Multi-agent systems will drive field of artificial intelligence digital application, using principles of component-based engineering, distributed decision making, parallel and distributed computing, autonomous computing, and advanced methods of interoperable/Integrate.
Operation of an agent-based system is based on interactions of autonomous, optionally self-interested, and loosely coupled entities – agents.
Processes characterised by materiel decomposition or possible computation distribution, can be solved by multi-agent systems very well. Moreover, the multi-agent system offers superb run-time digital integration capacity and changing reconfiguration, and autonomous delegation abilities.
There are several typical application areas of the agent technologies relating to product manufacturing. In production must solve highly complex planning problems need to control quickly changing, unpredictable and unstable processes.
In production there is also potential for agent-based diagnostics, repair, reconfiguration and replanning. In the domain of virtual organisations and supply chain processes, there are requirements for forming business alliances, planning long-term/short-term cooperation deals, including reconfigured supply chains.
Here we also can use multiple agent technologies for agents’ self-unique info capacity maintenance and specification of service interoperability across the supply chain. In the domain of digital network-based business agents, technologies can be used for shopping, information retrieval and searching, remote access to information and remote system control.
Another important application domain is logistics. Multi-agent systems can be used for directing production and materiel transit/handling, optimal planning and scheduling, especially in cargo transit and, military manoeuvres, etc. There is a nice match of the agent technologies with mobile operators networks, also simulation and predication of alarm situations, prevention to overload and intrusion detection.
For production support, creating an agent certification process can be successfully used for an integration of computing equipment already existing in the enterprise. Existing facilities can be extended by newly designed agents for planning, info transfer and digital visualisation.
For physically distributed production units, it's advantageous to decompose and distribute the planning problem. System can utilise established predictive digital tools for distributed planning and replanning.
Agents usually form local plans, optimise them local and later merge them e.g. by negotiation and selecting options. Another advantage of agent-based approach is its ability to process relevant production stats distributed across the entire enterprise or supply chain.
The classical approach when info collected and processed centrally is difficult especially when batches are voluminous and change frequently. Distributed approach allows proactive processing at the place of their origin and to exchange only necessary results.
The agent-based technology certainly does not provide an uncomplicated solution of planning problems. However, the concept allows integration of heavy-duty artificial intelligence problem solvers--such as constraint satisfaction systems and linear programming tools by its transformation into specialised agents.
Multi-agent solutions exist for low-level scheduling or control systems as well as product-configuration and quote phases to be used for short- and long-term production planning and supply chain administration.
Multi-agent systems on intra-enterprise level and extra-enterprise level are independent in the digital population point of view. Agents used on intra-enterprise level are operating inside an enterprise represents many units or processes in the unit. On extra-enterprise level, whole unit is represented by a single agent, providing all abilities and services, available in the company.
If agents on both levels are used, a special agent can exists that bind both levels together. For digital application purposes, both levels can be modeled together to study and improve their abilities.
Multi agent systems can be used also for a digital simulation and modeling of the production process or the supply chain, where they easily simulate an independence of involved parts. These tools can help to answer non-trivial tasks – how changes in single component will affect the production process or supply chain as a whole.
Multi-agent systems are robust and provide easy integration of digital behaviour with existing computing systems. Agents technologies are suitable for domains with the following properties:
1. Domains exist for applying of multi-agent systems in production support
2. Intra-enterprise production planning
3. Extra-enterprise production planning
4. Production simulation.
5. Highly complex systems to be controlled
6. Distributed information not available centrally
7. Domains with quickly changing scenarios and problem specification
8. High number of heterogeneous systems to be openly integrated
9. Cooperation of independent units
10. Coordination of virtual organisation.
Top 10 Situational Class Identification of Virtual Reality Enterprises
To support function of Virtual Enterprise-- independent of Virtual Enterprise size there is a need for a Virtual Enterprise coordinator. to monitor distributed business process Job Status and comparing it to Virtual Enterprise plans as described in the contracts.
In the case that an enterprise fails to perform its duties, the Virtual Enterprise must be reconfigured to replace the failing enterprise with another one. To support this functionality it is nice to have a distributed digital process plan/model tool to allow for re-planning and re-scheduling of business processes.
1. Typical virtual enterprise to include large scale engineering systems involved in system build-- emphasis is put on operation of virtual enterprise and on the support for business process definition and supervision.
2. Network topology situations show variable characteristics some enterprises can join or leave the alliance according to the phases of the business process or other market factors.
