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

Nephrolithiasis disease is commonly related with the low activity performance, i.e., the regular performance of physical activity can reduce the risk of kidney stones. Sensors available in off-the-shelf mobile devices may handle the control and recognition of the activities performed, including the energy expenditure and their identification. This paper identifies the common values that should be measured during the treatment of this disease, including water consumption (with regular registration), daily calories intake (defined by a professional) and urinary pH (measured with test strips), which may be combined with the measurement of the energy expenditure and the activities performed. As the treatment and prevention of the Nephrolithiasis disease includes the performance of hard physical activity and the regular trip to the toilet, where this identification provides a control of the evolution of the treatment. The combination of these concepts and the use of the technology may increase the control and speed of the treatment.

Home Health Care (HHC) services are growing worldwide and, usually, the home care visits are manually planned, being a time and effort consuming task that leads to a non optimized solution. The use of some optimization techniques can significantly improve the quality of the scheduling solutions, but lacks the achievement of solutions that face the fast reaction to condition changes. In such stochastic and very volatile environments, the fast re-scheduling is crucial to maintain the system in operation. Taking advantage of the inherent distributed and intelligent characteristics of Multi-agent Systems (MAS), this paper introduces a methodology that combines the optimization features provided by centralized scheduling algorithms, e.g. genetic algorithms, with the responsiveness features provided by MAS solutions. The proposed approach was codified in Matlab and NetLogo and applied to a real-world HHC case study. The experimental results showed a significant improvement in the quality of scheduling solutions, as well as in the responsiveness to achieve those solutions.

Human support is thought to increase adherence to internet-based interventions for common mental health disorders, but can be costly and reduce treatment accessibility. Embodied virtual agents may be used to deliver automated support, but while many solutions have been shown to be feasible, there is still little controlled research that empirically validates their clinical effectiveness in this context. This study uses a controlled and randomized paradigm to investigate whether feedback from an embodied virtual agent can increase adherence. In a three-week ecological momentary assessment smartphone study, 68 participants were asked to report their mood three times a day. An embodied virtual agent could mirror participant-reported mood states when thanking them for their answers. A two-stage randomization into a text and personalized visual feedback group, versus a text-only control group, was applied to control for individual differences (study onset) and feedback history (after two weeks). Results indicate that while personalized visual feedback did not increase adherence, it did manage to keep adherence constant over a three-week period, whereas fluctuations in adherence could be observed in the text-only control group. Although this was a pilot study, and its results should be interpreted with some caution, this paper shows how virtual agent feedback may have a stabilizing effect on adherence, how controlled experiments on the relationship between virtual agent support and clinically relevant measures such as adherence can be conducted, and how results may be analyzed.

Nowadays, recommender systems are a key tool in sectors such as online sales, video playback and music on demand or book recommendation systems. This paper proposes a personalized route recommendation system for users of electric vehicles, specifically for e-bike users. Around the world e-bikes have become a real alternative to other motorized modes of transport and they are used for daily commuting. A multi-agent system is used to manage the information produced by the system, which generates route recommendations for users based on the routes they had travelled previously. Recommendations are provided to users through a smart-phone application, which is in charge of registering the data on the routes users travel.

The paper proposes multi-agent technology and a prototype system with together-acting UAVs for joint survey missions. The prototype makes it possible to connect UAVs in a united swarm, proposes coordinated flight plans and adaptively re-configures plans due to disruptive events. The approach to organization of program agents within a prototype subsystem is described. A series of simulation experiments and several flight tests were conducted to evaluate the effectiveness of the distributed scheduling mechanism. The aim of the current and future developments is creation of complex solutions for coordinated management of UAVs for precise agriculture.

In this paper we propose an airline marketplace, modeled as a multi-agent system with an automated negotiation mechanism, where airlines can announce availability of resources (aircraft or aircraft and crew) for lease and other airlines can go there to contract resources to fill gaps in the operation, typically due to disruptions and/or an unexpected increase on the operation. The proposed negotiation occurs in several rounds, where qualitative comments made by the buyer agent on proposals sent by the sellers enables these to learn how to calculate new proposals, using a case-based reasoning methodology.

