Multi-agent System Approach Research Articles

Multi-agent system approach for improved real-time visual summaries of geographical data streams

Interdisciplinary efforts are needed to support visual exploration and analysis of spatio-temporal data streams from sensor networks, depending on the size and complexity of data that must be analysed before being displayed. For this, several emerging approaches have been proposed known as 'Visual Summaries' of datasets that will help users find what is most important and interesting to visualise in the mass of available information. In this paper, we present an approach for generating automatically the visual summaries in real time. For this, we have adopted chorem-based visual representations of territories issued from both geometric and semantic generalisation. Our approach relies on a multi-agent framework; an extraction knowledge agent is able to extract important spatiotemporal patterns of data streams coming from a sensor network as interesting regions, and a visualisation agent which displays those patterns as simplified maps. We validate this model with an example taken from the meteorology field.

Potential of a Multi-Agent System Approach for Production Control in Smart Factories

Digitalization and Industry 4.0 allow for the development of new approaches to deal with classic industrial problems, such as production planning and control. In this context, Multi-Agent Systems (MAS) are a promising approach to exploit the new technologies in order to achieve improved planning and control performance. This paper applies a MAS approach to control the production in job shop manufacturing systems. Within a simulation study based on a real industrial case, the approach achieved a good performance compared to the standard scheduling approach applied by the considered company.

Multi-agent system approach for improved real-time visual summaries of geographical data streams

Interdisciplinary efforts are needed to support visual exploration and analysis of spatio-temporal data streams from sensor networks, depending on the size and complexity of data that must be analysed before being displayed. For this, several emerging approaches have been proposed known as 'Visual Summaries' of datasets that will help users find what is most important and interesting to visualise in the mass of available information. In this paper, we present an approach for generating automatically the visual summaries in real time. For this, we have adopted chorem-based visual representations of territories issued from both geometric and semantic generalisation. Our approach relies on a multi-agent framework; an extraction knowledge agent is able to extract important spatiotemporal patterns of data streams coming from a sensor network as interesting regions, and a visualisation agent which displays those patterns as simplified maps. We validate this model with an example taken from the meteorology field.

OVAP: A strategy to implement partial information sharing among supply chain retailers

This paper analyses the impact on supply chain performance of adopting different strategies to implement partial information sharing among heterogeneous retailers. Supply chains are modelled using a multi-agent systems approach. We find that the strategy adopted to construct the partial information sharing structure (i.e., the retailers who share information) has a significant impact on supply chain performance. We propose a practical strategy, named Order VAriance Prioritization (OVAP), which gives priority to the retailers with higher order variance. OVAP outperforms the worst (i.e. naive) implementation method by 27.2% and 7.8% with respect to the levels of bullwhip and average inventory.

Heterogeneity and Self-Organization of Complex Systems Through an Application to Financial Market with Multiagent Systems

Multiagent systems (MAS) provide a useful tool for exploring the complex dynamics and behavior of financial markets and now MAS approach has been widely implemented and documented in the empirical literature. This paper introduces the implementation of an innovative multi-scale mathematical model for a computational agent-based financial market. The paper develops a method to quantify the degree of self-organization which emerges in the system and shows that the capacity of self-organization is maximized when the agent behaviors are heterogeneous. Numerical results are presented and analyzed, showing how the global market behavior emerges from specific individual behavior interactions.

Decision Support Tool for Supply Chain Configuration Considering New Product Re-Design: An Agent-Based Approach

In the decline phase of product lifecycle, industrials need to re-design their products to introduce new functions and/or customers’ new preferences. These changes may not only affect the product’s bill of material, but also its supply chain network. Consequently, new supply chain costs are generated. This paper addresses the problem of supply chain configuration considering new product re-design using a multi-agent system (MAS). The objective of the system is to ensure good collaboration between two different points of view, supply chain partners and product designers, to make better decisions. To model the proposed system, we select the multi-agent system engineering (MaSe) methodology. The MAS framework contains three types of agents, namely, “product design agent” and “supply chain agents” which are fitted with optimization tools. These tools allow costs’ optimization and selection of supply chain means (suppliers, technologies, etc.). Finally, the system contains a “communication agent” acting like a mediator; it facilitates data exchange between designers. To support distributed decision-making, two models of mixed integer linear programming are adopted and implemented within the framework for supply chain optimization. The overall MAS approach was tested in simulation with a case study. The objective of the simulation is to choose among three product alternatives the cheapest one based on its supplying and production costs, under capacity constraints. The MAS was able to find the best product alternative among three alternatives proposed by product design team and select optimal supply chain means. The optimal supply chain contains two suppliers: one machine and one subcontractor to satisfy customer’s demand.

