No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Multi-Agent System Based Approach to Intelligent Process Automation Systems
Abstract
A more promising technology to integrate existing software systems and their functionalities and to add assistant systems
for the shop floors is to be found in software agents. Concerning with the applications of agent technology to intelligent
process automation systems, a flexible and extensible approach based on multi-agent system (MAS) and distributed planning
technique to reach for increased flexibility and fault-tolerance in process monitoring and control operations is proposed.
The monitoring operations are aimed at combining information from different sources depending on the monitoring tasks. The
control operations are supervisory control tasks performing either in both sequential and iterative forms. The agent layer
is used for monitoring the operations of the lower-level automation systems and reconfiguring its control logic. It operates
as a distributed planning and plan execution system to increase the operational flexibility of the whole process automation
systems. The software architecture and its functionalities and components for the implementation of the proposed approach
are also presented. The design strategy and the discussion on the proposed approach are provided.
KeywordsAgent-Automation system-Control operation-Monitoring operation-Software agent
ResearchGate has not been able to resolve any citations for this publication.
In this paper, we summarize advantages of using agent technology in the design and operation of industrial control systems.
To support appropriate responses to dynamically generated events, we divided the agent operation into two main levels: planning
and real-time control. A higher-level software agent covers the planning phase and low-level distributed control agent provides
the real-time control. The proposed architecture is targeted on applications where real-time response is essential but also
more sophisticated higher-level control mechanisms are needed for efficient control. An implementation of such system is briefly
described.
An approach to extend process monitoring with the help of information agents (IA) handling semantic data is presented in this
paper. According to this approach, an operator of a process automation system can configure monitoring tasks that a group
of IAs performs proactively. The monitoring tasks are assumed to be composites which refer to several process observations
and their logical relations. The purpose of these composite monitoring tasks is to enhance the work of the operator by letting
him to supervise process phenomena at a higher level of abstraction instead of following a large amount of simple measurement
data. The monitoring agents operate as a multi-agent system consisting of agents with capabilities to combine both numerical
and symbolic information from several data sources. The agents can setup and execute user configured monitoring tasks cooperatively.
The approach is illustrated with test scenarios using data from an industrial paper making process.
Agent-based systems technology has generated lots of excitement in recent years because of its promise as a new paradigm for conceptualizing, designing, and implementing software systems. This promise is particularly attractive for creating software that operates in environments that are distributed and open, such as the internet. Currently, the great majority of agent-based systems consist of a single agent. However, as the technology matures and addresses increasingly complex applications, the need for systems that consist of multiple agents that communicate in a peer-to-peer fashion is becoming apparent. Central to the design and effective operation of such multiagent systems (MASs) are a core set of issues and research questions that have been studied over the years by the distributed AI community. In this article, I present some of the critical notions in MASs and the research work that has addressed them. I organize these notions around the concept of problem-solving coherence, which I believe is one of the most critical overall characteristics that an MAS should exhibit.
An approach to intelligent process automation based on distributed planning agents is presented in this paper. According to this approach a higher-level agent-based automation layer supervises an ordinary process automation system. The purpose of the agent layer is to monitor the operation of the lower-level automation system and semi-autonomously reconfigure its control logic when needed. The agent layer operates as a distributed planning and plan execution system, which creates and runs reconfiguration sequences on the ordinary automation system in order to adapt it to various situations. In this way the approach aims to increase the operational flexibility of the whole automation system. The approach is demonstrated with a laboratory test process where process startup scenarios have been imitated. Experiences from the initial test runs are also described.
Existing modeling frameworks for manufacturing system control can be classified into hierarchical, heterarchical, and hybrid control frameworks. The main drawbacks of existing frameworks are discussed in this paper. A new hybrid modeling framework is also described. It is a hybrid of the two: hierarchical and heterarchical frameworks. In this proposed framework, entities (e.g., parts) and resources (e.g., material handling devices, machines, cells, departments) are modeled as holonic structures that use intelligent agents to function in a cooperative manner so as to accomplish individual, as well as cell-wide and system-wide objectives. To overcome the structural rigidity and lack of flexibility, negotiation mechanisms for real-time task allocation are used. Lower-level holons may autonomously make their negotiations within the boundary conditions that the higher-level holons set. Horizontal, as well as vertical decisions, are made between various levels of controllers, and these are explicitly captured in the model.