This paper presents the optimization algorithm Artificial Bee Colony (ABC) firstly introduced by in 2005 and proposed for optimizing numerical problems. ABC is the swarm-based meta-heuristic algorithm inspired by intelligent behavior of honey bee colonies. In this paper, ABC has been applied on solving the flight disruption problem, by swapping aircraft and/or cancelling/delaying flights, and its performance has been shown through experimentation. The environment and data for experiments are provided by MASDIMA, Multi-Agent System for DIsruption MAnagement developed by LIACC (Laboratory of Artificial Intelligence and Computer Science).

This paper presents a Virtual Environment Mapping (VEM) module assembled in an indoor drone in order to be used by creative industries. This module is in charge of allowing users to capture the environment where the final recording will take place. Having a virtual representation of the environment allows both the photography director and the director to test different camera trajectories, points of view, speeds and camera configuration without the need to be physically in the recording set. The digitization of the scene will be performed with a 3D camera on-board of the drone. This paper discusses the overall VEM architecture, taking into account the requirements it has to fulfil. It will also present a working demo of the system, with the communication infrastructure in place and with a proof of concept of the main components of system.

We study the impact of the network topology on various market parameters (volatility, liquidity and efficiency) when three populations or artificial trades interact (Noise, Informed and Social Traders). We show, using an agent-based set of simulations that choosing a Regular, a Erdös-Rényi or a scale free network and locating on each vertex one Noise, Informed or Social Trader, substantially modifies the dynamics of the market. The overall level of volatility, the liquidity and the resulting efficiency are impacted by this initial choice in various ways which also depends upon the proportion of Informed vs. Noise Traders.

The economic impact of fraud is wide and fraud can be a critical problem when the prevention procedures are not robust. In this paper we create a model to detect fraudulent transactions, and then use a classification algorithm to assess if the agent is fraud prone or not. The model (BOND) is based on the analytics of an economic network of agents of three types: individuals, businesses and financial intermediaries. From the dataset of transactions, a sliding window of rows previously aggregated per agent has been used and machine learning (classification) algorithms have been applied. Results show that it is possible to predict the behavior of agents, based on previous transactions.

This paper proposes a model based on particle swarm optimization to aid electricity markets players in the selection of the best player(s) to trade with, to maximize their bilateral contracts outcome. This approach is integrated in a Decision Support System (DSS) for the pre-negotiation of bilateral contracts, which provides a missing feature in the state-of-art, the possible opponents analysis. The DSS determines the best action of all the actions that the supported player can take, by applying a game theory approach. However, the analysis of all actions can easily become very time-consuming in large negotiation scenarios. The proposed approach aims to provide the DSS with an alternative method with the capability of reducing the execution time while keeping the results quality as much as possible. Both approaches are tested in a realistic case study where the supported player could take almost half a million different actions. The results show that the proposed methodology is able to provide optimal and near-optimal solutions with an huge execution time reduction.

The negotiation is one of the most important phase of the process of buying and selling energy in electricity markets. Buyers and sellers know about their own trading behavior or the quality of their products. However, they can also gather data directly or indirectly from them through the exchange information before or during negotiation, even negotiators should also gather information about past behavior of the other parties, such as their trustworthiness and reputation. Hence, in this scope, reputation models play a more important role in decision-making process in the undertaken bilateral negotiation. Since the decision takes into account, not only the potential economic gain for supported player, but also the reliability of the contracts. Therefore, the reputation component represents the level of confidence that the supported player can have on the opponent’s service, i.e. in this case, the level of assurance that the opponent will fulfil the conditions established in the contract. This paper proposes a reputation computational model, included in DECON, a decision support system for bilateral contract negotiation, in order to enhance the decision-making process regarding the choice of the most suitable negotiation parties.

