Multi-agent Model Research Articles

Renewable energy effects on energy management based on demand response in microgrids environment

With further penetration of low-carbon energy conversion, microgrids (MGs) have become a necessary tool for expanding the consumption of renewable energies. In this paper, an optimal operation model for a microgrid-based multi-agent system is proposed. The goal is to save the total energy cost, which is expressed as a sum of locally observable convex functions. Therefore, improving the operational efficiency of microgrids is the key to promoting renewable energy development. This paper develops a three-layer multi-agent system model considering energy storage system and power thermal load demand response to solve the energy management problem of microgrids. In order to investigate the effect of energy storage system and demand response in microgrids, this paper designs three simulation cases, namely infrastructure case, energy storage case and demand response case. In order to prove the effectiveness of the proposed method, this paper uses the proposed method to solve three cases and compare the result with other meta-heuristic algorithms. The comparison results show that: (1) the multi-agent system model can realize the joint optimization of "resource, network, load and storage". (2) The introduction of energy storage and demand response system in microgrids can stabilize the output. Renewable energy units promote the use of renewable energy and reduce the overall operating cost of microgrids. Moreover, the results clearly demonstrate that the proposed algorithm has far better performance than other optimization methods. Also, the analysis obtained from the results shows that the cost is reduced by 1.82%. PV and WT output increased by 14.54% and 2.42%. In addition, their standard deviation decreases after ESS participation. The proposed approach is very effective through a simulation case study, which shows high potential for applications.

Combining MAS-GiG Model and Related Problems to Optimization in Emergency Evacuation

Emergency evacuation is of paramount importance in protecting human lives and property while enhancing the effectiveness and preparedness of organizations and management agencies in responding to emergencies. In this paper, we propose a method for evacuating passengers to safe places with the shortest possible evacuation time. The proposed method is based on a multi-level multi-agent MAS-GiG model combined with three problems. First, constructing a path map to select the shortest path; second, dividing the space of the experimental environment into smaller areas for efficient management, monitoring, and guiding evacuation; the third, adjusting the speed to handle collision issues and maintain distance between two or more groups of evacuees while moving. We extend our previous study by establishing groups based on the location of passengers and using a MAS-GiG model to guide evacuation. We compare the proposed method with our previous method to provide specific evaluations for the research and research in the future. We tested two methods in the departure hall, first floor, Danang International Airport, Vietnam.

Exploring agent-based chatbots: a systematic literature review

In the last decade, conversational agents have been developed and adopted in several application domains, including education, healthcare, finance, and tourism. Nevertheless, chatbots still need to address several limitations and challenges, especially regarding personalization, limited knowledge-sharing capabilities, multi-domain campaign support, real-time monitoring, or integration of chatbot communities. To cope with these limitations, many approaches based on multi-agent systems models and technologies have been proposed in the literature, opening new research directions in this context. To better understand the current panorama of the different chatbot technology solutions employing agent-based methods, this Systematic Literature Review investigates the different application domains, end-users, requirements, objectives, technology readiness levels, designs, strengths, limitations, and future challenges of the solutions found in this scope. The results of this review are intended to provide researchers, software engineers, and innovators with a complete overview of the current state of the art and a discussion of the open challenges.

Spatiotemporal transmission of infectious particles in environment: A case study of Covid-19

This study explores the dynamic transmission of infectious particles due to COVID-19 in the environment using a spatiotemporal epidemiological approach. We proposed a novel multi-agent model to simulate the spread of COVID-19 by considering several influencing factors. The model divides the population into susceptible and infected and analyzes the impact of different prevention and control measures, such as limiting the number of people and wearing masks on the spread of COVID-19. The findings suggest that reducing population density and wearing masks can significantly reduce the likelihood of virus transmission. Specifically, the research shows that if the population moves within a fixed range, almost everyone will eventually be infected within 1 h. When the population density is 50%, the infection rate is as high as 96%. If everyone does not wear a mask, nearly 72.33% of the people will be infected after 1 h. However, when people wear masks, the infection rate is consistently lower than when they do not wear masks. Even if only 25% of people wear masks, the infection rate with masks is 27.67% lower than without masks, which is strong evidence of the importance of wearing a mask. As people's daily activities are mostly carried out indoors, and many super-spreading events of the new crown epidemic also originated from indoor gatherings, the research on indoor epidemic prevention and control is essential. This study provides decision-making support for epidemic preventions and controls and the proposed methodology can be used in other regions and future epidemics.

