In the context of the growing complexity and uncertainty of the modern economy, the industry of complete sets of electrical equipment has seen a significant decline in profits and increased competition between enterprises. In some cases, none of the enterprises in the market of complete sets of electrical equipment is often able to fully fulfill a large order from Gazprom or LUKOIL (for example, for the supply of complete electrical substations), which requires a wide range of purchased and manufactured products. The new complex task that arises is to automate the process of forming chains of cooperation of such enterprises, which would be carried out in real time at the very moment of forming a request from a large customer, taking into account the current load, competencies, resource capacities and limitations of each enterprise and the possibilities of their cooperation.
Multi-agent systems (MAS) have been an area of high expectations of the industrial IT community. However, in reality, these expectations are still not met and, in practice, the industry very rarely uses the MAS design methodologies, technologies, and software tools despite the appearance of many new classes of applications for which the MAS paradigm could be the perfect match. This paper analyzes the barriers and trends of the mismatch between the recent industrial anticipations and the real state of the practical use of MAS. It identifies engineering problems with very little re-use of code that currently stops economics of scale and impedes the extensive industrial MAS deployment and the ways to overcome them.
For over a quarter of a century, multi-agent systems havebeenconsideredasoneofthemostpromisingtechnologies for conceptualization, software development and implementation of Artificial Intelligence (AI) solutions. Not so widely but multi-agent technology is also considered as a way of designing complex adaptive systems based on bio-inspired principles of selforganization and evolution. However, in practice, the industry rarely uses multi-agent technology, despite the appearance of new classes of applications for which it is the perfect match, for example, smart cyber-physical systems with digital twins of controlled objects. The paper analyzes the recent anticipations and real achievements in the practical use of multi-agent systems at the industrylevel. Italsoidentifiestheengineeringproblems thatcurrentlyimpedetheextensiveindustrialimplementation of multi-agent systems and technologies as well asthewaystoovercomethem. Finally,prospectsfordevelopment of these technologies are evaluated up to the level of industrial implementation.
In this paper, Smart City is described as a live and constantly developing complex adaptive system operating in an uncertain environment with many participants and actors involved. The vision of the “Smart City 5.0” concept as an ecosystem of smart services based on multi-agent technology is presented. It is characterized by the cooperation of Artificial Intelligence systems and humans, and can harmoniously balance all spheres of life and contradictory interests of different city actors. In this concept, each smart service is presented by an autonomous agent. They can compete or cooperate with each other through a service bus and interact both vertically and horizontally on the basis of specialized protocols. Top-level services can be constructed as autonomous multi-agent systems of a lower level, where an agent can recursively reveal a new service for itself. The paper describes the design principals and the general architecture of the digital platform including the basic agent of smart service, the architecture and basic principles of smart city ontologies and knowledge base. The paper shows how the platform can support the decisionmaking life cycle for managing any urban object and the adaptive behaviour of Smart City 5.0 is compared with the fixed scenarios Smart City 4.0.
The paper proposes conceptual model of advanced digital platform for adaptive management of enterprises within the next generation of digital economy in the upcoming era of Industry 5.0. It analyzes existing digital platforms and their limitations due to their centralized and hierarchical management style supported. The paper considers the concept of digital ecosystem as an open, distributed, self-organizing “system of systems” of smart services capable of coordinating decisions and automatically resolving conflicts through multi-party negotiations. It proposes classification of services to be provided by the introduced advanced digital platform and describes their functions. It substantiates the leading role of multi-agent systems as the basic software architecture and technology for developing applications managed by the introduced digital platform. The paper results are applicable to many modern industrial enterprises.
In the paper, the new vision of “Smart City 5.0” is presented. It is based on a previously developed model of Smart City 4.0 and implementing the concept of the complex adaptive system for balancing conflict interests of different city actors. These actors can include business, transport, energy and water supply providers, entertainment and other services and can be unified based on resource and demand model. The paper describes the general principals, functionality and the architecture of the digital multi-agent platform for creating eco-system of “Smart City 5.0”. It is designed as holonic p2p network of smart services and technological components for supporting demand-resource relations. It is shown that in proposed eco-system smart services can interact both vertically and horizontally supporting competition and cooperation behavior on the basis of specialized protocols of p2p network. In the future, each smart service is considered as an autonomous cyber-physical multi-agent system which can be decomposed on a lower level of smaller services recursively. The first prototypes of smart services and their interaction are presented, the next steps for future research work are outlined.