The paper presents the results of research carried out in recent years in the area of modeling and simulation of production and assembly systems. The main goal of the article is to show the practical application of ABS in the simulation modeling of production and assembly systems as well as the benefits resulting from it. The most important points of the article are the presentation of the originality of this approach in the context of traditional methodologies for building simulation models of manufacturing systems, the discussion of solutions proposed in commercial simulation programs, suggesting a language describing the behavior of agents carrying out the flow of materials in the production process and showing the model built in the described way.

Industrial processes are facing major changes with the arrival of a new revolution: Industry 4.0. By introducing blockchain technology on this environment, conditions are met to accelerate and improve the concepts associated with this new revolution. By looking at industries as an intelligent ambient, where there is a big amount of data being exchanged and created, is possible to gather data and create knowledge about the interactions, and other entities. In this work we propose a model that uses blockchain and multi-agent systems to help represent an entity in a network of entities and help the decision-making process by providing additional knowledge.

Agents technology is popular in many different scientific area. One of them is logistics and supply chain. Agent is characterized by many features, as: mobility and intelligence, autonomy, the ability to monitor the environment at all times. Authors in this article focus on aspect of agents in logistics and supply chain. In this area 3 types of agents can be distinguished: searching agents, monitoring agents and managing agents. Describing the agents topic author decided to present basic information about Discrete-Event Simulation (DES) and Agent Based Simulation (ABS), which are used in logistics also. In this article are presented examples of using each mentioned method of simulation in logistics area.

Nowadays, electronics can be found in almost every available device. At the core of electronic devices there are Printed Circuit Boards (PCB). To create a suitable PCB there is the need of complying with several constraints, both concerning electrical and layout design. Thus, the design rules related to the PCB manufacturing and assembly are very important since these restrictions are fundamental to ensure the creation of a viable physical PCB. Electrical Computer Aided Design (ECAD) tools are able to automatically verify such rules, but they only consider a subset of the total required rules. The remaining rules are currently manually checked, which may increase the occurrence of errors and, consequently, increase the overall costs in designing and in the manufacturing process of a PCB. Being the design a crucial phase in the manufacturing procedure, a software system that automatically verifies all design rules and produce the corresponding assessment report is fundamental. Such software system is addressed in this paper.

We build an agent-based system for supporting correlation analysis between human behavioural and non-behavioural patterns. A novel social norm specification language is leveraged to create an interaction model based communication engine for choreographing distributed systems, offering a communication environment for multiple interacting players. Categorising sets of players based on their interaction behaviours allows labelling the other patterns, which the system uses to further its understanding relationship between the two traits. While existing analysis methods are manually applied, non-user-editable and typically opaque, the system offers an end-to-end computing framework and protocols which are modifiable for specific users. Evaluation for this system relies on tests for categories of people who are mentally depressed, where traditional questionnaire-based methods are superseded by methods that use more objective behavioural tests. This approach to evaluation through behavioural experimentation is intended not only to classify sub-types of depression cases which would facilitate elucidation of aetiology but evaluates system performance in a real-world scenario.

This paper describes several steps towards the development of an online agent (or socialbot) that provides emotional support to stressed users. In particular, we present an empirical study that was conducted with the aim to investigate how people help each other to cope with stressful situations via online social networks. To this end, around 10.000 tweets about stressful situations were collected. Then, using crowdsourcing, these tweets were classified into stress categories, and supportive replies to them were collected, which were also classified into categories. Contingency tables were constructed in order to explore which types of support were most frequently used in which circumstances. The resulting values can be used as parameters for a previously developed algorithm that automatically constructs support messages. This allows our agent to generate supportive messages that are more similar to the support messages that human beings send via social media.