Modelling the long-term consequences of implementing hormone-free reproductive management on the sustainability of a dairy sheep farm

Livestock production systems are strongly influenced by societal constraints which induce changes in their overall chain value organization. In dairy sheep farming, practices such as the male effect (ME) supported by new precision tools (heat detector) can be used to stop the use of hormonal treatments (HT) for heat induction and synchronization. However, the effectiveness of these “hormone-free” (HF) methods for managing ewes’ reproduction remains more uncertain than that of HT. Thus, introduction of HF practices in farms currently managed with HT could imply collateral impacts on the overall functioning of the production system and affect its long-term sustainability. In this work, a new multi-agent model (called REPRIN'OV) was designed to simulate consequences of introducing a HF reproduction management on a series of biotechnical, economic and environmental farm performance indicators, including farmer workload. Three reproduction management scenarios were simulated over five consecutive years for the case study of a French dairy sheep farm considered as conventional (i.e. using HT for reproduction management). They were defined as: use of HTs and insemination (HAI); no use of HTs while preserving insemination practice (HFAI); no use of HTs and direct mating only (i.e., no insemination; HFM). Results confirm that implementing HF reproductive management without accompanying it with further changes in farm management, could negatively impact several indicators of the farm sustainability in the medium and long-term. Indeed, the HF scenarios resulted in overall lower farm biotechnical performances (e.g., −4% for the fertility rate; P < 0.05) due to a more variable heat induction. An increase in farmer's workload during the lambing period (+34% and +33% of h/L of milk in HFAI and HFM, respectively; P < 0.05) was also observed, mainly due to more staggered lambing periods. As for the simulated environmental indicators, they were increased, mostly due to the reduction of the total milk production in HF’s (e.g., +6% of farm methane emissions per L of milk for both HFAI and HFM; P < 0.05). Nevertheless, positive outcomes were also simulated in the HFs as ewes lambing peaks were lower than in HAI suggesting, e.g., a possibility to improve ewes and lamb surveillance during lambing. HF scenarios also led to higher gross margin (GM) per litre of milk produced because no HT was used, reducing the production costs compare to HAI. In addition, implementing insemination in HFAI improved farm milk production and quality (i.e., fat and protein content), which resulted in improved total GM compared to HFM. Still, the reduction in the total milk production in HFs was too important compared with the improvement of the GM per litre, resulting in a reduced total GM in these scenarios (-4% and −5% in HFAI and HFM, respectively; P < 0.05). Overall, simulation results showed that to warrant sustainable introduction of HF strategies (including ME), new frameworks in the organization of key operations at the farm level should be explored (e.g., extending reproduction or milking periods length) and trade-offs should be discussed, considering all involved stakeholders of the dairy sheep industry and the chain’s value requirements. The REPRIN’OV model, may provide an effective tool to simulate the collateral impacts of establishing HF reproductive management and could be used as a decision support system to accompany the ambitious transition towards a HF dairy sheep industry in Europe and beyond.

Complex behavior of individuals and collectives in a social system: An introduction to exploratory computational experimental methodology based on multi-agent modeling.

