A process-based control for evolvable production systems

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia Electrotécnica e de Computadores

Nowadays, companies in a challenging environment are compelled to adapt to the rapid changes in the manufacturing business. The search for new processes to create products with short life-cycles at low cost, while keeping the same levels of productivity and quality is greater than ever. This has generated the need to create even more agile manufacturing systems, which could easily adapt to the market changes at a low cost. Advances in information technologies have allowed manufacturing systems to achieve new levels of agility, opening the doors to new approaches. These same advances helped companies in several sectors other than manufacturing to gain e ectiveness through the synchronization of the processes of their several departments by using Business Process Management tools. This thesis proposes a system that reacts and adapts itself to di erent production orders by means of recon guration. To reach this goal, the concept of Business Process Management was used. This concept, already used in many companies, allows them to model their inner behaviours with processes that can be changed according to their needs. A manufacturing system using this may become equally agile and alter its functioning in accordance with the needs of other departments of the same company. To create the system presented in this thesis it was used a multi-agent architecture based on process execution. Each agent contains a knowledge base, used by its processes,that stores internal or external information. This system may be used not only in the manufacturing shop oor, but also in any other areas within a company. This thesis also presents an application of the system to the shop oor, based on the Evolvable Production Systems concept, in which each agent represents a manufacturing resource that o ers a given set of services useful to the production process. The resources,by means of the agents, may aggregate among themselves to execute services together. Keywords: Manufacturing system, multi-agent system, ontology, process, BPM, EPS.