We simulate the impact of influencers in the adoption of digital multi-sided platforms. We consider four metrics to identify influencers: degree, betweenness, closeness and page rank, and we test how they shape the adoption, prices, and profits of digital multi-sided platforms using an agent-based model. We simulate the market adoption with and without those influencers. We find that adoption is lower and grows slower without influencers. This result is also valid even when one side has influencers, but the other one has not. Depending on the network we assume, the role of influencers is different. In some cases, the launching fails, in others, it is slower only. We also find that prices are sensitive to influencers. However, the effect on prices depends on which centrality measure we consider. Companies use prices as a tool to counterbalance the influence of influencers in profits. Lastly, we show that profits are very sensitive to influencers, without them, profits are lower.

It is possible to detect the presence of residents in a home by monitoring its energy consumption. Currently, the state of the art provides us with a number of approaches. Some studies leverage intrusive systems which require user interaction. Others employ sensors to detect the presence of people in a non-intrusive way. In this article, we propose the use of a sensor network for measuring electric energy consumption in a home. A multi-agent system is used to manage the data generated by the deployed sensor network in an intelligent way. A non-intrusive occupation monitoring algorithm was designed to determine when a house is occupied and when it is empty.

This research explores information fusion and data mining techniques and proposes a methodology to improve predictions based on strong associations among agricultural products, which allows prediction for future consumption in local markets in the Andean region of Ecuador using spatial prediction techniques. This commercial activity is performed using Alternative Marketing Circuits (CIALCO), seeking to establish a direct relationship between producer and consumer prices, and promote buying and selling among family groups.

Multi-agents are being increasingly used in many areas, especially in Health Systems. In a game the agents’ paradigm provides more autonomy, intelligence and pro-activity. In this context, Multi-agents systems can be used to control the user performance, by adapting the interface to the difficulty tasks level. This paper aims at describing the development process of SMEC-3D, a cognitive stimulation game that integrates Virtual Reality and Multi-agent technologies. The SMEC-3D modeling process used i* (i-star) framework to model the goals, agents, domain, plans and tasks. The game objective is to improve attention and memory of patients with neuropsychiatric disorders. The paper describes the development process specially the agents’ methodology. The resulted SMEC-3D game showed that the combination of tools and programming languages applied to this experiment worked efficiently.

This research explores the capacity of Machine Learning techniques to detect anomalies and how incorporate this capacity to thinger.io platform. Thinger.io is a IoT opensource platform that allows to create an IoT environment using any hardware available on market. In this paper, several ML techniques are proposed to detect anomalies in the platform.

The Ambient Intelligence Development Environment (AIDE) tool allows developers of Ambient Assisted Living (AAL) solutions to test their projects in a virtual environment. The simulation reduces the onus on systems development in AAL, proving to be an adequate tool for the adoption of agile methodologies in AAL projects. This work presents the implementation of an extension of AIDE that enables designers to test their solutions with the end user through the virtual reality environment. AIDE-VR extends the simulations of the first tool to be used with the end users of the product in an immersive experience through the use of virtual reality.

This paper proposes a multi-agent approach for power grid load shedding programming. Known as the most complex machine ever made by man, power grid is an essential pillar of all national economies. Its complexity and size make it vulnerable. Load shedding is usually an emergency control process against electrical networks with low production capacity. In this study, a system named MASLA, a Multi Agent System based Load Shedding Algorithm, which explores the load shedding planning is proposed. In electrical distribution system, power is delivered to customers through feeders which are a combination of electrical lines and medium voltage transformer substations. Depending on its power, a feeder may serve very large number of customers of different types (industrial, residential, …). Hence, Feeder Agent and Load Agent are defined in a power grid and a negotiation takes place between them, considering customer’s level of priority to select which one will be significantly impacted in a possible emergency outage. The model is implemented using a multi agent platform and a case study using the Senegalese power grid revealed that the proposed approach is useful and feasible for companies facing frequent disruption of electricity supply.

