A Hybrid Agent-Oriented Infrastructure
`for Modeling Manufacturing Enterprises
`Weiming Shen and Douglas H. Norrie
`Division of Manufacturing Engineering, The University of Calgary
`2500 University Dr. NW, Calgary, AB, Canada T2N 1N4
`E-mail: [wshen | norrie]@enme.ucalgary.ca
`ABSTRACT
`Manufacturing enterprises are now moving towards open architectures for integrating their
`activities with those of their suppliers and customers within wide supply chain networks.
`Traditional knowledge engineering approaches with large scale or very large scale knowledge
`bases are not suited for such widely distributed systems. Agent-based technology provides a
`natural way for resolving this problem. This paper presents a hybrid agent-oriented infrastructure
`for modeling manufacturing enterprises so as to integrate design, planning, scheduling,
`simulation, execution, material supply, and marketing services into a distributed intelligent open
`environment. In this paper, we discuss the requirements for next generation of manufacturing
`enterprises, and describe the main features of the proposed general infrastructure and the
`functions of its components. A machine-centered dynamic scheduling and rescheduling
`mechanism is then detailed and a prototype implementation is presented.
`Keywords: Enterprise integration, knowledge engineering, distributed manufacturing systems,
`manufacturing scheduling, agent, mediator.
`1 INTRODUCTION
`Manufacturing enterprises are now moving towards open architectures for integrating their
`activities with those of their suppliers and customers within wide supply chain networks. To
`compete effectively in today's markets, manufacturers must be able to interact with customers,
`suppliers, and services rapidly and inexpensively. Traditional knowledge engineering approaches
`with large scale or very large scale knowledge bases are inappropriate because of the highly
`distributed nature of the systems. Agent-based technology derived from Distributed Artificial
`Intelligence provides a natural way for resolving this problem.
`At The University of Calgary, we are now working on the MetaMorph II project whose enhanced
`capabilities will embody lessons learned from our previous research work. This paper describes
`our ongoing MetaMorph II project and presents its prototype implementation. The rest of this
`paper is organized as follows: Section 2 discusses the requirements for next generation of
`manufacturing enterprises; Section 3
`introduces agent-based
`technology for modeling
`manufacturing enterprises; Section 4 presents our MetaMorph project; Section 5 describes the
`dynamic scheduling and rescheduling mechanisms developed for MetaMorph II; Section 6
`presents the prototype implementation; Section 7 gives concluding remarks and perspectives.
`2 REQUIREMENTS FOR NEXT GENERATION OF MANUFACTURING
`ENTERPRISES
`Manufacturing strategy has shifted rapidly over the past ten years to support global
`
`1
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 1
`
`

