Highlights of Practical Applications of Agents, Multi-Agent Systems, and Sustainability - The PAAMS Collection

Ambient Assisted Living (AAL) systems must provide adapted services easily accessible by a wide variety of users. This can only be possible if the communication between the user and the system is carried out through an interface that is simple, rapid, effective, and robust. Natural language interfaces such as mobile multimodal conversational agents fulfill these requisites, as they can emulate communication capabilities of a human being including several communication modalities, such as speech, tactile and visual interaction. In this paper, we present a multimodal application for Android mobile devices aimed to patients suffering from Alzheimer. This application helps them to enhance their memory and to stimulate their cognitive abilities.

In this paper, a computational model of a human agent is presented which describes the effect of social support on mood. According to the literature, social support can either refer to the social resources that individuals perceive to be available or to the support that is actually provided in problematic situations. The proposed model distinguishes between both roles of social support. Simulation experiments are done to analyze the effect of the different types of support in different scenarios. It is shown that support can help to reduce the induced stress and thus can contribute to healthy mood regulation and prevention of depression. This presented model provides a basis for an intelligent support system for people with mood regulation problems that take the social network of people into account.

A common conversation between an older adult and a nurse about health-related issues includes topics such as troubles with sleep, reasons for walking around nighttime, pain conditions, etc. This dialogue emerges from the participating human’s lines of thinking, their roles, needs and motives, while switching between topics as the dialogue unfolds. This paper presents a dialogue system that enables a human to engage in a dialogue with a software agent to reason about health-related issues in a home environment. The purpose of this work is to conduct a pilot evaluation study of a prototype system for human-agent dialogues, which is built upon a set of semantic models and integrated in a web application designed for older adults. Focus of the study was to receive qualitative results regarding purpose and content of the agent-based dialogue system, and to evaluate a method for the agent to evaluate its behavior based on the human agent’s perception of appropriateness of moves. The participants include five therapists and 11 older adults. The results show users’ feedback on the purpose of dialogues and the appropriateness of dialogues presented to them during the interaction with the software agent.

The Flexible Job Shop scheduling Problem (FJSP) is an extension of the classical Job Shop scheduling Problem (JSP) that allows to process operations on one machine out of a set of alternative machines. It is an NP-hard problem consisting of two sub-problems which are the assignment and the scheduling problems. This paper proposes a holonic multiagent model based on a combined genetic algorithm and tabu search for the FJSP. Firstly, a scheduler agent applies a Neighborhood-based Genetic Algorithm (NGA) for a global exploration of the search space. Secondly, a cluster agents set uses a local search technique to guide the research in promising regions. Numerical tests are made to evaluate our approach, based on two sets of benchmark instances from the literature of the FJSP: Kacem and Hurink. The experimental results show the efficiency of our approach in comparison to other approaches.

We study a bi-objective problem of scheduling the jobs from two agents on a single parallel-batching machine with the objective of minimizing the makespans of the both agents. All jobs have an equal processing time and non-identical job sizes. We define a boundary of Pareto-optimal set and then present two metrics to evaluate the quality of a non-dominated set. Based on two different strategies, two heuristics are proposed to generate non-dominated sets. Furthermore, we compare the performance of the heuristics with the widely used non-dominated sorting genetic algorithm (NSGA-II), and evaluate the quality of the obtained non-dominated set based on the proposed metrics. The results show that the proposed heuristics outperform NSGA-II, and the obtained non-dominated set is very close to the Pareto-optimal set.

The article is a presentation of the research results regarding selected activity coordination mechanisms. The research was carried out independently and within the framework of a research project. Reference coordination models, which serve further simulation works, are their result. The article consists of several parts. The first part discusses the most important issues regarding coordination theory. The second part discusses activity coordination in complex systems, multi-agent systems. Selected activity coordination mechanisms and their comparisons were presented in the third part. The article is concluded with a summary.

This study deals with scheduling groups of jobs, their arrival and delivery, and individual processing of each of them. All jobs in a group should be delivered at the same time after processing. One of the objectives is to minimize the average delivery time of the group containing that job (waiting period). The new way of modeling and solving the decision problem - a multi-agent hybrid approach is presented. This approach includes the design and implementation of two agent types: MP-agents (Mathematical Programming) and CLP-agents (Constraint Logic Programming). The iterative algorithm for solving the model under dynamic emergence of new orders is also included. This structure enables managers to ask all kinds of questions.

In addition, the paper proposes new functionalities based on the CLP environment and numerical experiments for illustrative examples.