The concept of distributed generation and renewable energy has increased the need for using Smartgrids which are electric micro-grids having intelligent characteristics that provide autonomy to the system not only to improve their operation but also to make it easier for users their management. The aim of this paper is to propose a fuzzy-based Multi-Agent Model to control a micro-grid by determining optimal operation states based on real-time process conditions and energy market dynamics. The Prometheus methodology is used for the MAS architecture design and development. The implementation of the system is carried out using Java, the JADE framework and the JFuzzyLogic library. Based on the proposed fuzzy MAS model, a prototype was implemented and validated through a case study. Results obtained demonstrate the effectiveness of this approach to automatically manage the states of a micro-grid when connected to external grid in a dynamic energy market environment. It is also possible to extend this application for different micro-grid applications involving other power generation, storage, and consumption capabilities.

The key mechanisms for purchasing and selling electrical energy include electricity pools and bilateral contracts. This article is devoted to bilateral contracting, which is modeled as a negotiation process involving an iterative exchange of offers and counter-offers. It focuses on coalitions of end-use consumers and describes a case study involving five schools for children located in England. The schools decide to ally into a coalition to strengthen their bargaining positions and, hopefully, to obtain better tariffs. To this end, they rely on a coordinator agent, who is defined from the group of five schools, by selecting either the “most powerful” school or the “best negotiator” school. The coordinator takes decisions according to either a “majority” rule, a “consensus” rule, or an “unanimity” rule. The simulations are performed with an agent-based system, called MATREM (for Multi-agent TRading in Electricity Markets). Although preliminary, the results suggest that coalition formation and management is beneficial to end-use customers, since the price agreed in the new forward contracts is more favorable to these agents, particularly when the coordinator is the “best negotiator” agent and considers the “unanimity” decision rule.

Micro-grids are known as a means of localization of renewable energy production and consumption. However, due to the intermittent nature of renewable energy sources, one of the main challenges in Micro-grid energy control and management is to handle any deviation from the prior forecasted power generation/consumption. Our proposed distributed multi-agent algorithm tries to handle power imbalance situations in a PV-based grid connected Micro-grid through optimizing a combination of storage usage, load curtailment, and main grid power purchase. In this model, the users’ consumption preferences are considered as an important factor in the decision making. We first devise a community of consumers with various energy usage preferences and then investigate the performance of our proposed algorithm over multiple scenarios having different users’ reactions to the energy conservation requests. The results obtained show the convergence and feasibility of the proposed algorithm. Moreover, the cost of imbalance handling is considerably reduced, preserving the level of satisfaction in the community, as the inconvenience effect of load curtailment is compensated by paying back to the consumers.

Doorstep scams are scams, often happening at the front door, in which a con artist has a convincing, but fraudulent, story with the purpose of coming into your house and/or stealing money. Various campaigns to educate people exist, but they do not focus on the verbal skills people can use to prevent themselves from becoming a victim. This paper describes the conceptual design of a proposed training application. This application will provide an agent-based learning environment for high-risk doorstep scam victims. In order to create a training application, field research has been done to the content and progress of doorstep scams, which is used to create interactive scenarios.

The Semantic Web offers a structure to support generation of significant learning contents for Web based intelligent learning environments by using public knowledge bases known as public ontologies, available on the Web. Prior research has therefore been undertaken into allowing agents societies to navigate through these knowledge bases, in search of answers to queries. In addition, research is emerging into using this knowledge to contribute to the area of education, in terms of creating virtual learning environments. This work proposes a model for agents that allows access to ontologies related to a given domain of knowledge available on the Web, allowing these agents to use this knowledge in the construction and formulation of questions for the production of relevant and updated content for the student. Several efforts have been made to integrate agents with ontologies, which allow a greater knowledge for the agent based on a local ontology. However, no proposal has yet combined the ability to use the semantic data available on the Web in conjunction with a consolidated BDI agent framework for the production of meaningful content for virtual learning environments. Therefore, this work proposes a model for a virtual learning environment that uses agents developed using the Jason interpreter, with its ability to access ontologies available on the Web to update its belief base and generate significant content for the student. To validate this approach, a case study of an educational quiz is presented that uses this information to identify questions and check the answers obtained.

