Multi-agent System Model Research Articles

A Category-theoretic Approach to Organization-based Modeling of Multi Agent Systems on the Basis of Collective Phenomena and Organizations in Human Societies

This paper presents an idea of using category theory for developing organizational multi-agent systems taking inspiration from collective phenomena and organizations in human societies. Category theory is used for studying and formalizing organizations and collective phenomena in human societies with the aim of capturing their logics into categorical models. Afterward, the captured models are mapped categorically to categorical MAS organizational models. This way of thinking allows studying properties of result MAS organizational models as well as properties of organizations in human societies such as stability and adaptation before taking them as landmarks for developing MAS organizational models.

Model checking real-time conditional commitment logic using transformation

A new logical language for real-time conditional commitments called RTCTLcc has been developed by extending the CTL logic with interval bounded until modalities, conditional commitment modalities, and fulfillment modalities. RTCTLcc allows us to express qualitative and quantitative commitment requirements in a convenient way. These requirements can be used to model multi-agent systems (MASs) employed in environments that react properly and timely to events occurring at time instants or within time intervals. However, the timing requirements and behaviors of MASs need an appropriate way to scale and bundle and should be carefully analyzed to ensure their correctness, especially when agents are autonomous. In this paper, we develop transformation algorithms that are fully implemented in a new Java toolkit for automatically transforming the problem of model checking RTCTLcc into the problem of model checking RTCTL (real-time CTL). The toolkit engine is built on top of the NuSMV tool, effectively used to automatically verify and analyze the correctness of real-time distributed systems. We analyzed the time and space computational complexity of the RTCTLcc model checking problem. We proved the soundness and completeness of the transformation technique and experimentally evaluated the validity of the toolkit using a set of business scenarios. Moreover, we added a capability in the toolkit to automatically scale MASs and to bundle requirements in a parametric form. We experimentally evaluated the scalability aspect of our approach using the standard ordering protocol. We further validated the approach using an industrial case study.

多智能体系统支持下的耕地非农化流转决策优化研究

多智能体系统支持下的耕地非农化流转决策优化研究

Trust-Based Situation Awareness: Comparative Analysis of Agent-Based and Population-Based Modeling

In recent years, to comprehend and analyze complex systems, multiagent systems modeling and simulation are being widely used across various disciplines. Two major approaches used for multiagent systems modeling and simulation are agent-based modeling (ABM) and population-based modeling (PBM). In multiagent community, it is a silent assumption that both approaches represent similar dynamics for large population size. One of the recent studies from literature has reported similar results for a model of situation awareness spread in multiagent systems. Trust is a significant factor that affects agents’ communication, and consequently it controls spread of situation awareness among agents in a multiagent system. Hence, current work firstly extends the reported model of situation awareness spread from literature, to incorporate interagent trust for both ABM and PBM. Later, these extended models are used for comparative evaluation of both approaches. Various simulation experiments for different population sizes (small and large) as well as population types (homogenous and heterogeneous) are conducted and analyzed. Results of these experiments show that for large and homogeneous population, ABM approximates behavior of PBM, but for even slightly heterogeneous population, these approaches do not produce similar results irrespective of population sizes. Thus, the current study reports that, under some conditions, ABM and PBM produce similar results for trust-based situation awareness spread in multiagent systems, but this assumption does not hold true at large.

Modelling and analysis of multi-agent systems using UPPAAL SMC

This paper proposes a novel approach to modelling and analysis of complex multi-agent systems. The approach is based on actors and asynchronous message passing, and exploits the UPPAAL statistical model checker (SMC) for the experiments. UPPAAL SMC is interesting because it automates simulations by predicting the number of executions capable of ensuring a required output accuracy, it uses statistical techniques (Monte Carlo-like simulations and sequential hypothesis testing) for extracting quantitative measures from the simulation runs, and it offers a temporal logic query language to express property queries tailored to the application needs. The paper describes the approach, clarifies its structural translation on top of UPPAAL SMC, and demonstrates its practical usefulness through modelling and analysis of a large scale and adaptive version of the Iterated Prisoner's Dilemma (IPD) problem. The case study confirms known properties, namely the emergence of cooperation under context preservation, that is when the player interaction links are preserved during the game, but it also suggests some new quantitative measures about the temporal behaviour which were not previously pointed out.

