Modelo fuzzy cascata multiatributos e preditivo para despacho de AGVs em FMS

• Ano de defesa: 2016
• Autor(a) principal: Caridá, Vinicius Fernandes
• Orientador(a): Morandin Júnior, Orides
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
• Palavras-chave em Português: Despacho de AGVs; Inteligência artificial; Sistemas flexíveis de manufatura
• Palavras-chave em Inglês: Fuzzy logic; Prediction; Automated Guided Vehicletems; Automated Guided Vehicle; AGV control
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/7985

In recent years, manufacturers are increasingly applying automation techniques with the aim of increase their efficiency to remain competitive. The material handling is an essential activity in any process of production and its effectiveness has severe impacts on production costs. Systems of automated guided vehicles (AGVs) have become an important strategic tool for factories and automated warehouses. In a very competitive business scenario, they can increase productivity and reduce costs. The management of these AGVs is the key to a transport system that ensures the improvements envisioned by the industry. One of the main problems encountered in the management of AGVs is the dispatch decision. This paper proposes a vehicles dispatch model based on a fuzzy cascade system for consideration of multiple attributes of the factory and a structure based on state space that enables the extraction of information of future states of the industrial production system. The objective is to reduce makespan and tardiness values of the production system. Two factory scenarios are simulated and tests are performed of the model and five other methods of dispatch. A statistical validation is realized of the results in which corroborates with 97% confidence the hypotheses of the work.