Agent-Based Systems for Intelligent Manufacturing: A State-of-the-Art Survey

Agent technology has been considered as an important approach for developing distributed intelligent manufacturing systems. A number of researchers have attempted to apply agent technology to manufacturing enterprise integration, supply chain management, manufacturing planning, scheduling and control, materials handling, and holonic manufacturing systems. This paper gives a brief survey of some related projects in this area, and discusses some key issues in developing agent-based manufacturing systems such as agent technology for enterprise integration and supply chain management, agent encapsulation, system architectures, dynamic system reconfiguration, learning, design and manufacturability assessments, distributed dynamic scheduling, integration of planning and scheduling, concurrent scheduling and execution, factory control structures, potential tools and standards for developing agent-based manufacturing systems. An extensive annotated bibliography is provided.     

Engineering framework for agent-based manufacturing control

MAS (multi-agent systems) and HMS (holonic manufacturing systems) are enabling the vision of the Plug & Play Factory and paving the way for future autonomous production systems. This paper reviews the state of the art in implementations of agent-based manufacturing systems, and identifies the lack of engineering tools as a technological gap for widespread industrial adoption of the paradigm. The lack of tools limits the implementation of agent-based manufacturing systems within reach of only a handful of domain experts. One of the current challenges for the design and implementation of intelligent agents is the simulation and visualization of the agent societies. This issue is significant as soon as the software agent is embedded into a mechatronic device or machine resulting in a physical intelligent agent with 3D-mechanical restrictions. These mechanical restrictions must be considered in the negotiations between agents in order to coordinate the execution of physical operations. This paper presents an engineering framework that contributes towards overcoming the identified technology gap. The framework consists of a comprehensive set of software tools that facilitate the creation, simulation and visualization of agent societies. The 3D framework is innovative in fully emulating the deployed agents, recreating multi-agent negotiations and societies that coordinate and execute control of assembly operations. The documented research describes the methodology for the 3D representation of individual physical agents, the related identified objects present in the interaction protocols, and the assembly features and clustering algorithms.     

Supporting decision making process with “ideal” software agents - What do business executives want?

According to Simon’s (1977) decision making theory, intelligence is the first and most important phase in the decision making process. With the escalation of information resources available to business executives, it is becoming imperative to explore the potential and challenges of using agent-based systems to support the intelligence phase of decision-making. This research examines UK executives’ perceptions of using agent-based support systems and the criteria for design and development of their “ideal” intelligent software agents. The study adopted an inductive approach using focus groups to generate a preliminary set of design criteria of “ideal” agents. It then followed a deductive approach using semi-structured interviews to validate and enhance the criteria. This qualitative research has generated unique insights into executives’ perceptions of the design and use of agent-based support systems. The systematic content analysis of qualitative data led to the proposal and validation of design criteria at three levels. The findings revealed the most desirable criteria for agent based support systems from the end users’ point view. The design criteria can be used not only to guide intelligent agent system design but also system evaluation.     

Applications of agent-based systems in intelligent manufacturing: An updated review

Agent technology has been recognized as a promising paradigm for next generation manufacturing systems. Researchers have attempted to apply agent technology to manufacturing enterprise integration, enterprise collaboration (including supply chain management and virtual enterprises), manufacturing process planning and scheduling, shop floor control, and to holonic manufacturing as an implementation methodology. This paper provides an update review on the recent achievements in these areas, and discusses some key issues in implementing agent-based manufacturing systems such as agent encapsulation, agent organization, agent coordination and negotiation, system dynamics, learning, optimization, security and privacy, tools and standards.     