Although the multi-agent model has been used to analyze several economic and management problems, and the research results are regarded more profoundly, they all rely on certain scenarios. Once the scenarios are shifted to an unknown one, the results cannot be matched. In this paper, a new research method named exploratory computational experiment is introduced to resolve the problems coming from the social complex system, where individual's behaviors are irrational, diverse, and complex, and collective behavior is dynamical, complex, and critical. Firstly, the foundation of the computational experiment is introduced, then several important problems, how individuals make the decision under complex environment, how collective behavior have emerged when different conflicts co-exist, and how to evaluate collect behaviors, are analyzed. To specify this new method, two examples of how to design a scientific mechanism to make the traffic system more effective and how is the evolution law of giant components in scale-free networks if the parameters are changed continuously. The results show that multi-agent modeling based on irrational behaviors controlled by individual dynamical game radius and memory length limited can describe the social problem more accurately, the exploratory computational experiment can give us more profound conclusions.

Accelerating the integration of ChatGPT and other large‐scale AI models into biomedical research and healthcare

AbstractLarge‐scale artificial intelligence (AI) models such as ChatGPT have the potential to improve performance on many benchmarks and real‐world tasks. However, it is difficult to develop and maintain these models because of their complexity and resource requirements. As a result, they are still inaccessible to healthcare industries and clinicians. This situation might soon be changed because of advancements in graphics processing unit (GPU) programming and parallel computing. More importantly, leveraging existing large‐scale AIs such as GPT‐4 and Med‐PaLM and integrating them into multiagent models (e.g., Visual‐ChatGPT) will facilitate real‐world implementations. This review aims to raise awareness of the potential applications of these models in healthcare. We provide a general overview of several advanced large‐scale AI models, including language models, vision‐language models, graph learning models, language‐conditioned multiagent models, and multimodal embodied models. We discuss their potential medical applications in addition to the challenges and future directions. Importantly, we stress the need to align these models with human values and goals, such as using reinforcement learning from human feedback, to ensure that they provide accurate and personalized insights that support human decision‐making and improve healthcare outcomes.

Ultimately Bounded TSK Fuzzy Management in Urban Traffic Stream Mechanism: Multi-Agent Modeling Approach

In this paper, a novel control approach based on the selection of the type of traffic light and the period of the green phase to accomplish an optimum balance at intersections is proposed. This balance should be flexible to the static behavior of time, and randomness in a traffic situation, which the goal of the proposed method is to reduce traffic volume in transportation, the average delay for each vehicle, and control over the crash of cars. The proposed method was specifically investigated at the intersection through an appropriate timing of traffic lights by sampling a multi-agent system. It consists of a large numbers of intersections, each of which is considered as an independent agent that exchanges information with each other, and the stability of each agent is provided separately. The robustness against uncertainties, scalability, and stability of the closed-loop overall system are the main merits of the proposed methodology. The simulation results show that the fuzzy intelligent controller in this multi-factor system which is a TS fuzzy is more useful than scheduling in the fixed-time method and it reduces the lengths of vehicles queuing.

Exploratory policy analysis for electric vehicle adoption in European countries: A multi-agent-based modelling approach

To reach climate neutrality goals, European countries need to reduce their transportation sector emissions. To this end, implementing effective incentive policies to accelerate electrical vehicle (EV) adoption plays a vital role. With this study, we highlight the important role of governments, showing that even with the provision of certain mild non-financial incentives, such as charging infrastructure development, EV adoption rates can be significantly increased. We develop a multi-agent model of EV adoption within European countries up to the year 2030. This integrated framework can capture the interplay between technical, financial, and social aspects of the EV adoption process. We find that an annual increase of only 10% in the charging infrastructure incentives over ten years can increase the average European EV adoption rate to 46%. We observe that countries that have both a low level of charging point density and a high population density would benefit more from the provision of charging infrastructure incentives. Countries with lower overall achieved EV shares, on the other hand, are found to be relatively insensitive to such provisions. We also characterize how a higher level of charging point density will lead to more rapid EV adoption.