Flexible manufacturing systems have long been touted as an application area for supervisory control theory. Unfortunately, due to the typical exponential growth of state space with the number of interacting subsystems, concurrent systems such as manufacturing applications have, for the most part, remained beyond the reach of existing supervisory control theory tools. This paper demonstrates how, by imposing a hierarchical, modular, interface-based architecture on the system, significant gains can be made in the size of applications that can be handled by supervisory control theory. We first review hierarchical interface-based supervisory control, providing the theory necessary to motivate the creation of well-defined automata-based interfaces between components. This architecture permits the verification of global safety (controllability) and nonblocking properties to be decomposed into a set of local checks, each of which only involves an individual component subsystem and its interface automata. The paper then provides a detailed description of how the theory can be applied to the design and verification of a flexible manufacturing system work cell. The work cell model is based on the Atelier Intere´tablissement de Productique flexible manufacturing workcell, a system that has been previously studied in the literature with limited success.
Looks at how cooperative multiagent systems designed for monitoring and diagnosing real-time nuclear power plant failures can help operators better anticipate operational problems
The aim is to improve the monitoring and control of district heating systems through the use of agent technology. In order to increase the knowledge about the current and future state in a district heating system at the producer side, each substation is equipped with an agent that makes predictions of future consumption and monitors current consumption. The contributions to the consumers, will be higher quality of service, e.g., better ways to deal with major shortages of heat water, which is facilitated by the introduction of redistribution agents, and lower costs since less energy is needed for the heat production. Current substations are purely reactive devices and have no communication capabilities.
A multi-agent system architecture for coordination of just-in-time production and distribution is presented. The problem to solve is two-fold: first the right amount of resources at the right time should be produced, and then these resources should be distributed to the right consumers. In order to solve the first problem, which is hard when the production and/or distribution time is relatively long, each consumer is equipped with an agent that makes predictions of future needs that it sends to a production agent. The second part of the problem is approached by forming clusters of consumers within which it is possible to redistribute resources fast and at a low cost in order to cope with discrepancies between predicted and actual consumption. Reallocation agents are introduced (one for each cluster) to manage the redistribution of resources. The suggested architecture is evaluated in a case study concerning management of district heating systems. Results from a preliminary simulation study show that the suggested approach makes it possible to control the trade-off between quality-of-service and degree of surplus production. We also compare the suggested approach to a reference control scheme (approximately corresponding to the current approach to district heating management), and conclude that it is possible to reduce the amount of resources produced while maintaining the quality of service. Finally, we describe a simulation experiment where the relation between the size of the clusters and the quality of service was studied.
In this article an agent-based framework is presented for designing and implementing multicontroller systems. These systems are getting a lot of attention within the field of control engineering, as some researchers have argued that they are suitable for solving complex control problems in a practical manner (Johansen and Murray-Smith, 1997). A practical and structured framework for designing and implementing these architectures is still lacking. The agent-based framework presented in this article provides such a framework and can be used to design and implement hierarchical structured multi-controller systems that consist of a set of heterogeneous control algorithms that are combined by heterogeneous techniques. Furthermore, the architecture is open such that controllers can be easily added, removed or changed. Finally, a case study is given to illustrate the framework.
ARCHON™ (ARchitecture for Cooperative Heterogeneous ON-line systems) is Europe’s largest project in the area of Distributed Artificial Intelligence (DAI). It has devised a general-purpose architecture, software framework and methodology which has been used to support the development of DAI systems in a number of industrial domains. Some examples of the applications to which it has been successfully applied include: electricity distribution and supply, electricity transmission and distribution, control of a cement kiln complex, control of a particle accelerator, and control of a robotics application. The type of cooperating community that it supports has a decentralised control regime and individual problem solving agents which are large grain, loosely coupled, and semi-autonomous.