The selection of educational material that allows to address the diversity of students in the class-room is a problem faced by teachers in their daily work. The difficulty increases as you wish to attend to a greater number of traits in the trainees, for example, their predominant learning styles and their conditions of disability, to cite some cases. A valid alternative to include different profiles of students and to evaluate the type of materials available is to simulate students with different characteristics and evaluate the relevance of available educational resources. This article presents an interactive tool that supports the teacher in this task, simulating the behavior of students through agents, providing each one with different learning styles and special education needs related to any disability condition. The multi-agent system recommends learning objects (LO) stored in a repository, according to the preferences of agents, which were guide by rules defined according to their characteristics and associated with the metadata of the resources. The appropriate materials are chosen for all students selected by the teacher.

Monitoring classrooms using cameras is a non-invasive approach of digitizing students’ behaviour. Understanding students’ attention span and what type of behaviours may indicate a lack of attention is fundamental for understanding and consequently improving the dynamics of a lecture. Recent studies show useful information regarding classrooms and their students’ behaviour throughout the lecture. In this paper we start by presenting an overview about the state of the art on this topic, presenting what we consider to be the most robust and efficient Computer Vision techniques for monitoring classrooms. After the analysis of relevant state of the art, we propose an agent that is theoretically capable of tracking the students’ attention and output that data. The main goal of this paper is to contribute to the development of an autonomous agent able to provide information to both teachers and students and we present preliminary results on this topic. We believe this autonomous agent features the best solution for monitoring classrooms since it uses the most suited state of the art approaches for each individual role.

Nowadays, the need of Intelligent Transportation Systems software tools and services for Sustainable Transportation is urgent. This paper proposes an ontology specially tailored for Intelligent Transportation System characterization. The main features of the proposed ontology is the ability to incorporate sustainable variables when characterizing a transportation system, and the coverage of open fleet concepts together with its dynamic features. Moreover, it is enhanced to facilitate its integration with other intelligent components that in a wider and complete application tool can provide intelligent computation over the data specified with the proposed ontology.

Nowadays, the advance of the technology has allowed to develop applications and systems to facilitate the daily life of the people. One of the most used field by thousands of people every day is to generate routes to go from one place to another, obtaining not only the route according to the means of transport that the user selects, but also can get recommendations of places of interest that can be found along the way. Unfortunately, these applications do not consider the profile of the end users, and generate the same routes for a person who has a disability as for those who do not. In this article, we propose a model to create a recommendation system based on the user profile to generate automatic and personalized routes on foot or on public transportation for people with disabilities.

The integration of geolocation, big data and cognitive agents has become one of the most boosting business tools of the digital era. By definition, geolocation represents the use of different technologies in a variety of applications to help locate humans and objects. To really achieve smart services, companies also require accessing huge volumes of related information to draw meaningful conclusions. With big data, it is possible to establish connections between a wide range of associated information, and use it to improve available services or create new ones. Today, the influence of geolocation, cloud data science and involved cognitive agents impacts many application fields, which include: safety and security, marketing, beacon technology, geofencing, location-sensitive services, transportation and logistics, healthcare, urban governance, intelligent buildings and smart cities, intelligent transport systems, advanced driver assistance systems, and autonomous and semi-autonomous vehicles. To address these challenges, this paper presents a general associative-cognitive architecture framework to develop goal-oriented hybrid human-machine situation-awareness systems focused on the perception and comprehension of the elements of an environment and the estimation of their future state for decision-making activities. The architecture framework presented emphasizes the role of the associated reality as a novel cognitive agent and the involved semantic structures, to improve the capabilities of the corresponding system, processes and services. As a proof of concept, a particular situation awareness agent for geolocation of vehicles in tunnels is shown, that uses cloud data association, vision-based detection of traffic signs and landmarks, and semantic roadmaps.