Self-organizing multi-agent systems for the control of complex systems

Because of the law of requisite variety, designing a controller for complex systems implies designing a complex system. In software engineering, usual top-down approaches become inadequate to design such systems. The Adaptive Multi-Agent Systems (AMAS) approach relies on the cooperative self-organization of autonomous micro-level agents to tackle macro-level complexity. This bottom-up approach provides adaptive, scalable, and robust systems. This paper presents a complex system controller that has been designed following this approach, and shows results obtained with the automatic tuning of a real internal combustion engine.

A multi-agent systems approach for sustainable supplier selection and order allocation in a partnership supply chain

Recently, incorporating sustainability into the buyer-supplier sourcing decisions has achieved a considerable amount of attentions among researchers and industrial enterprises who are attempting to move towards sustainable production. Moreover, by investigating further in the buyer-supplier relationships, the literature suggests that proper communication and structured information exchange are important components in establishing a long-term partnership and maintaining such a relationship. Toward this end, a Multi-Agent Systems (MASs) approach is proposed as a mean of automating and facilitating the process of sustainable supplier selection and order allocation (SSS&OA) resulting in a more co-operative partnership. This research shows that financial performance of manufacturing companies adopting environmental and social sustainability in their operations strategy enhanced their competitive advantage that can lead to long-term sourcing relationships for the buyer-supplier dyad. Additionally, it was also shown that applying MASs to the SSS&OA problem can be utilized as an approach to facilitate communications and automate information exchange processes in Supply Chains (SCs) where suppliers and manufacturer are looking to maintain a long-term SC partnership. The applicability of the developed MAS approach and its incorporated sustainable supplier evaluation and order allocation models is demonstrated using an adopted practical scenario from an industrial case study operating in the electronics sector in medical device industry.

SWARM INTELLIGENCE APPROACH FOR SIMULATION MODELING OF DISTRIBUTED POWER SYSTEMS

The approach for multiagent systems’ simulation models’ adaptation is proposed. It uses the methods of swarm intelligence and fuzzy logics to describe the collective behavior of decentralized system agents. The practical use of this approach is demonstrated by the simulation model of a distributed power system, as well as the use of swarm intelligence for some other systems.

Real-time battery state of charge estimation in smart grid application by Multi Agent System

This paper presents a SOC estimation model in real-time applications such as smart grid. Authors list the existing SOC estimation methods and propose a SOC model to overcome the limitations of the precedent methods presented in the literature. Furthermore, this paper proposes a scalable, self-adaptive, and generic SOC model to meet the smart grid requirement. Multi-agent system approach is proposed to manage the smart grid and a co-simulation is proposed enabling the physical variables modeling while the discrete decision-making is given by developing agents under Jade framework. The results analyses have shown a SOC model which depends on the battery use conditions that can match with all batteries technologies and therefore can be suitable for smart grid applications.

Системы управления знаниями и многоагентные системы

Работа посвящена анализу подходов к построению систем управления знаниями научно-образовательных организаций. Рассмотрены особенности СУЗ вузов и проблемы их реализации. Подчеркивается важность мотивации персонала в отчуждении знаний. Показана перспективность многоагентного подхода к созданию СУЗ вузов, в том числе с точки зрения мотивации персонала. Сформулированы основные принципы построения многоагентной СУЗ. Поставлена задача моделирования такой СУЗ в реальных условиях. Рассмотрена возможная функциональность агентов многоагентной СУЗ.