The manufacturing industries are now experiencing fierce pressure of competition from every corner on this planet. In addition, the advancement in computer networks and information technologies has been gradually reshaping the manufacturing companies by shifting from the industrial age to the information era. Due to these elevated competitiveness and advanced computer technology, a number of new manufacturing and management strategies (e.g., CE or CIM) have emerged for the innovation of manufacturing enterprises. Although they have different definitions and scopes, there are several common issues: inter-enterprise functions integration; inter-enterprise resources integration; and collaboration. This paper proposes a new multi-agent system (MAS) architecture to support the inter-enterprise functions/resources integration and collaboration over the networked environment, including the hybrid agent architecture and hybrid network architecture. In contrast to the existing agent architectures, the proposed agent architecture enables agents to exhibit the hybrid (continuous and discrete) behavior and interactions. In addition, our network architecture is more suitable for building the large-scale distributed manufacturing systems that are prone to dynamic random changes of their environment. Based on the proposed MAS architecture, a collaborative product development environment is implemented as a starting point, and a multidisciplinary team-oriented design problem is illustrated to provide the vision of the proposed MAS architecture.
Despite a great deal of research, there still exists a number of challenges before making agent-based computing a widely accepted
paradigm in software engineering practice. In order to realize an engineering change in agent oriented software engineering,
it’s necessary to turn agent oriented software abstractions into practical tools for facing the complexity of modern application
areas. The paper presents a customizable development architecture for multi-agent systems, which can empower the developer
to assemble a methodology tailored to the given project by putting appropriate meta models together, much like developers
building applications from third party off-the-shelf components. To exemplify its feasibility and effectivity, the construction
of C4I system is presented as a case study.
This paper proposes a framework for the implementation of multiagent system for hierarchical control of complex process control
systems based on OPC technology that is widely applied to the automation control systems. This framework is proposed with
utilization of OPC technology in both continuous-event part and discrete-event part by incorporating with XML for the negotiation
and cooperation in the environments of multiagent system. The framework design criteria are also described. The comparison
of the proposed framework with existing frameworks is made to demonstrate that the proposal is reliable and feasible in order
to apply to agent-based process control applications.
The paper presents hierarchical and multi-agent control and information system with real-time update of self organising database
as well as with negotiation capability for control events and decisions. A practical application is presented, that utilizes
the OPC technology in the continuous-time part, and scripts using XML in the discrete-time part of the system for negotiation
and cooperation in multi-agent environment. This feature is applied for the improvement of a non-conventional biotechnological
process control in the pilot plant.
This paper focuses on the design of multiagent systems using object oriented design patterns. In earlier papers the distinctions and relationships between agents and objects were discussed; these findings are employed here as a foundation for a multiagent system architecture. Patterns are flexible, reusable, and effective object oriented designs. The role of design patterns for multiagent systems is discussed, and a design is proposed that addresses agent concurrency, virtual migration, collaboration, and reasoning. The paper yields a software architecture for multiagent systems and compares it to other approaches. Key software engineering issues for multiagent systems, such as reusability, testability, and maintainability, are addressed by the proposed design.
An approach to combine information access and control operations in a process automation system extended with multi-agent
system technology is presented in this paper. According to this approach a multi-agent system supervises an ordinary process
automation system by performing higher-level information access and control operations. The information access operations
are aimed for actively combining information from different sources depending on the monitoring tasks of the users. The control
operations of the multi-agent system are supervisory control tasks performed either in sequential or iterative fashion. The
expected benefit of the multi-agent system is enhanced adaptability of the automation system and increased situation awareness
of its users. The architecture of the multi-agent system is based on the BDI agent model and utilization of so-called ontologies.
An approach for engineering applications for this kind of a multi-agent system is also discussed. The approach is demonstrated
with results from experiments performed with industrial test data and a laboratory test process.
Ontologies are needed for agent communication as they provide means for describing intended semantics of the language used
for expressing the content of messages. We show what types of ontologies and their descriptions are used in the area of multi-agent
systems for manufacturing and compare them with the ontology modeling languages in the semantic web area. We present a framework
that enables adding semantics of OWL ontologies used in the semantic web to the ontologies without any formal explicit description
used in the manufacturing domain (expressed usually in XML). The ontologies with explicitly defined semantics can be used
for purposes that require formal semantics such as for automated reasoning over the source ontologies from the manufacturing
domain or for integration of manufacturing and semantic web agents. This integration is one of the key issues for successful
development of agent-based solutions for manufacturing as well as virtual enterprises.