A multi-agent model of urban microgrids: Assessing the effects of energy-market shocks using real-world data

The shift towards renewable energy sources (RES) in energy systems is becoming increasingly important. Residential energy generation and storage assets, smart home energy management systems, and peer-to-peer (P2P) electricity trading in microgrids can help integrate and balance decentralized renewable electricity supply with an increasingly electrified power, heat, and transport demand, reducing costs and CO2 emissions. However, these microgrids are difficult to model because they consist of autonomous and interacting entities, leading to emergent phenomena and a high degree of complexity. Agent-based modeling is an established technique to simulate the complexity of microgrids. However, the existing literature still lacks real-world implementation studies and, as a first step, models capable of validating the existing results with real-world data. To this end, we present an agent-based model and analyze the corresponding microgrid performance with real-world data. The model quantifies economic, technical, and environmental metrics to simulate microgrid performance holistically and, in line with state-of-the-art research, consists of self-interested, autonomous agents with specific load profiles, RES generation, and demand-response potential. The model can simulate a P2P marketplace where electricity is traded between agents. In the second part of the paper, we validate the model with data from a medium-sized German city. In this case study, we also compare microgrid performance in 2022, during the energy market crisis in Europe, with historical data from 2019 to assess the effects of energy market shocks. Our results show how microgrids with P2P trading can reduce electricity costs and CO2 emissions. However, our trading mechanism illustrates that the benefits of energy-community trading are almost exclusively shared among prosumers, highlighting the need to consider distributional issues when implementing P2P trading.

Mean-field optimal control in a multi-agent interaction model for prevention of maritime crime

We study a multi-agent system for the modeling maritime crime. The model involves three interacting populations of ships: commercial ships, pirate ships, and coast guard ships. Commercial ships follow commercial routes, are subject to traffic congestion, and are repelled by pirate ships. Pirate ships travel stochastically, are attracted by commercial ships and repelled by coast guard ships. Coast guard ships are controlled. We prove well-posedness of the model and existence of optimal controls that minimize dangerous contacts. Then we study, in a two-step procedure, the mean-field limit as the number of commercial ships and pirate ships is large, deriving a mean-field PDE/PDE/ODE model. Via Γ-convergence, we study the limit of the corresponding optimal control problems.

Managing human involvement in an energy community: Application to a subway station

Energy communities become increasingly diverse, as sharing mechanisms and artificial intelligence become an important component of the cooperation between people and energy systems. This cooperation can play an important role at municipality level, where transportation systems that consume huge quantities of energy (i.e. subway networks) could become more efficient and sustainable, with the active involvement of people. In this context, we propose a method to manage the involvement of people in the cooperation mechanism established between an energy community and a subway station. The method relies on a multi-objective recommendation strategy that provides the members the optimal times to commute in order to provide a positive environmental and economical impact at municipality level, while also considering human dissatisfaction. The strategy is simulated in a multi-agent model, where agents represent commuting passengers. Results show that important economical and environmental performances may still be obtained when both the system performance and human dissatisfaction are minimised.

Learning asymmetric encryption using adversarial neural networks

We propose a multi-agent adversarial neural networks model where a sender (Alice) and a receiver (Bob) are able to learn to use a pair of public/private keys in order to protect their communication from one or more attackers/eavesdroppers. Existing work in the field required shared symmetric information between Alice and Bob before initiating the training process. To the best of our knowledge, this is the first work in which Alice and Bob with asymmetric information can train themselves to protect their communication. Our initial model setup contains five agents: sender Alice, receiver Bob, eavesdropper Eve and two neural networks (we call them public keys generator and private keys generator) that, based on a (secret) random noise from Bob, will generate a pair of public/private keys that allows Alice to encrypt a message with the public key and Bob to decrypt the message with the private key while preventing Eve from decrypting the secret message using the public key. We show that the neural networks are able to establish a communication and secure it from Eve. Finally, we consider adversaries stronger than Eve to model leakage attacks, chosen plaintext attacks (CPA) and test the distinguishability between ciphertexts. The last three experiments show that neural networks (with asymmetric information) can secure the communication providing stronger security guarantees and resilience to leakage attacks which may include leakage from the private key.