Nowadays, transportation is a critical sector of our lives, not only for the movement of people, but also to be capable to move goods around the world. Although providing such services can be seen as a very tiny problem in our society, behind it, there is a complex sector that requires sophisticated models and specific software to analyse a vast amount of information coming from different sources in order to provide a sustainable and efficient service. Given such a complex field, various issues have come up like the search of optimised routes, efficient assignment of vehicles, reduction of gas emissions, cost optimization problems, etc. In most cases, the provided approaches are focused on addressing the optimization problem considering fleets with identical features. In this work, we present HVSRec, a heterogeneous fleet semantic recommender system that integrates mechanisms to manage vehicles of different nature and characteristics efficiently. The platform is aimed to connect customers that request a transportation of a certain good with drivers that are offering transportation services with their own vehicle.

Bike-sharing systems are becoming very popular in big cities. They provide a cheap and green mean of transportation used for commuting and leisure. Being a shared limited resource, it is common to reach imbalanced situations where some stations have either no bikes or only empty slots, thus decreasing the performance of the system. To solve such situations, trucks are typically used to move bikes among stations in order to reach a more homogeneous distribution. Recently, research works are focusing on a complementary action to reduce imbalances consisting in incentivizing users to take (or return) bikes from stations with many bikes rather than those with few bikes, e.g. by fare discounts. In this paper, we present simulator for analyzing bike-sharing systems. Several user generation distributions can be configured. The simulator is specifically designed with the aim of evaluating incentive-based rebalancing strategies. The paper describes in detail the characteristics and potential of the simulator, including several experiments.

Agent-based modeling and simulation are some powerful techniques that are widely used with success for analyzing complex and emergent phenomena in many research and application areas. Many different reasons are behind the success of such techniques, among which an important mention goes to the availability of a great variety of software tools, that ease the development of models, as well as the execution of simulations and the analysis of results. However, the agent models provided by such tools do not offer the features of the computational agents found in multi-agent systems or distributed artificial intelligence techniques. Therefore, it is difficult to use such tools to model complex systems defined by autonomous, proactive and social entities. This paper presents an actor software library, called ActoDeS, for the development of concurrent and distributed systems, and shows how it can be a suitable mean for building flexible and scalable ABMS applications.

Airport slot allocation is a combinatorial allocation problem involving different complex and autonomous systems. Nowadays, airport slots are allocated in a two-stage process: primary allocation is performed according to a set of administrative rules and for each airport independently, while secondary allocation is based on trading mechanisms. Several studies have raised inefficiencies in these processes. To enhance the airport slot allocation process we use an auction-based market. More specifically, we present an airport slot allocation mechanism based on a price-setting auction that has been implemented and evaluated by means of Agent-Based Modelling (ABM) and simulation techniques. The solutions obtained using our approach are compared and assessed with the ones obtained using linear programing, showing that market mechanisms can be an efficient alternative to the current administrative procedure.

Multi-agent systems are commonly used in various research fields such as artificial intelligence, operational research, simulation, biology, ... However, this diversity often requires that the system and agents in it are created from scratch for each new research project. In addition to the fact that this forces the developers to code similar elements anew each time, this can introduce non-negligible biases (e.g. an information accessible to every agent which shouldn’t be or a scheduler executing twice due to a user interface design failure). To avoid this, we propose in this paper AMAK, a framework developed in Java$$^\textsf {TM}$$ to facilitate the design and development of a multi-agent system. First, we present the particularity of Adaptive Multi-Agent Systems. Secondly, a state of the art of the main tools and software aiming at facilitating the development of such systems is discussed. Then, we develop the architecture of the framework and the main features. The use of the framework is illustrated with an application for socio-technical ambient systems. And finally, we conclude with the perspectives of this work.

