Tese
Um simulador para análise de redes de produção da indústria 4.0
Fecha
2019-12-16Autor
Pereira, Adriano
Institución
Resumen
The evolution of technology and production methods has led to revolutions in industry fields.
The first three Industrial Revolution were marks of these advances, and they keep being
studied after their occurrence. In 2011, at Germany, the term Industry 4.0 was created. It
refers to the Fourth Industrial Revolution, which has been studied before it happens, or while
it is happening, differently of the previous three. Industry 4.0 uses technologies such as
Internet of Things and Cyber-Physical Systems to build smart-factories, where machines and
human interact and make decision based on real time information. The interaction of these
actors makes huge and complex production networks, which study is important to the advance
of Industry 4.0; nevertheless, this study is too expensive in the real environment. In this
context, modeling and simulation techniques can be used to allow analysis and evaluation of
different production scenarios. This work had as goal the development of a software simulator
for the analysis of Industry 4.0 production networks. Therefore, Design Science Research
method was used. This method is used when the research aims to build constructs to solve
relevant problems, based on scientific rigor and the extension of the solution to a class of
problems. The simulator development began with literature review and requirements
definition. After that, an Industry 4.0 production network model was created and, then, three
algorithms were proposed to define nodes and paths to execute production tasks in the
network. The algorithms were based on depth-first search, breadth-first search and genetic
algorithm metaheuristic. Both production model and the algorithms were validated. The
simulator was implemented as a software, and tests were performed to check different
algorithms and network structures behaviors, as well as networks behaviors with failures.
Among the three proposed algorithms, genetic algorithm obtained better results regarding to
tasks execution. Besides, it was proved that the simulator can be used to evaluate different
scenarios, base on production analysis in different networks configurations, as well as in the
presence of failures. Then, the work achieved its goal, and it can be used in different areas of
Industry 4.0 production networks study, being a theoretical framework to these networks analysis.