Enhancing Workflow Management in the Manufacturing Information System of a Small-Medium Enterprise: An Agent-Based Approach

One shortcoming of manufacturing information systems is their inability to integrate and to enhance different information and related functionalities, such as those found in workflow management. Software agents, once added as a new software layer to an existing system, can overcome this problem. Of the fields applying software agent technology, manufacturing, and especially workflow management, is perhaps the most promising for the development of distributed systems. This potential is reflected in the several ongoing research projects that focus on workflow management and share the final aim of integrating and coordinating plant and business activities.The approach presented in this paper can be regarded as an agent-based architecture applied to an existing workflow management system in order to manage new functionalities, e.g., customer relationship management in electronic commerce. This approach is particularly suitable for small medium enterprises with simple manufacturing information requirements and simple or nonexistent computer based workflow management.The model and some important features of system implementation, such as the knowledge modeling graphical user interface, the communication model, and a simple practical example of the architecture's use, are described.     

Modeling of agile intelligent manufacturing-oriented production scheduling system

Agile intelligent manufacturing is one of the new manufacturing paradigms that adapt to the fierce globalizing market competition and meet the survival needs of the enterprises, in which the management and control of the production system have surpassed the scope of individual enterprise and embodied some new features including complexity, dynamicity, distributivity, and compatibility. The agile intelligent manufacturing paradigm calls for a production scheduling system that can support the cooperation among various production sectors, the distribution of various resources to achieve rational organization, scheduling and management of production activities. This paper uses multi-agents technology to build an agile intelligent manufacturing-oriented production scheduling system. Using the hybrid modeling method, the resources and functions of production system are encapsulated, and the agent-based production system model is established. A production scheduling-oriented multi-agents architecture is constructed and a multi-agents reference model is given in this paper.     

ManufAg: a multi-agent-system framework for production control of complex manufacturing systems

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.     

Agent技术研究现状及其在企业集成中的应用

1.引言分布式人工智能(DAI)的研究和网络化分布环境的普及推动了Agent的理论、技术,特别是多Agent的理论、技术的发展,因为它为分布开放系统的分析、设计和实现提供了一个崭新的途径。自从Minsky在1986年出版的“思维的社会”中提出Agent的概念后,智能代理技术随着计算机科学的发展     

Intelligent products: The grace experience

Product intelligence is a new industrial manufacturing control paradigm aligned with the context of cyber-physical systems and addressing the current requirements of flexibility, reconfigurability and responsiveness. This paradigm introduces benefits in terms of improvement of the entire product׳s life-cycle, and particularly the product quality and customization, aiming the customer satisfaction. This paper presents an implementation of a system of intelligent products, developed under the scope of the GRACE project, where an agent-based solution was deployed in a factory plant producing laundry washing machines. The achieved results show an increase of the production and energy efficiency, an increase of the product quality and customization, as well as a reduction of the scrap costs.     

自动化科学与技术发展方向

本文结合中国自动化科学与技术的发展状况和中国绝大多数大学设有自动化专业的现状,借鉴自动化科学与技术发展历程中的成功经验,结合国家社会经济发展和国家安全对自动化系统的未来需求,以生产制造系统、重要运载工具和人参与的信息物理系统为主要对象,以自动化系统的发展方向—智能自主控制系统、智能优化决策系统和智能优化决策与控制一体化系统的愿景功能为目标,以研究实现愿景功能的建模、控制与优化新算法和新的自动化系统的设计方法和实现技术以及结合重大应用领域开展的应用研究为主线,提出了自动化科学与技术的发展方向,并结合新兴应用领域对自动化科学与技术的需求与挑战,提出了未来自动化科学与技术的发展方向.     

基于人机一体的混合智能创新设计

如何应用智能技术进行创新设计是制造领域的研究热点。该文在分析和总结当前智能设计方法的基础上,将进化思想引入创新设计过程,以人类设计师为主导,将改进遗传算法与基于实例推理的产品设计方法结合起来,集成人机智能进行产品设计,实现了基于人机一体的混合智能创新设计。     

Multiagent framework for lean manufacturing

We have developed the manufacturing agent-based emulation system as an open framework for design and analysis of discrete manufacturing systems. MABES currently supports the transition from traditional to lean manufacturing in two major functions: analysis of alternative agent-based scheduling and control approaches that can be implemented across the extended enterprise; and real-time collaboration of design teams during manufacturing line design and analysis stages. MABES bases its support for these functions on two system paradigms: distributed agents and synchronous collaboration     