`

`competitiveness, new product innovation and introduction, and rapid market responsiveness. The
`next generation of manufacturing systems will be time oriented versus cost or even quality
`based. Such manufacturing systems should meet following fundamental requirements:
`Enterprise Integration
`In order to support its global competitiveness and rapid market responsiveness, an individual
`manufacturing enterprise has to be integrated with its related management systems (e.g.,
`purchasing, orders, design, production, planning, control, transport, resources, personnel,
`materials, quality, etc.) which are,
`in general, heterogeneous software and hardware
`environments. Such integration may be realized via tactical planning systems that rely heavily on
`distributed knowledge-based systems to link demand management directly to resource and
`capacity planning.
`Cooperation
`Manufacturing enterprises have to fully cooperate with their suppliers and customers for material
`supply, parts fabrication, final product commercialization, and so on. Such cooperation should be
`in an efficient and quick-response manner.
`In a cooperative system, dynamic chains of events are embedded in concurrent information
`processes. Requirements imposed by customer orders, managerial decisions, and design stages
`are integrated with the production planning and resource allocation tasks in a complex
`framework that incorporates high-level decisions into the planning activities. This is essentially a
`cooperative, concurrent information-processing environment. Cooperation is an imperative
`requirement for any complete functional model for advanced manufacturing systems.
`Integration of humans with software and hardware
`People and computers need to be integrated to work collectively at various stages of the product
`development, with access to required knowledge and information. Heterogeneous sources of
`information must be integrated to support these needs and enhance the decision capabilities of
`the system. Bi-directional communication environments are required to allow effective, quick
`communication between human and computers to facilitate their cooperation.
`Agility
`Economic globalization and expanding market expectations are rapidly transforming the
`environment for manufacturing. Considerable attention must be given to reducing product cycles
`to be able to respond more quickly to customer desires. In this new scale of economic
`transformation, corporations are progressively reorienting their strategies to expand their share of
`the market and to integrate “Agile” manufacturing into their production facilities.
`Agile manufacturing is the ability to adapt in a manufacturing environment of continuous and
`unanticipated change and thus is a key component in manufacturing strategies for global
`competition. To achieve agility, manufacturing facilities must be able to establish convenient
`associations with heterogeneous partners. Ideally, partners are contracted with “on the fly” only
`for the time required to complete specific tasks. This type of interaction can also be used to plan
`long-term strategies. Agility will bring greater flexibility to the manufacturing organization
`without incurring large or diverted industrial investments.
`Scalability
`Scalability is an important property for advanced manufacturing systems. Scalability means that
`more resources can be incorporated into the organization as required. This property should be
`available at any working node in the system and at any level within the nodes. Expansion of
`
`2
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 2
`
`

`

`resources should be possible without disrupting organizational links previously established.
`To identify and incorporate new components into the system, organizational knowledge
`registries are required. When new physical components arrive in the system, representative
`entities are created to act as counterparts to the components throughout their life cycles. The
`ability to add new components incrementally allows the system to respond flexibly to a wide
`variety of requests. For example, the system might dynamically add increased intelligence and
`manufacturing capacity to supply a rapidly expanding market or reduce capacity to adjust
`downwards during low demand periods. When physical components are removed from the
`system for maintenance or other reasons, the listing of these components is removed from the
`system registry. Robust registration mechanisms are needed to provide ongoing integration of
`new components or the removal of existing ones
`Dynamic reconfiguration
`Both human beings and artificial entities in manufacturing systems need to be more alert to
`environmental changes. Every stage in manufacturing planning is affected by dynamic variations
`coming from either internal or external sources. In conventional manufacturing systems, the
`input from customers triggers a sequence of events, starting with planning operations. At this
`level, requests are processed according to preestablished stages (which include specification of
`product design, material management, manufacturing capacity planning and availability, and
`preparation of production costs). The planning process also triggers requirements for
`subcontracting external services.
`Conventionally, the planning process progressively advances through a series of sequential
`evaluations that correspond to system conditions from an earlier time. Therefore, any subsequent
`variations in the state of the environment can make these plans invalid. This type of sequential
`system thus becomes expensive, since there is a tendency to reuse computing resources for
`redundant and repetitive evaluations.
`Eventually, there is a transition from the planning process to a second major area of
`manufacturing control activity in which manufacturing plans and tasks are allocated execution
`times. More variations are introduced at this stage, which affect the stability of the system and its
`ability to execute plans according to schedule. This complex stage restricts the ability of the
`system to reconfigure to cope with dynamic and unforeseen changes.
`Expandability is possible in such conventional manufacturing systems, but it requires major
`reconfiguration of the system. Tightly coupled interconnections among the system’s existing
`components make adjusting each processing module to new component availability very tedious.
`Similarly, the removal of components implies a considerable readjustment of the system.
`Knowledge capitalization and distribution
`The efficient capture and distribution of knowledge pertaining to each aspect of the organization
`- finance, marketing, design, and manufacturing - coupled with its effective use, will result in
`startling advances in market research, product and process development, production planning
`and scheduling, and ultimately customer responsiveness.
`The more obvious problems in information processing observed in conventional manufacturing
`systems are highly centralized information management and production control, which
`corresponds to the need to maintain an overall system view in order to minimize costs (and so to
`win over more of the market). Centralized databases are commonly used to accumulate system
`information for establishing production plans and forecasting future requirements. Powerful
`centralized computers process large amounts of data to create production plans and schedules.
`Transactions among various resources are also forced to pass through a centralized control unit.
`
`3
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 3
`
`

