Mihai PARLEA

Title: Contributions to the Design of Semi-heterarhical Fabrication Control Systems Using the Intelligent Product Concept

Supervisor: Professor Theodor BORANGIU

Final presentation in 2011

The present economy trends demand that production cells equipped with robotized workstations offer the best possible performance and behave predictable over time. The only way that can guarantee the best possible performance is a hierarchical (or centralized) driving strategy, which executes an off-line generated schedule controlled by a central entity. Unfortunately, perturbations occur when driving such cells, such as failure or stock depletion of a robot workstation. This means that the entire cell needs to stop, at least until the planning is recalculated, with bad consequences on the production schedule. As reality is seldom as deterministic as this, the hierarchical strategies quickly become inefficient when the controlled cell has to align to a stochastic behavior that can change the objective from batch level optimization to adaptability in the face of perturbation end to real-time balancing of the resources load (Sauer, 2008).

This situation led researchers to design new shop-floor control architectures that self-organize the access to resources, featuring agility and adaptability to frequent production changes and efficiency in resource utilization. New research directions in manufacturing control have been proposed, which are centered on product-driven automation. This control paradigm is based on the "intelligent product" concept (McFarlane et al., 2002; Meyer et al., 2008).

So emerges the justification for the research focus of this PhD thesis, namely to implement multiple production control strategies for an intelligent cell, each one defined by a certain centralization degree, a certain production planning horizon and a certain perturbation rejection capacity.