The aim of the paper is to present a simulation model of transportation which includes different disturbances. The paper describes in detail modeled route and way of modeling some disturbances. In article authors presented influence of distinguished distortion to time of realization transport task. Authors create own object with using Discrete Event Simulation and Agent Based Simulation approach. The research highlights of the performed works are as follows: showing influence of disturbances to transportation time and present a description to build objects, which represent disturbances.

This paper is concerned with designing a multi-agent approach for evaluating new products and selecting product portfolio. In today’s companies it is widespread to execute many new product projects simultaneously. As these projects require resources that are available in the limited quantities, there is the need to select the most promising set of new product for development. The evaluation of new product projects involves many agents that analyse the customer requirements and information acquired from an enterprise system, including the fields of sales and marketing, research and development, and production. The company’s resources, performance metrics, and the identified relationships are stored in knowledge base that is specified according to the framework of constraint satisfaction problem. The relationships are sought with the use of fuzzy neural system and described in the form of if-then rules.

The paper presents the results of research performed in the area of modeling and simulation of warehouse operations in a supply chain. Author defines in details main problems concerning warehousing. “Old” and “new” approaches to modeling and simulating are compared based on available simulation programs on the market. Main goal of the paper is to answer the question how to prepare in a very short time the motion of forklifts in a defined simulation model based on picking lists. Author proposes a method based on mixing DES (Discrete Event) and ABS (Agent Based Simulation). As result the library of agents objects (LogABS) is described and explained.

The lack of growth in Europe is one of the reasons behind the dire economic situation facing many European countries today. This economic situation can also be seen as the cause of political turmoil in many countries and the consequences could be radical as in Greece with the success of a far left party. The social situation in Europe is equally worrying and enterprises have been severely impacted. They have to resist the crisis and prepare for the future, hoping to find effective solutions very quickly to enable them to reorganize and be more competitive during and after the crisis. The research of ICAM (School of Engineering) in industrial organization aims to satisfy the needs of enterprises, public and private organisms, establishments and departments. It is based on GRAI Methodology, one of the three main methodologies for enterprise modeling. GRAIMOD is a software tool being developed by ICAM for supporting GRAI Methodology. A general typology is proposed for enterprises and general public entities, then for facilitating the improvement during the design phase a reference model is proposed for each enterprise, public establishment or department domain. This paper presents the latest concepts of GRAIMOD. The combination of CBR reasoning with multi-agent systems is also presented. An example is given to illustrate how to use concepts for developing the tool and improving enterprise performance.

This work presents a new agent based simulation tool specifically designed to study ecological networks. It includes many unique features like genetics, evolution, space-explicit simulation domain, flexible environmental modeling, etc. Written in C++, it yields a high performance experience and allows extremely large and complex simulations to be run, with up to hundreds of thousands of individuals moving and interacting in different ways to feed, reproduce or attack each other. It can be used to study ecosystems at different scales, from microscopic to superior animals whether alive or extinct.

This work takes place in the framework of GEODIFF project (funded by CNRS) and deals with the general issue of the social behavior modeling of pest insects with a particular focus on Bark Beetles. Bark Beetles are responsible for pine trees devastation in North America since 2005. In order to stem the problem and to apply an adapted strategy, one should be able to predict the evolution of the population of Bark Beetles. More precisely, a model taking into account a given population of insects (a colony) interacting with its environment, the forest ecosystem, would be very helpful. In a previous work, we aimed to model diffusive phenomenons across the environment using a simple reactive Multi-agent System. Bark beetle use pheromones as a support for recruitment of other bark beetles in the neighborhood in order to achieve a mass attack over a tree. They are first attracted by the ethanol or other phytopheromones emitted by a sick, stressed or dead tree and reinforce the presence of other individuals amongst the targeted tree. Both ethanol and semiochemicals are transported through the forest thanks to the wind, thermic effects and this advection phenomenon is modulated by the topology of the environment, tree and other obstacles distribution. In other words, the environment is involved in the process of a bark beetle attack. The first modeling we used to tackle our objective was not spatially explicit as long as free space propagation only was taken into account (isotropic phenomenon) with no constraint imposed by the environment such as wind. This article is intended to take into account such physical phenomenons and push the modeling one step further by providing predictions driven by measures provided by a Geographical Information System.

We provide an overview of agent-based and network-based computational models in archaeology. Then we suggest a sample model of gradual spatial dispersion of late Iron Age settlement network regarding the probable existence of central sites and settlement hierarchies. The model is based on archaeological research hypotheses and fragmented archaeological evidence of sites in Central Europe. The aim of the model is to enable experimenting with relevant combinations of parameters and triggers and to provide the dynamic picture of the emergence of the prehistoric settlement network.