`

`All activities in conventional manufacturing systems are limited by the accuracy and stability of
`the centralized processing components, yielding a fragile infrastructure.
`Although centralized and sequential information-processing systems have in the past minimized
`hardware and software costs, their central structure is not suited to the inherent distributed nature
`of concurrent information flow in agile manufacturing.
`Distribution of production knowledge will enhance system modularity and facilitate both
`integration and reconfiguration. The increased modularity reduces the complexity of organizing
`knowledge by maintaining knowledge locally. Information held locally can be processed
`concurrently, thus avoiding the limitations of sequential information processing
`Concurrent Engineering
`Ensuring the manufacturability of the product constitutes the first step in implementing
`concurrent engineering. Geometric and functional specifications, availability of raw materials,
`and the capability and availability of shop-floor resources each has a major influence on
`manufacturability. A design may be manufacturable under one combination of product
`requirements and shop-floor resources, but not under another. The selection and availability of
`stock material from which the part will be manufactured influences the number of intermediate
`steps required, and hence the production cost. The capability and availability of shop-floor
`resources impact the process plan to be used, and again the production cost. Thus, all of these
`aspects need to be considered simultaneously for effective concurrent engineering.
`3 AGENT-ORIENTED APPROACHES FOR MODELING
`MANUFACTURING ENTERPRISES
`3.1 Knowledge Sharing and Reuse
`From the perspective of knowledge intensive engineering, we can view all relevant aspects of an
`organization domain in terms of 'knowledge'. This applies to the structure and nature of the
`organization itself, the data used within different components of the organization and the flow of
`this data through the organization along with the value added to it during the execution of the
`organizational tasks. This knowledge exists in the form of data (factual assertions about the
`organization and its tasks) which may reside in existing information systems (such as databases),
`and in the form of specific 'business rules' applied to this data in order to carry out some function
`within the organization, either relating directly or indirectly to the tasks at hand. The focus of
`knowledge systems techniques is the explicit representation of this organizational knowledge in a
`form that is optimal for effective reasoning about the tasks in the organization, as well as for the
`representation of this information to those assigned the execution of line tasks (Davis and Oliff
`1988).
`Previously, large scale or very large scale knowledge bases have been often advocated for
`engineering applications including design, manufacturing, operations, and maintenance, because
`these activities require an extremely huge amounts of and various kinds of knowledge (Forbus
`1988). But, according to Tomiyama et al (1995), not only the quantity of knowledge but also the
`quality of knowledge in terms of sharability and reusability of knowledge is crucial. Knowledge
`intensive engineering aims at both the amount and flexibility of knowledge. A single knowledge
`base can make inferences in a particular circumstance but it may hard-fail. Therefore just having
`a large scale or very large scale knowledge base alone is not enough for modern manufacturing
`whose robustness and reliability are actually important.
`Knowledge in modern manufacturing must be well organized and should be able to be flexibly
`
`4
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 4
`
`