Complexity of modern resource management is analyzed and related with a number of decision makers, high variety of individual criteria, preferences and constraints, interdependency of all operations, etc. The overview of existing methods and tools of Enterprise Resource Planning is given and key requirements for resource management are specified. The concept of autonomous Artificial Intelligence (AI) systems for adaptive resource management based on multi-agent technology is discussed. Multi-agent model of virtual market and method for solving conflicts and finding consensus for adaptive resource management are presented. Functionality and architecture of autonomous AI systems for adaptive resource management and the approach for measuring adaptive intelligence and autonomy level in these systems are considered. Results of delivery of autonomous AI solutions for managing trucks and factories, mobile teams, supply chains, aerospace and railways are presented. Considerable increase of enterprise resources efficiency is shown. Lessons learned from industry applications are formulated and future developments of AI for solving extremely complex problems of adaptive resource management are outlined.
Nowadays the world is surviving the fourth industrial revolution named Industry 4.0, which combines physical world of real things with their “virtual twins”. The man with his intellect, creativity and will lies beyond this ideology. Now the image of a new paradigm of Industry 5.0 could be seen. It involves the penetration of Artificial Intelligence in man’s common life, their “cooperation” with the aim of enhancing the man capacity and the return of the man at the “Centre of the Universe”. The paper outlines modern technologies – from IoT up to emergent intelligence, being developed in organizations where authors work. The convergence of these technologies, according to our minds, will provide the transformation from Industry 4.0 to Industry 5.0.
This paper analyses the new requirements for real time resource management systems based on multi-agent technology. It shows the growing demand for developing autonomous systems which combines resource allocation, scheduling, optimization, communication with users and control in one cycle and can respond rapidly to unexpected events in real time. To solve the problem the cyber-physical multi-agent systems are considered. The paper also dwells on the new impact which such systems bring into design of modern systems on the way from smart Internet of things – to new organizations and ways of user motivation.
A thorough analysis of data on aircraft lifecycle revealed inadequacy of current lifecycle management methods in the face of increased complexity of the Internet-based global market. A new method for managing lifecycle has been developed by authors and their teams using concepts and principles of the emerging complexity science with the aim of reducing lead times and costs. Centralised control has been replaced with distributed decision-making empowering all lifecycle stakeholders. The solution described in this paper is the first of its kind and it represents a genuine advance in knowledge, which leads to considerable reduction in design/production lead times and decrease in the lifecycle cost. The method has been validated in a variety of applications.
In this paper modern methods of scheduling and resource optimization based on the holonic approach and principles of “Swarm Intelligence” are considered. The developed classes of holonic agents and method of adaptive real time scheduling where every agent is connected with individual satisfaction function by the set of criteria and bonus/penalty function are discussed. In this method the plan is considered as a un-stable equilibrium (consensus) of agents interests in dynamically self-organized network of demands and supply agents. The self-organization of plan demonstrates a “swarm intelligence” by spontaneous autocatalitical reactions and other not-linear behaviours. It is shown that multi-agent technology provides a generic framework for developing and researching various concepts of “Swarm Intelligence” for real time adaptive event-driving scheduling and optimization. The main result of research is the developed approach to evaluate the adaptability of “Swarm Intelligence” by measuring improve of value and transition time from one to another unstable state in case of disruptive events processing. Measuring adaptability helps to manage self-organized systems and provide better quality and efficiency of real time scheduling and optimization. This approach is under implementation in multi-agent platform for adaptive resource scheduling and optimization. The results of first experiments are presented and future steps of research are discussed.
The modern problem of real-time resource management to increase enterprise efficiency is considered. A new look at the dynamic self-organizing processes based on multi-agent technologies in building and revising schedules by events in real time is suggested. Schedule is considered as a flexible network of operations of demand and resource agents. This schedule is formed during the interactions of basic agent classes that set and break the dynamic links between each other, depending on the events and changing situation in the real world.
A thermodynamic model of demand–resource network (DRN) dynamics is introduced. There is a similarity to Ilya Prigogine’s non-linear thermodynamics theory which allows us to explain the phenomenon of unstable equilibrium emergence, order and chaos, catastrophes, bifurcations and other non-linear events that are significant to the self-organizing processes control in multi-agent systems (MAS).
Managing Complexity. WITPress. — 2014
Managing Complexity is the first book that clearly defines the concept of Complexity, explains how Complexity can be measured and tuned. The thesis of the book is that complexity of the environment in which we work and live offers new opportunities and that the best strategy for surviving and prospering under conditions of complexity is to develop adaptability to perpetually changing conditions.An effective method for designing adaptability into business processes using multi-agent technology is presented and illustrated by several extensive examples, including adaptive, real-time scheduling of taxis, see-going tankers, road transport, supply chains, railway trains, production processes and swarms of small space satellites. Additional case studies include adaptive servicing of the International Space Station; adaptive processing of design changes of large structures such as wings of the largest airliner in the world; dynamic data mining, knowledge discovery and distributed semantic processing.