Self-Regulation in Self-Organising Multi-agent Systems for Adaptive and Intelligent Manufacturing Control

In this paper, we explore the potential of distributed satisfaction techniques as to provide self-regulated manufacturing control. This work relies on a DisCSP-based modeling distributed among agents (e.g. machines) having enough and reasoning capabilities to cooperate and negotiate for a committed schedule. This approach is used to dynamically regulate the system (the network of machines) when perturbations occur (machine break-out, operator or container unavailability, or even priority command). Thus, for these machines, embodied intelligence and autonomy are a mean to provide a more flexible and adaptive manufacturing network. In this paper, we present two different multi-agent models and two extensions of well-known DisCSP solvers. Experiments using a dedicated simulation platform, MASC, are presented and discussed.