3. Duration of some alliances of virtual enterprises established towards a single business opportunity, and are dissolved at the end of such process
4. Many sectors have established supply chains with an almost fixed structure-- little variation in terms of suppliers or clients during the virtual enterprise life cycle consider temporary interaction with non-member enterprises such as occasional suppliers
5. Supporting infrastructure must handle many virtual enterprise participation spaces and cope with strict cooperation and information visibility rules, to preserve the requirements of every individual enterprise
6. Dominant company defines "the rules of the game" and imposes its own standards on others in terms of business process models, information exchange mechanisms and access rights
7. Different organisation can be found in some supply chains, without a dominant company so nodes cooperate on an equal basis, preserving their autonomy, but joining their core competencies.
8. Once successful alliance is formed, companies may realise the mutual benefits of joint control of resources and skills tends to create joint coordination structure
9. Visibility scope related to the topology and coordination i.e., how far, along the network can one node see the virtual enterprise configuration like direct neighbors ie, suppliers, clients
10. Monitoring of order fulfillment, planning, scheduling, workload distribution are examples of advanced task supervision and virtual enterprise coordination to include extensive visibility scope agreed in enforced contracts among all members
Top 10 Work Plan Composition Steps from Task Force on Architectures for Virtual Enterprise Integration
Integration of virtual enterprises must be developed and their use must be populated through examples and application experiences. The objective is to develop and validate a step forward in the state of the art of Digital Architectures for Enterprise Integration.
First objective consists of architecture section selection to describe and present all the necessary activities to establish, carry out and complete an enterprise integration programme for any kind of enterprise.
Requirements and components are put together by digital standards teams. Any kind of proposal for an enterprise integration reference architecture can be evaluated under certification criteria.
Although established architectures have many good points, all these architectures can be improved, since they have not completely generated the necessary digital modeling techniques and adequate execution tools for the different kinds of enterprises.
One particular architecture has been focused in the problem of virtual production enterprise integration
1. Create methods describes whole life cycle of a virtual production enterprise, including the design transactions among potential partners as part of the strategic activity.
2. Establish set of Reference Models to allow the representation of virtual production relationships to include design system teams, the operational business process and external constraints
3. Stand up performance measurement systems to help in assessment, decision-making and control of the production virtual organisation.
4. Utilise computer engineering tools to solve specialised problems of the production business.
5. Employ Information Infrastructure model to support all the Virtual production Enterprise activities.
6. Build existing complementary approaches in only one architecture.
7. Improve result architecture incorporating new techniques, methods, models and templates.
8. Validate usability and application, carry out real enterprise integration projects, mostly in sectors with small and medium-sized enterprises
9. Organise knowledge and experience obtained in primary architecture
10. Develop particular architectures and specialised tools focus on necessities of every type of enterprise activity.
Top 10 Production Planning System Based on Major Job Site Work Activities
All production activities can be separated into planning and control of daily activities being digitally structured along with work objects.
Digital behavioural structure of production order and information generation based on grouping of planning and control of daily activity aspect. Each item is lined up with considered activities, objects and each corresponding subsystems.
Digital framework for production work flow system is designed where each team activities and those objects are connected with administration functions.
1. Make priority Job Site planning composed of facility layout/process outlines
2. Determine production along with build capacity based on facility layout
3. Utilise Work space to form plan for Job Site specification assign based on process plan
4. Generate work order description by input of block design metrics
5. Select production techniques process planning
6. Design product work sequence parameters
7. Estimate lead time of each production process by control of available resources/capacity.
8. Assign schedule plan of production strategy and materiel procurement of Job Site
9. Create short-term and mid-term schedule to consider available resources
10. Assess Production volume of each Product and estimated Labour
Top 10 Limitations of Engineering Digital Informatics for Design/Use of Structures/Systems
Many weapons system products have lifecycles spanning multiple decades e.g., aircraft, ships, power generation equipment with design repositories and digital product lifecycle systems models readable most of the time.
Digital product models can have longer lifespan than information formats, application and computing platforms used to create the model. And info must be writable as well as readable if a digital product model, or its supporting information, needs editing at some point during the product lifecycle.
Engineering informatics facilitates practice of engineering to achieve military objectives supporting semantic codification, organisation, exchange, sharing, decision-making, storage, and retrieval of digital objects characterising the multi-disciplinary domain of engineering.
It is critical for engineers that digital models and systems they build today be extensible and reusable by subsequent generations of tech workers.
Many difficult problems must be addressed, since it requires combining many different types of technologies, e.g., information science, product engineering, and other engineering specialties
Even without addressing issues specific to engineering, the general problem of long-term digital preservation has several important issues to address.
1. Complex and open-ended
2. Long-term archiving requirements
3. Cost/benefit model to rationalise archiving
4. Establishing formal standards
5. Building application domains
6. Long-term retention of knowledge
7. Efficiency of archival procedures
8. Definition of policy guidelines
9. Metrics and archival method/protocol
10. Institutional support for archiving