`

`applied to different kinds of applications. Figure 1 compares three possible different types of
`knowledge sharing architectures (Tomiyama et al 1995). Figure 1(a) depicts a situation with
`independent knowledge bases. In this case, the 'strength' of knowledge is just a sum of each of
`independent knowledge bases. Integrated knowledge bases can be represented as in Figure 1(b).
`Here, the knowledge bases can be applied to various situations and the 'strength' of knowledge is
`near maximum. However, this requires having a platform with a uniform language. The Cyc
`project (Lenat and Guha 1989) is an example of this approach. In Figure 1(c), independent
`knowledge bases can communicate and form an interoperable situation, although the 'strength' of
`knowledge might be weaker than that in Figure 1(b). The entire knowledge base is a federation
`or a set of loosely coupled intelligent agents. This approach has recently been used by projects
`like SHADE (McGuire et al 1993), PACT (Cutkosky et al 1993), DESIRE (Brazier and Treur
`1996), and DIDE (Shen and Barthès 1997).
`
`(c) interoperable knowledge bases
`(b) integrated knowledge bases
`(a) independent knowledge bases
`Figure 1. Knowledge sharing architectures (Tomiyama et al 1995)
`
`The second architecture (Figure 1(b)) may suffer from the lack of uniform knowledge
`representation. Unless carefully designed, the platform language cannot cover everything. The
`third architecture (Figure 1(c)) may overcome this problem, if the framework is abstract enough
`to incorporate the different types of ontology each agent uses. However, it does not completely
`avoid the problem, because communication among agents requires at least understanding what
`other agents are talking about.
`The same communication requirements justify the representation model used in SHADE
`(McGuire et al 1993). The model, called KIF (Knowledge Interchange Format) (Genesereth &
`Fikes 1992), is a machine-readable version of first order predicate calculus, with extensions to
`enhance expressiveness. KIF specifications define syntax and semantics; ontology defines the
`problem-specific vocabulary. Agents exchange sentences in KIF using the shared vocabulary.
`To support the sharing and reuse of formally represented knowledge among AI systems, it is
`useful to define the common vocabulary in which shared knowledge is represented (Patil et al
`1992). A specification of a representational vocabulary for a shared domain of discourse
`definitions of classes, relations, functions, and other objects is called an ontology (Gruber 1993).
`The need for a shared ontology is a direct result of the multidisciplinary nature of engineering.
`There are many different views of a design (function, performance, manufacturing), each with a
`different language. However, the various perspectives typically overlap, necessitating the sharing
`of information if design is to proceed concurrently and cooperatively. For information to be
`shared, there must be a commonly understood vocabulary. A detailed discussion on ontology can
`be found in (Gruber 1993). An application of ontology in enterprise modeling was proposed by
`Fox et al (1996).
`In design applications, it is necessary to represent knowledge at several levels: domain
`knowledge associated with the particular vocabulary used in the design domain, but also general
`
`5
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 5
`
`