Partner selection and performance evaluation framework for a construction-related virtual enterprise: a multi-agent systems approach

ABSTRACTTo exploit market opportunities, disparate enterprises can pool their core competencies to form a temporary organisation. This inter-organisational collaboration of enterprises is commonly referred to as a virtual enterprise (VE). However, the VEs’ partner selection problems influence their overall performance. Due to the limited construction-specific studies on the evaluation and selection of partners, the current research sought to address this knowledge gap by investigating the partner selection problem. The partners comprising 10 construction companies carrying out project tasks for a large building in Nairobi, Kenya, were evaluated. Data were collected using a sequential mixed-methods approach consisting of a focus group interview, questionnaires and algorithms. The study proposed the use of the partner selection and performance evaluation technique. This technique combines fuzzy approximate reasoning with the conventional analytic hierarchy process algorithm, designed to deal with imprecise evaluators’ judgments. A multi-agent systems approach was chosen to simulate VEs. Results show that the chosen techniques are both efficient and effective. In particular, the reduced group fuzzy analytic hierarchy process reduces the number of pairwise comparisons required when comparing a large number of attributes. Application to the construction management domain, in particular, is germane to the research.

Using A Multi-Agent System Approach for Monitoring Appropriateness Criteria in Medical Imaging

: Agent technology is an emerging and promising research area in software technology, which increasingly contributes to the development of value-added information systems for large healthcare organizations due to its capability to automatically modify themselves in response to changes in their operating environment. Multi agent system (MAS) approach is the construction of a complex system as a set of entities, called agents, interacting among them in order to perform the system tasks. Patie Maes, the director of MIT media lab, deems that agent is a computer system, which locates in dynamic and complex environment, can autonomically sense the environment and act accordingly to complete its tasks or goals. Agent is a computing entity with four features of autonomy, reactivity, interaction and initiative. A MAS is composed of a large number of agents and other computational artifacts. These agents are goal-oriented components, that is, they are modeled as entities that pursue goals and choose for execution those actions that will potentially contribute to satisfy them. These choices depend on their information about the environment, past experiences and themselves. Agents are also social because they need to interact with other agents to achieve the satisfaction of goals, and these interactions are modeled in terms of information, requests and informs. In healthcare, medical imaging is a service-oriented, data-intensive and technology-driven environment; therefore it requires combining many computerized systems with different user environments to provide quality efficient services in less time. In such environments, MAS is well-suited to describe the interactions between agents (e.g., information exchanged, related tasks or reasons for that interaction) and the way in which the agent adapts to changes in it (e.g., new roles or goals) or its environment. Medical imaging is a science but also a discipline of action that often requires a decision. The complexity of decision-making, especially in medical imaging, comes from the uncertainty, for example the uncertainty of knowledge, uncertainty about the protocols and the uncertainty of the procedure used. A multi-agent system (MAS) for decision-aiding support uses and combines distributed and heterogeneous systems to ease the extraction of useful information in order to select appropriate protocol or procedure.

A novel leader following consensus approach for multi-agent systems with data loss

The objective of this paper is to develop a novel consensus algorithm or protocol for multi-agent systems in the event of communication link failure over the network. The structure or topology of the system is modeled by an algebraic graph theory, and considered as a discrete time-invariant system with a second-order dynamics. Bernoulli distribution is applied to represent the data dropout during operation. The sufficient conditions for the stabilization controller design is developed by Lyapunov-based methodologies and Linear matrix inequality (LMIs) techniques. The feasibility of the given LMIs is analyzed to verify the stabilization of controller design, which ensures the MAS to achieve the consensus. Leader-following numerical simulations with a group of agents are successfully conducted based on the effect of data losses, initial values, communication weights, and number of agents to demonstrate the effectiveness of the novel consensus algorithm in this paper. Finally, experimental studies are carried out by using two Pioneer 3-DX mobile robots and one Pioneer 3-AT mobile robot as a team to verify the proposed work.

Multi-agent system approach applied to a manufacturer’s supply chain using global objective function and learning concepts

The supply chain efficiency is essential to guarantee the producer competitiveness. As this environment encompasses intelligent actors, it can be model as a multi-agent system. In order to evaluate the interactions, competition, strategies and their consequences in this context, this study aims to analyze the multi-agent system approach applied in a manufacturer’s supply chain, considering a pull production system. Then, this study starts with the literature review about multi-agent systems and its application to supply chain. After, the proposed system is explained, describing its architecture, organization, objectives, negotiation and learning aspects. All the assumptions, individual objective functions and global objective functions are presented in this section. Following, the interactions and implementations aspects are addressed. The negotiation design was inspired in the monotonic concession protocol and the Dutch auctions. The agents learning aspect is broached by the \(\upvarepsilon \)-greedy heuristic. Finally, the experiments design, results and conclusions are presented. The results show the agents’ behavior and the overall system evaluation.

