environmental requirements. This mechanism is aimed at reducing the costs at which electricity is purchased from the market.

Open Issues in Multiagent System Reorganization

In the last few years, the demand for technologies and methodologies for supporting dynamic and open Multiagent Systems has increased. Thus, several approaches for reorganization in Multiagent Systems have emerged to dynamically adapt agent organizations. Current approaches for reorganization provide support to allow agents to enter or exit the system, to change the roles played by agents, to modify the norms that regulate the system, etc. In an effort to identify the dearth of support in current approaches, this paper presents a comparison and analysis of the most relevant approaches for reorganization. Then, after an evaluation, we indicate some future research lines with regard to reorganization in Multiagent Systems.

Simulating a Collective Intelligence Approach to Student Team Formation

Teamwork is now a critical competence in the higher education area, and it has become a critical task in educational and management environments. Unfortunately, looking for optimal or near optimal teams is a costly task for humans due to the exponential number of outcomes. For this reason, in this paper we present a computer-aided policy that facilitates the automatic generation of near optimal teams based on collective intelligence, coalition structure generation, and Bayesian learning. We carried out simulations in hypothetic classroom scenarios that show that the policy is capable of converging towards the optimal solution as long as students do not have great difficulties evaluating others.