`

`engineering knowledge (e.g., associated with the use of engineering units). The particular
`representation of the corresponding ontology requires special models, mechanisms, languages,
`and tools.
`In distributed systems, it must be possible to exchange knowledge among agents, even if agents
`work in different specialties. The required formalisms for exchanging knowledge have been
`studied in various projects, and several proposals exist, such as the knowledge representation
`languages like KIF (Genesereth & Fikes 1992), in connection with ontologies. The exchange of
`knowledge in most existing systems, however, is carried out among human designers, using
`electronic mail.
`3.2 Agent-Oriented Approaches for Modeling Manufacturing Enterprises
`The requirements described in Section 2 necessitate decentralized manufacturing facilities whose
`design, implementation, reconfiguration, and manufacturability allows the integration of
`production stages in a dynamic, collaborative network. Such facilities can be realized through
`agent-oriented approaches (Wooldridge and Jennings 1995) using knowledge sharing technology
`(Patil et al 1992). The following paragraphs briefly review several interesting projects in this
`domain.
`The SHADE project (McGuire et al 1993) was primarily concerned with the information-sharing
`aspect of concurrent engineering. Rather than attempting to model the design process, it provides
`a flexible infrastructure for anticipated knowledge-based, machine-mediated collaboration
`among disparate engineering tools. SHADE is distinct from other approaches in its emphasis on
`a distributed approach to engineering knowledge rather than a centralized model or knowledge
`base. That is, not only does SHADE avoid the requirement of physically centralized knowledge,
`but the modeling vocabulary is distributed as well, focusing knowledge representation on
`specific knowledge-sharing needs.
`PACT (Cutkosky et al 1993) was a landmark demonstration of both collaborative research
`efforts and agent-based technology. The agent interaction relies on shared concepts and
`terminology for communicating knowledge across disciplines, an interlingua for transferring
`knowledge among agents, and a communication and control language that enables agents to
`request information and services. This technology allows agents working on different aspects of
`a design to interact at the knowledge level, sharing and exchanging information about the design
`independent of the format in which the information is encoded internally.
`SHARE (Toye et al 1993) was concerned with developing open, heterogeneous, network-
`oriented environments for concurrent engineering. It used a wide range of information-exchange
`technologies to help engineers and designers collaborate in mechanical domains.
`Recently, PACT has been replaced by PACE (Palo Alto Collaborative Environment)
`[http://cdr.stanford.edu/PACE/] and SHARE by DSC
`(Design Space Colonization)
`[http://cdr.stanford.edu/DSC/].
`FIRST-LINK (Park et al 1994) was a system of semi-autonomous agents helping specialists to
`work on one aspect of the design problem. NEXT-LINK (Petrie et al 1994) was a continuation of
`the FIRST-LINK project for testing agent coordination. Process-Link (Goldmann 1996) followed
`on from Next-Link and provides for the integration, coordination, and project management of
`distributed interacting CAD tools and services in a large project.
`MADEFAST (Cutkosky et al 1996) is a DARPA DSO-sponsored project to demonstrate
`technologies developed under the ARPA MADE (Manufacturing Automation and Design
`Engineering) program. MADE is a DARPA DSO long term program for developing tools and
`technologies to provide cognitive support to the designer and allow an order of magnitude
`
`6
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 6
`
`

`

`increase in the explored alternatives in half the time it takes to explore a single alternative today.
`SiFA (Brown et al 1995), developed at Worcester Polytechnic, is intended to address the issues
`of patterns of interaction, communication, and conflict resolution. DIDE proposed to use
`autonomous cognitive agents for developing distributed intelligent design environments (Shen
`and Barthès 1997). In AARIA (Parunak et al 1997a), the manufacturing capabilities (e.g. people,
`machines, and parts) are encapsulated as autonomous agents. Each agent seamlessly
`interoperates with other agents in and outside its own factory. AARIA uses a mixture of heuristic
`scheduling techniques: forward/backward scheduling, simulation scheduling, and intelligent
`scheduling. Scheduling is performed by job, by resource, and by operation. Scheduling decisions
`are made to minimize costs over time and production quantities.
`Saad et al (1995) proposed a Production Reservation approach by using a bidding mechanism
`based on the Contract Net protocol to generate the production plan and schedule. Maturana et al
`(1996) described an integrated planning-and-scheduling approach combining subtasking and
`virtual clustering of agents with a modified Contract Net protocol. RAPPID (Responsible Agents
`for Product-Process Integrated Design) (Parunak et al 1997b) at the Industrial Technology
`Institute was intended to develop agent-based software tools and methods for using market place
`dynamics among members of a distributed design team to coordinate set-based design of a
`discrete manufactured product. AIMS (Park et al 1993) was envisioned to provide to the US an
`integrated industrial base able to rapidly respond, with highly customized solutions, to customer
`requirements of any magnitude, thus reinstating the US as the world leader in manufacturing.
`ADDYMS (Architecture for Distributed Dynamic Manufacturing Scheduling) proposed by
`Butler and Ohtsubo (1992) was a distributed architecture for dynamic scheduling in a
`manufacturing environment. Pan and Tenenbaum (1991) proposed a software Intelligent Agent
`(IA) framework for integrating people and computer systems in large, geographically dispersed
`manufacturing enterprises. This framework is based on the vision of a very large number (e.g. 10
`000) computerized assistants, known as Intelligent Agents (IAs). Human participants are
`encapsulated as Personal Assistants (PAs), a special type of IA.
`Roboam and Fox (1992) proposed an Enterprise Management Network (EMN) to support the
`integration of activities of the manufacturing enterprise throughout the production life cycle with
`six levels: (1) Network Layer provides for the definition of the network structure; (2) Data Layer
`provides for inter-node queries; (3) Information Layer provides for invisible access to
`information spread throughout the EMN; (4) Organization Layer provides the primitives and
`elements for distributed problem solving; (5) Coordination Layer provides protocols for
`coordinating the activities of EMN nodes; and (6) Market Layer provides protocols for
`coordinating organizations in a market environment.
`4 MetaMorph PROJECT
`4.1 MetaMorph I
`MetaMorph (now referred to MetaMorph I) (Maturana and Norrie 1996) is a multi-agent
`architecture for intelligent manufacturing developed at The University of Calgary. The
`architecture has been named MetaMorphic, since a primary characteristic is its changing form,
`structure, and activity as it dynamically adapts to emerging tasks and changing environment.
`In this particular type of federation organization, intelligent agents can link with mediator agents
`to find other agents in the environment. Additionally, mediator agents assume the role of system
`coordinators by promoting cooperation among intelligent agents and learning from the agents’
`behavior. Mediator agents provide system associations without interfering with lower-level
`
`7
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 7
`
`

