Distributed intelligent systems: A paradigm shift

Centralized systems are unsuitable for large-scale systems integration because of their high reliance on centralized communication, high complexity, lack of scalability, and high cost of integration. The use of distributed intelligence technologies avoids these weaknesses by building intelligent systems with physical and software agents that operate autonomously to independently handle specialized tasks and cooperate to satisfy system-level goals and achieve a high degree of flexibility. This lecture provides an introduction to the technologies and applications of distributed intelligent systems for dynamically changing, networked environments. It describes how a peer-to- peer environment can be built to distribute the logistical and strategic requirements of a system, while improving robustness and scalability. The presenter will contrast centralized and distributed systems, survey development frameworks for distributed systems, describe recent advances based on multi-agent and holonic systems, and survey applications involving manufacturing automation, distributed scheduling, automated decision support, RFID tracking, and distributed energy systems. Specific examples of implementations in industrial environments will be provided.