Table of Contents Author Guidelines Submit a Manuscript

Recent Applications of Neurocomputing in Large-Scale Distributed Systems

Call for Papers

Neurocomputing is a research domain that deals with the development of models, techniques, and algorithms inspired by complex processes that take place in the human brain (e.g., knowledge acquisition through learning processes, communication between neurons, and information processing by neurons). Neurocomputing includes but is not limited to machine learning, computational learning theory, self-organizing systems, and neural networks; it can be successfully applied to solve various problems in optimization, classification and clustering, prediction, and pattern recognition.

With the recent technological advancements in the areas of IoT and distributed control, the research efforts are moving towards developing complex distributed systems able to administrate large-scale heterogeneous pervasive environments composed of sensors, actuators, smart devices, and agents (humans or software) which interact and collaborate. To offer support for the humans during their interactions and to allow them to concentrate on their tasks, it is essential for the distributed systems to become aware of their execution context and to be able to coordinate, optimize, and adapt their behavior to situation at hand. Nowadays, they are used in various domains ranging from energy and resources management to intelligent transportation, circular economy, e-health, and public safety.

Even though lately significant progress has been made in this area, the efficient management of distributed systems is still an open issue due to the concurrency of their main components, lack of global clock, and independent failure of components.

In this context, it has been noticed that neurocomputing provides many metaphors for designing decentralized self-organization, management, and optimization models, since the human neural system can optimally take decisions by acquisition of useful information, self-organization, reasoning, and learning. In our opinion, if the characteristics of human neural system are modelled and transferred to large-scale distributed systems, this will make them not only more efficient but also easier to design, deploy, and operate.

For example, in factory automation, the distributed systems used for production control are deployed in isolation without any connection with the factories business environment thus being unable to automatically adapt the production parameters in a near-real-time fashion considering factors such as product properties, costs, logistics, time, and sustainability. Neurocomputing-based methods may provide horizontal integration and optimization across the entire socioeconomic ecosystem of the factory, in which all the interested (possibly competing) stakeholders along the production chain will cooperate leveraging on the digital integration of their already deployed and vertically integrated CPS to take advantage of factories’ flexible production systems. Another example is the energy management sector in which the smart grid is usually composed of large heterogeneous and independent energy systems that interact to compete or cooperate. The increasing need for integrating larger shares of renewable energy poses severe problems because of their intermittent energy production and regionally distributed disposition requiring new system integration abstractions and advanced neurocomputing-based methods for optimal and cost-effective management.

The purpose of this special issue is to present the recent advancements in designing, development, and applications of neurocomputing-based methods (e.g., neural/bayesian networks, learning theory-based methods, and learning classifiers) for managing large-scale distributed systems.

Potential topics include but are not limited to the following:

  • Applications of neurocomputing-based methods for energy efficiency in data centers and cloud computing
  • Applications of neurocomputing-based methods for efficient management of demand response programs management in smart grids
  • Applications of neural networks and learning theory for efficient and distributed management of supply chains in circular economy
  • Applications of neural networks, learning theory, and learning classifiers in Industry 4.0
  • Applications of learning theory and neural networks in Ambient Assisted Living
  • Applications of neurocomputing-based methods for optimal coordination in disaster management
  • Applications of neurocomputing-based methods for solving optimization problems in smart cities

Authors can submit their manuscripts through the Manuscript Tracking System at

Submission DeadlineFriday, 2 March 2018
Publication DateJuly 2018

Papers are published upon acceptance, regardless of the Special Issue publication date.

Lead Guest Editor

Guest Editors