In this paper, we present a framework for the implementation of multi-agent-systems for production control of complex manufacturing
systems. We present the results of a requirement analysis for production control systems for complex manufacturing systems;
then we describe the framework design criteria. Our framework supports the inclusion of distributed hierarchical decision-making
schemes into the production control. Furthermore, in order to increase the coordination abilities of multi-agent-systems,
we follow the decision-making and staff agent architecture suggested in the PROSA reference architecture. We indicate the
usage of the framework for designing and implementing an agent-based production control system for semiconductor manufacturing
processes in a case study.
This thesis describes studies on application of multi-agent systems (acronym: MAS) to enhance process automation systems. A specification of an extended process automation system is presented. According to this specification, MAS can be used to extend the functionality of ordinary process automation systems at higher levels of control. Anticipated benefits of the specification include enhanced reconfigurability, responsiveness and flexibility properties of process automation. Previous research concerning applications of MAS in process automation has been more limited than in other fields of automation. There has been more research about this topic for example in the area of discrete manufacturing. As goal-oriented distributed systems with coordination capabilities MAS have been found applicable to a part of automation functions, e.g. modification of control logic in abnormal situations. However, when applying MAS to process automation the particular characteristics of this application domain need to be taken into account. The important role of continuous control in process automation needs to be considered. In this thesis, a specification of an agent platform for process automation is presented as a basis for applying MAS in this application domain. The specification extends a FIPA-compliant agent platform with process automation specific functionality. It utilises a hierarchical agent organisation, a BDI-agent model and qualitative reasoning. It also presents a model for programming MAS applications for process automation with techniques of distributed planning and search. Two applications are specified using the platform. One of these shows how the techniques of distributed planning can be applied in sequential control. The other provides a design model for supervisory continuous control applications using the techniques of distributed search. Experiments performed with a laboratory test environment using prototype implementations of the applications are presented. The experiments are able to demonstrate the feasibility of the approach in limited test scenarios. They also provide information about in which ways MAS techniques are able enhance the properties of process automation. As a result of the work presented in this thesis more knowledge has been gained about application of MAS in process automation. The specification of the agent platform for process automation and its applications provide a basis for further studies of this topic. Report / Helsinki University of Technology, Information and Computer Systems in Automation, ISSN 1456-0887; 11
A multi-agent architecture to automate complex industrial systems is described. The multi-agent system is created with artificial social behaviors to enable cooperation among the agents. Agents organize their respective views about the system to react to changes and to discover system capabilities. We explore the implications of using agent technology in the design and operation of a chilled water system. Our aim is to contribute with a set of guidelines on how to build autonomous agents for industrial automation.
The Fraunhofer Institute for Information and Data Processing has developed a novel production monitoring and control system (PMC) for DaimlerChrysler Bremen for its C-type car which will begin production in 2007. Called the Provis.Agent, it is the first agent-based PMC system for distributed real-time production monitoring that allows integration with other shop floor related applications.
. This paper describes a methodology that is being developed for designing and building agent-based systems for the domain of production control. In particular, this paper deals with the steps that are involved in identifying the agents and in specifying their responsibilities. The methodology aims to be useable by engineers who have a background in production control but who have no prior experience in agent technology. For this reason, the methodology needs to be very prescriptive with respect to the agent-related aspects of the design. 1 Introduction Software agents are on the verge of becoming a key control technology for largeseries production control systems. With ever shorter product life-cycles, decreasing product launch times, and increasing product variety, manufacturing processes must provide more product flexibility and higher volume scalability while maintaining high product quality and low manufacturing costs. Agent technology is well suited to addressing the con...
ARCHON^TM (ARchitecture for Cooperative Heterogeneous ON-line systems) is Europe's largest project in the area of Distributed Artificial Intelligence (DAI). It has devised a general-purpose architecture, software framework and methodology which has been used to support the development of DAI systems in a number of industrial domains. Some examples of the applications to which it has been successfully applied include: electricity distribution and supply, electricity transmission and distribution, control of a cement kiln complex, control of a particle accelerator, and control of a robotics application. The type of cooperating community that it supports has a decentralised control regime and individual problem solving agents which are large grain, loosely coupled, and semi-autonomous. This paper tackles a broad range of issues related to the application of ARCHON technology to industrial applications. Firstly, it gives the rationale for a DAI approach to industrial applications ...
this paper in which we analyse
Multiagent Systems. Magazine of American Association for
- K P Sycara
- K.P. Sycara
Agent Development Framework
- Jadf - Java