Formal Modelling of Real-Time Self-Adaptive Multi-Agent Systems

Formal Modelling of Real-Time Self-Adaptive Multi-Agent Systems

Симуляция поведения мультиагентных систем с ациклически взаимодействующими агентами

In this paper, we present an approach to model and simulate models of multi-agent systems (MAS) using Petri nets. A MAS is modeled as a set of workflow nets. The agent-to-agent interactions are described by means of an interface. It is a logical formula over atomic interaction constraints specifying the order of inner agent actions. Our study considers positive and negative interaction rules. In this work, we study interfaces describing acyclic agent interactions. We propose an algorithm for simulating the MAS with respect to a given interface. The algorithm is implemented as a ProM 6 plug-in that allows one to generate a set of event logs. We suggest our approach to be used for evaluating process discovery techniques against the quality of obtained models since this research area is on the rise. The proposed approach can be used for process discovery algorithms concerning internal agent interactions of the MAS.

A Decomposition-based Approach of Global Norms for Hierarchical Normative Systems

Holonic Multi-Agent System (HMAS) forms a promising approach to software engineering for the modeling and development of hierarchical autonomous systems (Intelligent transportation systems, Smart city management systems, etc.). Effectively, norms concepts are well adapted to the definition of HMAS. Its can be considered as a powerful way to specify the non-functional requirements of these types of systems. Non-functional requirements can conflict with each other (for example, cost and quality, comfort and economy). One of the main challenges currently faced in HMAS research is the coherence checking of norms that specify non-functional requirements. However verification approaches are limited by the state-space of the system under study. Specifying and verifying a global normative model in a single level is then complex and difficult. However, most normative models for multi-agent systems do not take into account the complexity of coherence verification algorithms of norms. One of solutions is to employ the use of coherent refinement process of global norms.This paper proposes a Global Norms Decomposition (GND) approach for hierarchical and critical autonomous systems. Indeed, the GND approach allows (i) the specification and verification of global norms in the abstract level of the studied system, and (ii) the successive refinement of these norms using a set of refinement rules that preserve properties of the system already proven in the highest level, in order to arrive finally at a concrete normative context which constitutes the behaviour model of the system. The GND approach allows the simplification of specification of norms, for an incremental specification using a refinement process, and the reduction of complexity of checking the coherence of norms, building verification using refinement rules. Our approach is also illustrated by a case study describing smart city management system.

On the Spatial Boundedness of Cellular RDA-nets

Cellular resource driven automata nets (CRDA-nets) are a generalization of the concept of two-level resource nets (Petri nets) with an introduction of an infinite regular system grid. This formalism is a hybrid of Petri nets and asynchronous Cellular Automata and is designed for modeling multi-agent systems with dynamic spatial structure. Spatial boundedness is a property that guarantees the preservation of the finiteness of “geometric dimensions” of the active part of the system (for example, the living space) during its lifetime. Three variants of spatial boundedness for cellular RDA-nets are defined: localization, bounded diameter and bounded area. The properties of the corresponding algorithmic problems are investigated, their undecidability in the general case is proved. A non-trivial criterion for the localization of an one-dimensional CRDA-net is proposed, based on the new concept of the RDA propagation graph. An algorithm is described for constructing a propagation graph, using the method of saturation of generating paths. A method for estimating the diameter of an 1-dim CRDA with a bounded propagation graph is presented.

Embedding ecological sensitivity analysis and new satellite town construction in an agent-based model to simulate urban expansion in the beijing metropolitan region, China

As the national center for politics and culture, Beijing’s urbanization level is very high. Many policies within the proposed Beijing–Tianjin–Hebei region coordinated development plan affect the expansion of urban land. Therefore, it is very important to predict urban land use changes in Beijing for planning and management purposes. In this paper, we integrated of MAS (multi-agent system) and CA (cellular automata) to simulate new satellite towns construction and ecological sensitivity impact on land use change. Physical and social driving factors were used in the combined model. The MAS involved the actions of three types of agent: regional authorities, property developers, and residents. The study used the CA model to simulate the neighborhood effects of urban land use, and the MAS model to simulate agents’ decisions. The new satellite towns and an ecological sensitivity analysis were embedded in the model to simulate the impact of decision making by the Beijing government on urban land expansion. Based on the land use data of 2005, the urban land area in 2010, 2015, 2020, and 2025 was predicted using the CA-MAS model. Urban expansion occurred faster during 2015–2025 than during the previous 10 years. Three land use types, i.e., cropland, woodland, and rural residential land, were the major sources of urban expansion. With respect to government decision making, the satellite towns were the priority areas of urban development, and urban development was restricted in ecologically sensitive areas. The New Districts of Urban Development were projected to become the main areas of future urban expansion in Beijing. The area designated for urban expansion around the ecologically sensitive areas was small. The results demonstrate that satellite towns and ecological sensitivity have large impacts on urban expansion. The results of this study will help to protect ecologically sensitive land, while enabling harmonious expansion of the city.