Multi-Agent System-Based Event-Triggered Hybrid Control Scheme for Energy Internet

This paper is concerned with an event-triggered hybrid control for the energy Internet based on a multi-agent system approach with which renewable energy resources can be fully utilized to meet load demand with high security and well dynamical quality. In the design of control, a multi-agent system framework is first constructed. Then, to describe fully the hybrid behaviors of all distributed energy resources and logical relationships between them, a differential hybrid Petri-net model is established, which is an original work. The most important contributions based on this model propose four types of event-triggered hybrid control strategies whereby the multi-agent system implements the hierarchical hybrid control to achieve multiple control objectives. Finally, the effectiveness of the proposed control is validated by means of simulation results.

H2 consensus control of time-delayed multi-agent systems: A frequency-domain method

An analytical H2 controller design approach of homogeneous multi-agent systems with time delays is presented to improve consensus performance. Firstly, a closed-loop multi-input multi-output framework in frequency domain is introduced, and a consensus tracking condition is given. Secondly, the decomposition method is utilized to simplify the analysis of internal stability and H2 performance index of the whole system to a set of independent optimization problems. Finally, the H2 optimal controller can be computed from all the stabilizing controllers. The contributions of the new approach are that the design procedure is conducted analytically for arbitrary delayed multi-agent systems, and a simple quantitative tuning way is developed to trade off the nominal performance and robustness. The simulation examples show the effectiveness of the proposed control strategy.

MATT: Multi Agents Testing Tool Based Nets within Nets

Testing is a software development activity, devoted to evaluating product quality and improving it by identifying defects and problems. Concerning multi agent systems, testing is a challenging task, which asks for new testing techniques dealing with their specific nature. The techniques need to be effective and adequate to evaluate agent’s autonomous behaviors and build confidence in them. The "Model Based Testing" (MBT) is a technique particularly interested among all existing solutions of tests. This latter is based on a system model, which produces abstract test cases. To run these last ones against systems under test,the abstract test cases have to be transformed to concrete ones. As part of this effort, we propose a model based testing approach for multi agent systems based on such a model called Reference net, where a tool, which aims to providing a uniform and automated approach is developed. The feasibility and the advantage of the proposed approach are shown through a short case study.

Distributed protection for smart substations based on multiple overlapping units

Taking advantage of the information sharing mechanism in smart substations, a distributed protection based on multiple overlapping units is introduced in this paper. The area division and the construction of protection units are studied based on the principle of differential protection. Mutual perception and information interaction among units, calculation with multisource information and coordinated protection schemes are proposed to improve the reliability and adaptability of the substation protection system. A multi agent system (MAS) approach is used to model the units' working process in various situations, such as communication failure and sampled data failure. Test results demonstrate that the method can accelerate the backup protection as well as reduce the power-off range.

Real-Time Smart Grids Control for Preventing Cascading Failures and Blackout using Neural Networks: Experimental Approach for N-1-1 Contingency

AbstractIn this paper, a novel intelligent control is proposed based on Artificial Neural Networks (ANN) to mitigate cascading failure (CF) and prevent blackout in smart grid systems after N-1-1 contingency condition in real-time. The fundamental contribution of this research is to deploy the machine learning concept for preventing blackout at early stages of its occurrence and to make smart grids more resilient, reliable, and robust. The proposed method provides the best action selection strategy for adaptive adjustment of generators’ output power through frequency control. This method is able to relieve congestion of transmission lines and prevent consecutive transmission line outage after N-1-1 contingency condition. The proposed ANN-based control approach is tested on an experimental 100 kW test system developed by the authors to test intelligent systems. Additionally, the proposed approach is validated on the large-scale IEEE 118-bus power system by simulation studies. Experimental results show that the ANN approach is very promising and provides accurate and robust control by preventing blackout. The technique is compared to a heuristic multi-agent system (MAS) approach based on communication interchanges. The ANN approach showed more accurate and robust response than the MAS algorithm.