In this paper we remarked the relevance of using of privacy-protection measures in trust models, since them imply conscious exchanges of opinions about third parties. Additionally to the already published description of the extra message exchanges that would protect privacy in a agent-mediated trust domain as ART testbed, we evaluate (through estimated weights) the communication and computation cost of these additional message exchanges at the level of their interactions. Although the effects of introducing them on the accuracy of trust valuations could also be considered in future works, in this contribution we focus our attention in an approach based on the weight of the information brought by an interaction.

This paper presents a practical application of Sample Selection techniques to model the process of selecting the next system response of a conversational agent. Our proposal deals with the important problem of imbalanced training data that is usually present in the selected application domain. This process is modeled as a classification task that takes the dialog history as input, and selects the next system response as output.

Our proposal improves the classifier’s performance

by automatically selecting examples that are difficult to classify during the training phase, considering the criteria of proximity to the border and the typicality of the examples. We present a practical application of this technique for a conversational agent providing railway information.

Simulation results support the usefulness of the proposed approach to provide the better selection of the responses of the conversational agent.

New technologies have become an important support for the monitoring of older people in outdoor environments by their caregivers. Smart phones equipped with a rich set of powerful sensors allowed the ubiquitous human activity recognition on mobile platforms at a low cost. Ambient Intelligence (AmI) is an emergent area that provides useful mechanisms that allows tracking elderly people through opportunistic sensoring using smartphone devices. This paper aims to show the second version of

SafeRoute

, an AmI system that fusions geo-localization sensors data embedded in smartphone devices for the monitoring of elderly people. This version improves functionalities of the previous one with the inclusion of new ones in the two components of this system: the Android OS application

CareofMe

and the web system

SafeRoute.

The proposed system merges localization data from GPS and Wifi sensors data in Android OS and includes the use of GoogleMaps functionalities in Android OS and web environments for provide alerts for caregivers.

This paper presents a new thresholding methodology for complex background images with an application to the courtesy amount of Brazilian bank checks. Courtesy amount images present a complex background and the proposal of an automatic thresholding process brings benefits to other steps in bank check clearance, such as the Optical Character Recognition (OCR). Experimental results showed that the proposed methodology yields good results, with average accuracy over 95 %, superior to standard methods from the literature.

The advances in Information and Communication Technologies (ICT) permits interactions among the computational and the physical elements of the smart-grid, and provide opportunities for novel energy management techniques allowing thus renewable energy integration and energy price minimization. Few of the current energy management schemes integrate storage aspects. In this paper, we propose an agent-based algorithm for better energy management in the smart-grid using a storage system. Furthermore, we propose a negotiation algorithm to help consumer choose the appropriate producer which provides him the needed energy at the lowest price. Simulation results show that our proposal minimizes the energy costs for each energy demand and reduces conventional energy utilization.

Finding an optimal coalition structure is a hard problem. In order to simplify this process, it is possible to explore some characteristics of the agents organization. In this paper we propose an algorithm that deals with a particular family of games in characteristic function, but is able to search in a much smaller space by considering organizational issues such as constraints in the number of participants. We apply this approach to the domain of smart grids, in which the aim is to form coalitions of electric vehicles in order to increase their reliability when supplying energy to the grid.

The programs and actions to rationalize energy consumption and increase energy efficiency, such as demand-side management (DSM) and demand response (DR), are receiving increasing attention. DSM involves the selection, planning, and implementation of measures intended to have an influence on the demand or customer-side of the electric meter. DR includes programs designed to encourage end-users to make short-term reductions in energy demand in response to price signals from the market or triggers initiated by electricity grid operators. This paper aims at studying DSM actions and DR programs to improve electricity efficiency in a public library (large building), taking into account a time of use (TOU) tariff. It presents a model of the library developed with the software DesignBuilder and using real data. It also proposes DSM actions to improve efficiency and reduce energy costs. Specifically, the analysis of the building equipment and their usage constraints lead to the development of load shifting and peak clipping mechanisms. Regarding DR programs, the library manager adopts a load direct control program proposed by a retailer and involving a reduction in consumption in the peak period of the day. The agents negotiate the terms and conditions of a bilateral contract, notably energy prices and contract duration. The DSM actions, the DR program and the negotiated rate allow the library manager to reduce consumption by 4 % and lead to savings of about 7 % in total annual costs.