复杂系统基于Agent的建模与仿真设计模式及软件框架

提出了一种基于Agent的建模与仿真设计模式——ABMS设计模式，阐述了ABMS设计模式的主要内容。该模式的提出有利于在建模与仿真领域以及软件开发界之间建立关于ABMS的对话联系与交流。然后设计了一种包含仿真服务模型ServerAgent和仿真模型结构SimAgent的ABMS分布仿真框架，该框架将最大限度实现仿真重用，并将支持大规模的基于Agent的复杂系统的仿真，包括军事对抗以及天战系统的仿真。该框架部分借鉴了HLA的思想，其中的ServerAgent提供底层的通用仿真服务。     

Applications of artificial intelligence in intelligent manufacturing: a review

Based on research into the applications of artificial intelligence (AI) technology in the manufacturing industry in recent years, we analyze the rapid development of core technologies in the new era of ‘Internet plus AI’, which is triggering a great change in the models, means, and ecosystems of the manufacturing industry, as well as in the development of AI. We then propose new models, means, and forms of intelligent manufacturing, intelligent manufacturing system architecture, and intelligent manufacturing technology system, based on the integration of AI technology with information communications, manufacturing, and related product technology. Moreover, from the perspectives of intelligent manufacturing application technology, industry, and application demonstration, the current development in intelligent manufacturing is discussed. Finally, suggestions for the application of AI in intelligent manufacturing in China are presented.     

Agent-based intelligent system development for decision support in chemical process industry

This paper presents an agent-based intelligent system to support coordinate manufacturing execution and decision-making in chemical process industry. A multi-agent system (MAS) framework is developed to provide a flexible infrastructure for the integration of chemical process information and process models. The system comprise of a process knowledge base and a group of functional agents. Agents in the system can communicate and cooperate with each other to exchange and share information, and to achieve timely decisions in dealing with various scenarios in process operations and manufacturing management. Process simulation, artificial intelligent technique, rule-based decision supports are integrated in this system for process analysis, process monitoring, process performance prediction and operation suggestion. The implementation of this agent-based system was illustrated with two case studies, including one application in process monitoring and process performance prediction for a chemical process and one application in de-bottlenecking of a site utility system.     

Manufacturing in the Digital Age: Exploiting Information Technologies for Product Realization

Information technology (IT) is both the key enabler for future manufacturing enterprises and a transformer of organizations and markets. By reducing barriers to collaboration, compressing lead time, eliminating physical movement, and enriching decision-making, IT helps manufacturers achieve their goal of meeting customer needs better, quicker, and cheaper. By providing global reach and easy connectivity, information technology has fostered cooperation while increasing market competition, and heightened customer expectations. Advances in computer and communication technologies combined with rapid changes in organizations have created new opportunities for exploiting information technologies in the entire product realization process. This paper explores these opportunities, and identifies promising directions for both basic and applied research. We first review important trends in organizations, markets, and information technologies—from increasing customer involvement and opportunistic organizational alliances to global reach and connectivity, enterprise integration, and virtualization. Adopting a process viewpoint of the product realization cycle, we translate these trends into high-impact IT applications in design and operations that offer rich potential for applied research and development. Underlying these applications are four broad classes of intelligent information processes—intelligent search, diagnosis and prognosis, collaboration, coordination, and negotiation, and understanding and learning. And, software agents provide an ideal platform to implement these processes. We briefly review developments in these basic research fields, and identify necessary scientific advances that are most important from the manufacturing perspective. Our goal is to synthesize streams of thought from many related disciplines in engineering, science, and management, and develop a framework for examining how information technologies can facilitate and influence manufacturing.     