`

`decisions unless critical situations occur. Mediator agents are able to expand their coordination
`capabilities to include mediation behaviors, which may be focused upon high-level policies to
`break the decision deadlocks. Mediation actions are performance-directed behaviors.
`Mediator agents can use brokering and recruiting communication mechanisms (Decker 1995) to
`find related agents for establishing collaborative subsystems (also called coordination clusters or
`virtual clusters) (see Figure 2). The brokering mechanism consists of receiving a request message
`from an intelligent agent, understanding the request, finding suitable receptors for the message,
`and broadcasting the message to the selected group of agents. The recruiting mechanism is a
`superset of the brokering mechanism, since it uses the brokering mechanism to match agents.
`However, once appropriate agents have been found, these agents can be directly linked. The
`mediator agent then can step out of the scene to let the agents proceed with the communication
`themselves. Both mechanisms have been used in MetaMorph I. To efficiently use these
`mechanisms, mediator agents need to have sufficient organizational knowledge to match agent
`requests with needed resources. Organizational knowledge at the mediator level is basically a list
`of agent-to-agent relationships that is dynamically enlarged.
`
`Agent
`
`Agent
`
`Mediator
`
`Mediator
`
`Agent
`
`Agent
`
`Agent
`
`Agent
`
`Agent
`
`Agent
`
`Agent
`
`Brokering Mechanism
`
`Recruiting Mechanism
`Figure 2. Brokering and Recruiting Mechanisms
`
`The brokering and recruiting mechanisms generate two relevant types of collaboration
`subsystems. The first corresponds to an indirect collaboration subgroup, since the requester agent
`does not need to know about the existence of other agents that temporarily match the queries.
`The second type is a direct collaboration subgroup, since the requester agent is informed about
`the presence and physical location of matching agents to continue with direct communication.
`One common activity for mediator agents involved in either type of collaboration is interpreting
`messages, decomposing tasks, and providing processing times for every new subtask. These
`capabilities make mediator agents very important elements in achieving the integration of
`dissimilar
`intelligent agents. Federation multi-agent architectures require a substantial
`commitment to supporting intelligent agent interoperability through mediator agents.
`In MetaMorph I (Maturana and Norrie 1996), mediators were used in a distributed decision-
`making support system for coordinating the activities of a multi-agent system. This coordination
`involves three main phases: (1) subtasking; (2) creation of virtual communities of agents
`(coordination clusters); and (3) execution of the processes imposed by the tasks. These phases
`are developed within the coordination clusters by distributed mediator agents together with other
`agents representing the physical devices. The coordination clusters are initialized through
`mediator agents, which can dynamically find and incorporate those other agents that can
`contribute to the task.
`
`8
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 8
`
`