Real-Time Coordinated Operation of Power and Autonomous Electric Ride-Hailing Systems

Integration of self-driving functions in electric vehicles is radically changing the transportation systems, and representing an opportunity for power utilities to develop innovative solutions for harnessing the spatio-temporal charging flexibility of autonomous electric vehicles (AEVs). This paper develops a multi-agent reinforcement learning model for intelligent real-time coordinated operation of interdependent autonomous electric ride-hailing system (AERS) and power distribution system (PDS), which enables the coordinated routing and charging of AEVs while ensuring the quality of service by meeting spatio-temporal passenger demand and regularly charging the batteries. The proposed model adopts a modified deep Q-network method with multi-agent rollout to determine the near-optimal solution for the coordinated operation problem that determines the routing and charging of numerous AEVs in real-time. The proposed model is tested on a 13-node transportation network, adopted from the Salt Lake City transportation system, and the IEEE 33-bus test power distribution system to showcase the efficiency of the proposed model in determining the real-time routing and charging decisions for AEVs while meeting the operational constraints of both AERS and PDS.

Collision-free parking recommendation based on multi-agent reinforcement learning in vehicular crowdsensing

The recent proliferation of Fifth-Generation (5G) networks and Sixth-Generation (6G) networks has given rise to Vehicular Crowd Sensing (VCS) systems which solve parking collisions by effectively incentivizing vehicle participation. However, instead of being an isolated module, the incentive mechanism usually interacts with other modules. Based on this, we capture this synergy and propose a Collision-free Parking Recommendation (CPR), a novel VCS system framework that integrates an incentive mechanism, a non-cooperative VCS game, and a multi-agent reinforcement learning algorithm, to derive an optimal parking strategy in real time. Specifically, we utilize an LSTM method to predict parking areas roughly for recommendations accurately. Its incentive mechanism is designed to motivate vehicle participation by considering dynamically priced parking tasks and social network effects. In order to cope with stochastic parking collisions, its non-cooperative VCS game further analyzes the uncertain interactions between vehicles in parking decision-making. Then its multi-agent reinforcement learning algorithm models the VCS campaign as a multi-agent Markov decision process that not only derives the optimal collision-free parking strategy for each vehicle independently, but also proves that the optimal parking strategy for each vehicle is Pareto-optimal. Finally, numerical results demonstrate that CPR can accomplish parking tasks at a 99.7% accuracy compared with other baselines, efficiently recommending parking spaces.

Sustainable Application of Automatically Generated Multi-Agent System Model in Urban Renewal

As cities expand, many old towns face the threat of being renovated or demolished. In recent years, the drawbacks of extensive urban renewal have become increasingly apparent, and the focus of urban development is gradually shifting from efficiency to quality. This study aims to combine urban renewal with emerging technologies to address the dilemma between efficiency and quality in urban renewal. The study found that algorithm models based on graph theory, topology, and shortest path principles neglect the influence of internal states and visual features on pedestrian activity, resulting in lower simulation accuracy. Although incorporating internal states and visual features into the core of the algorithm further improved the simulation accuracy, the model operates in a 3D environment with lower efficiency. To address the problems of insufficient simulation accuracy and low efficiency, this study proposes a dynamic pedestrian activity model based on a multi-agent system and incorporating visual features. The model simulates pedestrian daily activity paths using pheromones and virtual sensors as the core, and it was found that using Visibility Graph Analysis could accurately divide pheromones in the environment, thus obtaining more accurate simulation results. Subsequently, based on the optimized pedestrian model’s agent activity rules and dynamic pheromone theory, a model for automatically generating road paving in urban renewal projects was developed, and the generated results were verified for their rationality through design practice. This technology can effectively promote urban renewal and the preservation of historic neighborhoods, providing technical support for achieving sustainable urban development.