Electricity markets are systems for effecting the purchase and sale of electricity using supply and demand to set energy prices. Pool prices tend to change quickly and variations are usually highly unpredictable. Bilateral contracts allow market participants to set the terms and conditions of agreements independent of a market operator. This paper describes on-going work that uses the potential of agent-based technology to help addressing several important issues related to market models. Specifically, the paper is devoted to risk management in bilateral contracting of electricity. Two agents interact and trade according to the rules of an alternating offers protocol. The paper focuses on both risk attitude and risk asymmetry and how they can influence price negotiation. In particular, it describes the trading process, introduces strategies that model typical patterns of concessions, and presents several concession tactics. The article also presents a case study on forward bilateral contracting involving risk management: a producer agent and a retailer agent negotiate a three-rate tariff.

With the recent liberalization of electricity markets, market players need to decide whether to and how to participate in each electricity market type that is available to them. The search for the best opportunities to sell or buy the required energy is, however, not an easy task. Moreover, the changes that electricity markets are constantly suffering make this an highly dynamic environment, with huge associated unpredictability. Decision support tools become, therefore, essential for market players to be able to take the best advantage from market participation. This paper proposes a methodology to estimate the expected prices of bilateral contracts based on the analysis of contracts’ historic log. The proposed method is based on the application of a clustering methodology that groups the historic contracts according to their prices’ similarity. The optimal number of groups is automatically calculated taking into account the preference for the balance between the estimation error and the number of groups. The centroids of each cluster are used to define a dynamic fuzzy variable that approximates the tendency of contracts’ history. The resulting fuzzy variable allows estimating expected prices for contracts instantaneously and approximating missing values in the historic contracts log.

The consensus behind Smart Grids (SG) as one of the most promising solutions for the massive integration of renewable energy sources in power systems has led to the practical implementation of several prototypes and pilots that aim at testing and validating SG methodologies. The urgent need to accommodate such resources of distributed and intermittent nature and the impact that a deficient management of energy sources has on the global population require that alternative solutions are experimented. This paper presents a multi-agent based SG simulation platform that is connected to physical resources, so that realistic scenarios with palpable influence on real resources can be simulated. The SG simulator is also connected to the Multi-Agent Simulator of Competitive Electricity Markets (MASCEM), which provides a solid framework for the simulation of restructured electricity markets. Taking advantage on the complementarities between the simulators, a SG market is proposed, and a realistic simulation scenario, using two real buildings acting in a simulated SG is presented.

This paper presents an e-learning content selection model, based on multi-agent paradigm, aiming to facilitate the learning material reuse and adaptability on Learning Management Systems. The proposed model was developed according to a BDI multi-agent architecture, as an improvement of the Intelligent Learning Objects approach, allowing the dynamic selection of Learning Objects. A prototype was implemented to validate the proposed model, using the JADEX BDI V3 platform, and allowing to build improved learning experiences.

The present article has the objective to show some preliminary results of new Web interface developed to multi-agent system named Heraclito. The Heraclito is one Intelligent Tutoring System (ITS) who has as objective teach propositional logic to undergraduate students of Computer. This version was developed with Test Driven Development (TDD) technique. This approach allowed us to create an interface more friendly and, the access by Web Browser, turned the tool adaptive, may now been available to work on computers, tablets, smartphones and also on Smart TVs.

Regarding the educational contexts based on e-learning, Learning Objects (LOs) have arisen as a new conceptual model to organize the content. Thus, it is necessary to analyze the potential impact of LOs on knowledge appropriation processes using the quality concept. In this case, quality is understood as the level of relevance of the educational resources in the teaching-learning process, associated to educational goal and other characteristics. The level of significance can be determined by evaluating the characteristics of the object by a group of experts; however, this process is not an easy task because different criteria should be considered, such as teaching, interoperability, scalability and reusability. On the other hand, a factor to solve is the selection of experts. This paper presents different dimensions and metrics to evaluate, and an automatic mechanism for the correct selection of the experts. The validation is done through the development a multi-agent system.

The context-awareness concept, which is inherent to humans when performing any learning activity, becomes the main component for monitoring activities in virtual learning environments. The aim of this paper is to incorporate context-awareness services within an adaptive ubiquitous Multi-Agent System (U-MAS) learning environment intended for instructional planning and educational resource recommendation. The awareness agent developed that composes the system architecture provides the functionality that involves several context-awareness services. This characteristic allows both students and teachers at a given time be aware of their learning progress status during U-MAS execution. In order to validate the incorporation of context-aware services a prototype was built and tested through a case study. Results obtained demonstrate the effectiveness of using this kind of approaches in virtual learning environments which constitutes an attempt to improve learning processes.