`

`Another related project named ABCDE (Agent-Based Concurrent Design Environment) was
`developed in the same research group (Balasubramanian et al 1996). It was integrated with the
`MetaMorph I system through a Design Mediator. The ABCDE architecture includes an
`environment manager, feature agents, part agents, and CAD physical layers. ABCDE agents
`interact with the shop-floor mediator to obtain manufacturability assessments during the product
`design process. Human production managers may request manufacturability evaluations using
`either the CAD system or the design system interface.
`4.2 MetaMorph II
`Based on MetaMorph I, the MetaMorph II project started at the beginning of 1997. Its objective
`is to integrate design, planning, scheduling, simulation, execution, material supply, and
`marketing services into a distributed intelligent open environment. For this purpose, we propose
`a hybrid architecture combining and extending the architectures used in our previous projects
`MetaMorph I (Maturana and Norrie 1996), ABCDE (Balasubramanian et al 1996) and DIDE
`(Shen and Barthès 1995). In this hybrid architecture, the system is primarily organized at the
`highest level through 'subsystem' mediators (see Figure 3). Each subsystem is connected
`(integrated) to the system through a special mediator. Each subsystem itself can be an agent-
`based system (e.g., agent-based manufacturing scheduling system), or any other type of system
`like functional design system, knowledge-based material management system, and so on. Agents
`in a subsystem may also be autonomous agents at the subsystem level. Some of these agents may
`also be able to communicate directly with other subsystems or the agents in other subsystems.
`Manufacturing resource agents are coordinated by dynamically hierarchical mediators. For
`example, a shop floor Resource Mediator coordinates Machine Mediators, Tool Mediators,
`Worker Mediators, and so on (see Figure 4). A machine agent can also communicate and
`negotiate directly with a Worker Mediator, worker agents, a Tool Mediator and tool agents.
`
`Simulation
`Mediator 1
`
`Design
`Mediator 1
`
`Resource
`Mediator 1
`
`Resource
`Mediator 2
`
`Marketing
`Mediator
`
`Design
`Mediator 2
`
`Enterprise
`Mediator
`
`Simulation
`Mediator 2
`
`Execution
`Mediator 1
`
`Execution
`Mediator 2
`
`Marketing
`
`Design
`
`Planning & Scheduling
`
`Execution
`
`Figure 3. Functional Architecture of MetaMorph II
`
`Material
`Mediator
`
`Material
`Supply
`
`9
`
`Petitioner STMICROELECTRONICS, INC.,
`Ex. 1023, IPR2022-00681, Pg. 9
`
`

`

`Resource Mediator
`
`Machine Mediator
`
`Tool Mediator
`
`machine agents
`
`…
`
`Worker Mediator
`
`…
`
`…
`
`tool agents
`
`worker agents
`Figure 4. Organization of resource agents
`
`4.3 Characteristics of the MetaMorph II Architecture
`MetaMorph II is an extension of MetaMorph I in multiple dimensions (cf. Figure 3):
`1) Integration of Design and Manufacturing: Agent-based intelligent design systems will be
`integrated into the MetaMorph II. Some features and mechanisms used in the DIDE
`project and ABCDE project will be applied to develop this type of subsystem. Each such
`subsystem connects within MetaMorph II with a Design Mediator which serves as the
`coordinator of this subsystem and its only interface to the whole system. Several design
`systems can be connected to MetaMorph II simultaneously. Each design system may be
`an agent-based system or another type of design system.
`2) Extension to marketing: This will be realized by several easy-to-use interfaces for
`marketing engineers and end customers to request product information (performance,
`price, manufacturing period, etc), select a product, request modifications to a particular
`specification of a product, and send feedback to the enterprise.
`3) Integration of Material Supply and Management System: A Material Mediator will be
`developed to coordinate a special subsystem for material handling, supply, stock
`management etc.
`4) Improvement of the Simulation System: Simulation Mediators will be developed to carry
`out