An object-oriented constraints-based system for concurrent product development

This research work aims to develop an intelligent constraint-based system that enables designers to consider at the early stages of the design process all activities associated with product’s life cycle. One of the most important aspects of these activities is the evaluation and optimisation of manufacturing processes that require various type of information from the different aspects of product’s life cycle. This research article discusses the development of a prototype system for manufacturing process optimisation using a combination of both mathematical methods and constraint-programming techniques. This approach enables designers to evaluate and optimise feasible manufacturing processes in a consistent manner as early as possible during the design process. This helps in avoiding unexpected design iterations that wastage a great amount of time and effort, leading to longer lead-time. The development process has passed through the five major stages: Firstly, an intelligent constraint-based design system for concurrent product and process design has been developed. Secondly, a manufacturing process optimisation module has been constructed. Thirdly, the product features, processes, cost, time and constraints to be used for carrying out various design tasks has been represented in the format of constraints, frames, objects, and rules. Fourthly, the process optimisation and evaluation rules for the selection of feasible processes for complex features, and finally, the information management system that ensures consistency in information exchange and decision making activities have been developed.     

Development of an agent-based system for manufacturing control and coordination with ontology and RFID technology

Integrating physical objects with the corresponding enterprise applications any time any where is the essential issue for a real-time enterprise. This study proposes a multi-agent system framework called agent-based manufacturing control and coordination (AMCC) system, a agent-based framework using ontology and RFID technology to monitor and control dynamic production flows and also to improve the traceability and visibility of mass customization manufacturing processes. The capabilities offered by multi-agent systems to respond to RFID events in real-time and a broad class of agent design and coordination issue regarding just in time (JIT) and just in sequence (JIS) manufacturing processes are also exploited in this study. To validate the proposed framework, case study of a bicycle manufacturing company is used to demonstrate how the proposed framework can benefit its JIT production. Finally, an example prototype system is implemented to demonstrate the concept of the proposed framework.     

Research on the relations between cognition and intelligent transformation of executive teams in small and medium-sized manufacturing enterprises

Intelligent manufacturing is crucial to realize high-quality development and competitiveness improvement of the manufacturing industry. This study takes team cognition as the research perspective, seeks the antecedent variables and theoretical application scenarios of the intelligent transformation of small and medium-sized manufacturing enterprises, follows the theoretical paradigm of “cognition-behavior-effect” and contingency analysis logic of environmental factors to construct a relationship model that includes the cognitive complexity, cognitive flexibility, bricolage behavior and intelligent transformation of the executive team. An empirical research on the relationship of core variables are conducted based on the survey data from 195 small and medium-sized manufacturing companies. The results show that: bricolage behavior plays a partial mediating role between cognitive complexity and intelligent transformation and a complete mediating role between cognitive flexibility and intelligent transformation; cognitive complexity and cognitive flexibility positively regulate each other’s relationship between the bricolage behavior and intelligent transformation. The research reveals an important question of “how small and medium-sized manufacturing executives’ team cognition induces bricolage behavior and promotes intelligent transformation”, with a view to further improving team cognition theory and bricolage theory, thus providing valuable suggestions on breaking resource constraints and pursuing transformation and upgrading for small and medium-sized manufacturing companies.     

Agent-Based Supply Chain Integration

Supply chain management represents a critical competency in today's fast-paced, global business environment. However, in the current transition from EDI to Web-based supply chain technologies, much of the capability for process integration is being lost. And the integration of buyer and seller supply chain processes is critical for speed and responsiveness in today's hypercompetitive product and service markets. Intelligent agent technology offers the potential to overcome many limitations of current supply chain technologies. This paper presents intelligent supply chain agents that represent and autonomously conduct business on behalf of product users, buyers and vendors. We classify and present numerous extant agent applications and extend a technological framework to compare and contrast intelligent agents with other classes of information technology. We then describe an agent-based supply chain process design, along with its developmental techniques, and the structure and behavior of an agent federation used for integration in a major enterprise. We present results of this exploratory research in terms of technical feasibility and process performance in the enterprise context. This work is novel in that it integrates process-level knowledge from operational enterprises with distributed agent technologies. And it makes a contribution by demonstrating how agent-based supply chain integration can be effected along a large-scale, operational, inter-organizational process.