The project described in this paper proposes a model in which, through a videogame, players will become agents of change in a virtual world considering a Colombian post-conflict scenario. First we briefly describe a sample of games with similar approaches and later present the proposal making emphasis in explaining how the interaction between players is made, as well as the use of animated agents to provide narrative aspects. We also focus in one of the most important aspects of our proposal: a mechanism through which individual actions of each player, represented by an avatar, contribute to a collective construction. The metaphor used here is that, in the same way that players are able to collectively achieve the restauration of certain damaged zones within the virtual world, they would be able to contribute to peace in a Colombian post-conflict scenario.

During the last decade, learning management systems (LMS) have become a useful tool for students and teachers seeking to assist virtual learning processes. The aim of this paper is to present a proposal for the design and integration of a learning companion agent within the Claroline LMS intended to foster collaboration, assistance, support, and motivation of students. Using a role-based design method a learning companion agent was built having three main roles: Competitor, Fault Detector, and Collaborator. Through these roles the platform provides student-centered responses, educational resources, and the possibility to interact within a healthy competition learning environment. The companion agent was implemented and validated through a case study that includes a test virtual course and assessments for several registered students. It can be concluded that using a companion agent students become aware of their learning failures and overall performance of the virtual course.

Learning Object (LO) is a content unit being used within virtual learning environments, which -once found and retrieved- may assist students in the learning process. Such LO search and retrieval are recently supported and enhanced by data mining techniques. In this sense, clustering can be used to find groups holding similar LOs so that from obtained groups, knowledge-based recommender systems (KRS) can recommend more adapted and relevant LOs. In particular, prior knowledge come from LOs previously selected, liked and ranked by the student to whom the recommendation will be performed. In this paper, we present a KRS for LOs, which uses a conventional clustering technique, namely K-means, aimed at finding similar LOs and delivering resources adapted to a specific student. Obtained promising results show that proposed KRS is able to both retrieve relevant LO and improve the recommendation precision.

Research on agent interaction has attracted the attention of agent researchers for a long time. This paper revisits an old work to find uses of an argumentation protocol which borrows from social sciences to raise agreements among several agents. In this protocol, several agents discuss and provide a conclusion after some rounds of mutual information exchange. The work is reusing INGENIAS methodology to model and deploy the examples and illustrate benefits of this protocol proposal. It also analyses the requirements of this protocol for its application.

The purpose of this paper is to present the applicability of the

$$JaCalIVE$$

framework for developing human-agent societies. This kind of applications are those where virtual agents and humans coexist and interact transparently into a fully integrated environment. Specifically, the paper presents an ambient intelligence application where humans are immersed into a system that extracts and analyzes the emotional state of a human group. This social emotion is employed in order to try to maximize the welfare of that humans by playing the most appropriate music in every moment.

The current growth of information and communication technologies has promoted the development of tools in order to facilitate the process of e-recruitment; benefiting both recruiters as jobseekers. This paper presents a case-based Multi-Agent System which aims at integrating an ontology in order to select and to recommend adapted jobseekers to the recruiter job postings or vice versa. For this reason, the ontology considers the HR-XML standard for map-ping CVs in order to standardize the knowledge representation. The MAS was designed following the Prometheus Methodology and then a prototype has been implemented. A case study was performed within a testing phase in order to validate our work. As a result of this phase, we can prove the effectiveness of using this kind of technologies in the e-recruitment process.

A reflection about Roboethics and its declinations has been conducted starting from the analysis of the robot semantic and the cultural perception that has arisen towards these machines. The analysis faces also the meanings, the technological limits and the expectations about Robotics today, laying the foundations to define a design approach that put the man at the centre of the project, with its community and the context. As a case study is introduced Virgil, example of museum robotic activity carried out in the spirit of ethic design for a specific Cultural Heritage, which consist in the Savoia’s Royal Residences, in Piedmont, Italy.

Supporting group decision-making in ubiquitous contexts is a complex task that needs to deal with a large amount of factors to be successful. Here we propose an approach for a negotiation model to support the group decision-making process specially designed for ubiquitous contexts. We propose a new look into this problematic, considering and defining strategies to deal with important points such as the type of attributes in the multi-criteria problem and agents’ reasoning. Our model uses a social networking logic due to the type of communication employed by the agents as well as to the type of relationships they build as the interactions occur. Our approach intends to support the ubiquitous group decision-making process in a similar way to the real process, which simultaneously preserves the amount and quality of intelligence generated in face-to-face meetings and is adapted to be used in a ubiquitous context.