Design of a low carbon economy model by carbon cycle optimization in supply chain

IntroductionConcerning economic globalization, enterprises must work with the cooperative partner to obtain more profits and overall planning of the supply chain has become a new focus for enterprise development. This paper studies the joint emission reduction of the supply chain in green low-carbon economy development and achieve joint emission and economic cost reduction through the optimization of carbon emission and economic dispatch.MethodsThe paper firstly uses the multi-agent model to complete the fullcycle modeling of carbon emission and economic cost; Secondly, the simulated annealing-adaptive chaos-particle swarm optimization (SAACPSO) method is used to optimize various parameters in the model to achieve emission and cost reductionResultsThe results show that after the optimization, the economic cost is reduced by 0.07 and the carbon emission is also reduced by 0.16; Finally, the practical test of the model is conducted with the collected data from the local company. The results show that the multi-objective optimization model of a joint enterprise supply chain is significantly better than single optimization in terms of emission reduction.DiscussionIt provides new ideas for a green economy and technical support for the global planning of supply chain integration.

Effects of China's low-carbon policy under stochastic shocks-a multi-agent DSGE model analysis.

China has announced a target of achieving carbon peaking by 2030 and carbon neutrality by 2060. Therefore, it is important to assess the economic impacts and emission reduction effects of China's low-carbon policies. In this paper, a multi-agent dynamic stochastic general equilibrium (DSGE) model is established. We analyze the effects of carbon tax and carbon cap-and-trade policies under both deterministic and stochastic conditions, as well as their ability to cope with stochastic shocks. We found that (1) from a deterministic perspective, these two policies have the same effect. Every 1% cut in CO2 emissions will bring a 0.12% output loss, a 0.5% drop in demand for fossil fuels, and a 0.05% rise in demand for renewable energy; (2) from a stochastic perspective, effects of these two policies are different. This is mainly because economic uncertainty does not change the cost of CO2 emissions under a carbon tax policy, but it does change the price of CO2 quotas and the emission reduction behaviors under a carbon cap-and-trade policy; (3) from an economic volatility perspective, both two policies can act as automatic stabilizers. Compared to a carbon tax, a cap-and-trade policy can better ease economic fluctuations. The results of this study provide implications for policy-making.

Multi-Agent Modeling and Jamming-Aware Routing Protocols for Movable-Jammer-Affected WSNs.

Wireless sensor networks (WSNs) are widely used in various fields, and the reliability and performance of WSNs are critical for their applications. However, WSNs are vulnerable to jamming attacks, and the impact of movable jammers on WSNs' reliability and performance remains largely unexplored. This study aims to investigate the impact of movable jammers on WSNs and propose a comprehensive approach for modeling jammer-affected WSNs, comprising four parts. Firstly, agent-based modeling of sensor nodes, base stations, and jammers has been proposed. Secondly, a jamming-aware routing protocol (JRP) has been proposed to enable sensor nodes to weigh depth and jamming values when selecting relay nodes, thereby bypassing areas affected by jamming. The third and fourth parts involve simulation processes and parameter design for simulations. The simulation results show that the mobility of the jammer significantly affects WSNs' reliability and performance, and JRP effectively bypasses jammed areas and maintains network connectivity. Furthermore, the number and deployment location of jammers has a significant impact on WSNs' reliability and performance. These findings provide insights into the design of reliable and efficient WSNs under jamming attacks.

Energy price shocks and stabilization policies in the MATRIX model

The recent surge in energy prices in Europe has prompted governments to introduce policy measures to support households and businesses. This paper uses the MATRIX model, a multi-sector and multi-agent macroeconomic model calibrated on the Euro Area, to analyse the economic and distributional effects of different macro-stabilization policies in response to energy price shocks. We find that, without policies, a surge in fossil fuel prices leads to higher inflation, lower GDP, and slow recovery. Generalized tax cuts and household subsidies have no significant effects, while firm subsidies promote a faster recovery at the expense of financial instability in the medium term, leading to a second slump. However, this second-round effect can be mitigated with proper fiscal-monetary policy coordination. If timely adopted, a government-funded energy tariff reduction is the most effective policy in mitigating GDP losses at relatively low public costs, particularly when coupled with an extra-profit tax on energy firms. Energy entrepreneurs benefit from rising fuel prices in all scenarios, but workers and downstream firms’ owners benefit more from energy tariff cuts and windfall profits tax.