Security model of mobile multi-agent robotic systems with collective management

Security model of mobile multi-agent robotic systems with collective management

Empirical Evaluation of Similarity Coefficients for Multiagent Fault Localization

Detecting and diagnosing unwanted behavior in multiagent systems (MASs) are crucial to ascertain correct operation of agents. Current techniques assume a priori knowledge to identify unexpected behavior. However, generation of MAS models is both error-prone and time-consuming, as it exponentially increases with the number of agents and their interactions. In this paper, we describe a light-weight, automatic debugging-based technique, coined extended spectrum-based fault localization for MAS (ESFL-MAS), that shortens the diagnostic process, while only relying on minimal information about the system. ESFL-MAS uses a heuristic that quantifies the suspiciousness of an agent to be faulty. Different heuristics may have a different impact on the diagnostic quality of ESFL-MAS. Our experimental evaluation shows that 10 out of 42 heuristics (namely accuracy, coverage, Jaccard, Laplace, least contradiction, Ochiai, Rogers and Tanimoto, simple-matching, Sorensen-dice, and support) yield the best diagnostic accuracy (96.26% on average) in the context of the MAS used in our experiments.

Simulation of Land Use Changes in a Coastal Reclaimed Area with Dynamic Shorelines

Reclamation is capable of creating abundant land to alleviate the pressure from land shortages in China. Nevertheless, coastal reclamation can lead to severe environmental degradation and landscape fragmentation. It is quite important to monitor land use and cover change (LUCC) in coastal areas, assess coastal wetland change, and predict land use requirements. The siltation of tidal flats will result in the dynamic growth and continuous expansion of coastal areas. Therefore, the process of land change in coastal areas is different from that under the fixed terrestrial boundary condition. Cellular Automata and Multi-Agent System (CA-MAS) models are commonly used to simulate LUCC, and their advantages have been well proven under the fixed boundary condition. In this paper, we propose CA-MAS combined with a shoreline evolution forecast (CA-MAS-SEF) model to simulate the land change in coastal areas. Meanwhile, the newly increased area, because of the dynamic growth of tidal flats, is considered in the simulation process. The simulation results using the improved method are verified, and compared with observed patterns using spatial overlay. In comparison with simulation results that do not consider the expansion of tidal flats, the Kappa coefficient estimated while considering the dynamic growth of tidal flats is improved from 65.9% to 70.5%, which shows that the method presented here can be applied to simulate the LUCC in growing coastal areas.

Optimizing Energy Consumption in a Decentralized Manufacturing System

The deployment of modern information and communication technologies (ICT) within manufacturing systems leads to the creation of so-called cyber-physical production systems that consist of intelligent interconnected production facilities. One of the expected features of cyber-physical production systems is found to be the capability of self-organization and decentralized process planning in manufacturing. The functionality as well as the benefit of such self-organization concepts is yet to be proved. In this paper, the implementation of a virtual test field for the simulation of manufacturing systems based on a multi-agent system modeling concept is presented and used to evaluate a concept of decentralized process planning. Thereby, special focus is laid on the impact on energy consumption. The simulation results show the potential for energy reduction in manufacturing by a decentralized process-planning concept and yields hints for further development of such concepts.

A Framework for Implementing Formally Verified Resource-Bounded Smart Space Systems

Context-aware computing is a mobile computing paradigm that helps designing and implementing next generation smart applications, where personalized devices interact with users in smart environments. Development of such applications is inherently complex due to these applications adapt to changing contextual information and they often run on resource-bounded devices. Most of the existing context-aware development frameworks are centralized, adopt client---server architecture, and do not consider resource limitations of context-aware devices. This paper presents a systematic framework to modelling and implementation of resource-bounded multi-agent context-aware systems on Android devices. The proposed framework makes use of semantic technologies for context modelling and reasoning about resource-bounded context-aware agents, Android powered smartphones as development platform, a suitable communication model and declarative rule-based programming as a preferred development language.