Agent-based models (ABMs) need to populate a mega number of agents over a scalable simulation space in order to handle practical problems, (e.g., metropolitan traffic simulation and nationwide epidemic prediction). Although parallel and distributed simulation have steadily addressed their computational needs, non-computing scientists still tend to use GUI-rich, easy-to-use ABM interpretive platforms. This paper intends to identify the difficulty in using the current parallel ABM simulators and to propose their future improvements. For this purpose, we surveyed different ABM applications, modeled them as seven benchmark test cases, used them to analyze the agent descriptivity of parallel ABM simulators, and evaluated their execution performance affected by the current implementations.

The agent-based approach has been successfully used in the past years to model and simulate complex systems. We use this approach on a honeybee colony in a Dadant hive, where several tens of thousands of bees interact, in order to evaluate the impact of local actions at the bee-level (such as beekeeping practices) on the global system. In this article, we focus on the foraging activity, its recruitment mechanisms and the behaviour of foraging bees, and how these bees interact with the hive’s environment, greatly different in scale. We present a customizable, agent-compliant module called the Ecosystem Module, that aims at modelling and simulating the foraging, according to the local weather and the surrounding nectar sources. First results back up our model, showing that these recruitment mechanisms lead to a self-organizing process of the best available sources’ selection by the agents.

This paper introduces H-finder, a bio-inspired algorithm to build a decentralized and self-organized P2P information system in healthcare environment. The algorithm takes advantage of the properties of ant systems, in which a number of entities perform simple operations at the local level and an advanced form of “intelligence” emerges at global level. The work of ant-inspired agents is exploited to organize “metadata”, that is, documents containing healthcare information, like Electronic Health Records (EHR), in a network of clinical servers. Agents travel the P2P interconnections among servers and spatially sort metadata in order to organize similar metadata into neighbor hosts. Preliminary performance analysis proves that the proposed algorithm allows the healthcare information to be useful reorganized thus improving the data management.

We consider the problem of using only local communication to implement a distributed algorithm for large teams of mobile robots that searches space for locations of interest and distributes the robots across those locations according to quality. Toward this end, we take inspiration from insect societies that are able to coordinate without the use of pheromone trails. In particular, we focus on species that use only one-on-one local interactions to adaptively distribute scouts during nest-site selection tasks. Thus, there is a direct analogy between the insect communication mechanisms and peer-to-peer communication implementable in mobile, ad hoc networks of robots. Using chemical reaction networks as a conceptual bridge between behavioral descriptions from biology and event-triggered rules for robots, we develop a stochastic, biomimetic algorithm that achieves the desired goal. To validate our approach, we implement the algorithm on a large swarm of aerial, fixed-wing robots operating within the high-fidelity simulation package, SCRIMMAGE.

Bacterial colonies perform a cooperative and distributed exploration of the environmental resources. This paper describes how bacterial colony networks and their skills to search resources can be used as tools for mining association rules in static and stream data. The proposed algorithm is designed to maintain diverse solutions to the problems at hand, and its performance is compared to another well-known bacterial algorithm in both static and stream datasets.

Novelty search in evolutionary robotics measures a distance of potential novelty solutions to their k-nearest neighbors in the search space. This distance presents an additional objective to the fitness function, with which each individual in population is evaluated. In this study, the novelty search was applied within the differential evolution. The preliminary results on CEC-14 Benchmark function suite show its potential for using also in the future.

In this paper, we analyze the interplay of two robotic swarms applied to solve a target point search in a non-cooperative mode. In particular, we consider the case of two identical robotic swarms deployed within the same environment to perform dynamic exploration seeking for two different unknown target points. It is assumed that the environment is unknown and completely dark, so no vision sensors can be used. Our work is based on a robotic swarm approach recently reported in the literature. In that approach, the robotic units are driven by a popular swarm intelligence technique called bat algorithm. This technique is based on echolocation with ultrasounds, so it is particularly well suited for our problem. The paper discusses the main findings of our computational experiments through three illustrative videos of single executions.