A multi-agent based security policy for web related applications: a hybrid approach using context and ontology

In this paper, we present a multi-agent system model based, for securing computer systems that operate in complex and pervasive environments, such as web applications. These applications are characterised by unpredictable changes in terms of resources and circumstances. The choice of the multi-agent systems paradigm is motivated by its suitability to this type of problems, in terms of autonomy, flexibility and scalability. Therefore, we propose a hybrid approach considering the execution context to express the circumstances and conditions of the environment. The purpose is to better interact with users and respond effectively to the posed situation. Furthermore, the use of ontologies that enrich the considered system by semantics, strengthen the proposed approach by increasing the objectivity and overcome any ambiguities caused by the heterogeneity of the different intervening entities. So, the proposed model aims to respond to some security problems for information systems related to web, as well as to improve it in terms of relevance, clarity and objectivity.

On the Spatial Boundedness of Cellular RDA-nets

Cellular resource driven automata nets (CRDA-nets) is a generalization of the concept of two-level resource nets (Petri nets) with an infinite regular system grid. This formalism is a hybrid of Petri nets and asynchronous Cellular Automata and is designed for modeling multi-agent systems with dynamic spatial structure. Spatial boundedness is a property that guarantees the preservation of the finiteness of “geometric dimensions” of the active part of the system (for example, the living space) during its lifetime. Three variants of spatial boundedness for cellular RDA-nets are defined: localization, bounded diameter and bounded area. The properties of the corresponding algorithmic problems are investigated, their undecidability in the general case is proved. A non-trivial criterion for the localization of an one-dimensional CRDA-net is proposed, based on the new concept of the RDA propagation graph. An algorithm is described for constructing a propagation graph, using the method of saturation of generating paths. A method for estimating the diameter of an 1-dim CRDA with a bounded propagation graph is presented.

Modeling of Multi Agent System: From Analysis to Design

Multi Agent System (MAS) is an assembly of autonomous, proactive and dynamic entities called agents. These agents provide some service once an event is invoked. By providing the designated service the MAS achieves a specific goal. A logic based formal analysis of MAS is necessary to understand the system, its structure and dynamics. A systematic and sequential methodology is used for analysis of MAS. As a result of analysis several constructs in form of templates is obtained to represent the different facets of MAS. This is extremely necessary as it allows to explicitly identify how agents directly or indirectly interact or coordinate and collaborate with each other. This paper proposes an Architectural Design Language called Agent Description Language (ADL) that will be required to map these analysis templates to the design phase. Establishing the correctness of the structural transformation and correspondence from analysis to design phase constitutes the other side of the proposal. The proposed ADL has been used to get the corresponding classes of the templates generated. Jason, the universal MAS programming language has been used to implement the ADL classes to get the logical design which conforms to the conceptual design phase. The case study of Medical Emergency Service (MES) has also been taken to illustrate the proposals and findings.

Combining fault injection and model checking to verify fault tolerance, recoverability, and diagnosability in multi-agent systems

We present an automated technique that combines fault injection with model checking to verify fault tolerance, recoverability, and diagnosability in multi-agent systems. We define a general method for mutating a multi-agent systems model representing correct behaviour by injecting faults into it, and specification patterns based on temporal-epistemic formulas to reason about the correct and faulty behaviours of the mutated model. The technique is implemented in a toolkit that can be used for injecting automatically faults into a multi-agent systems program. The usefulness of the methodology is demonstrated by injecting a number of faults into a model of the IEEE 802.5 token ring LAN protocol and analysing the protocol's fault tolerance, by verifying a number of temporal-epistemic specifications.

A metamodeling approach for the identification of organizational smells in multi-agent systems: application to ASPECS

Software Quality is one of the most important subjects in the Process Development Software, especially in large and complex systems. Much effort has been devoted to the development of techniques and concepts to improve software quality over the years. We are especially interested on smells, which represent anomalies or flaws in the design/code that can have serious consequences in maintenance or future development of the systems. These techniques have a strong development in the Object Oriented paradigm, however, very few studies were conducted in the agent oriented paradigm. In this paper we focus on the detection of design smells applied to multi-agent systems models based on the organizational approach, named Organizational Design Smells (ODS). Early and automatic detection of these ODS allows reducing the costs and development times, while increasing the final product’s quality. To achieve this objective, validation rules were defined based in the EVL language. The approach is illustrated with two examples, their validation rules, and the refactoring solutions proposed.