Modeling, simulation and performance analysis of dispatch policy for open pit mines

• Ano de defesa: 2019
• Autor(a) principal: Flávio Henrique Batista de Souza
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA ELÉTRICA; Programa de Pós-Graduação em Engenharia Elétrica; UFMG
• Palavras-chave em Português: Dispatch process; Experiments design; Open pit mine; Simulation to discrete events; Stochastic timed Petri nets; Generalized stochastic Petri nets; Engenharia elétrica; Planejamento experimental; Redes de petri
• Link de acesso: http://hdl.handle.net/1843/30756

A structured and robust methodology of modeling and simulation can be available through modular stochastic timed Petri nets, including experiments that allow designers to simulate the processes involved in ore production on well-founded estimates. The research represents a relevant industrial production process optimization through a computational simulation structure, and a modeling of complex stochastic systems, constructed from real data of the process (extraction of knowledge based on data). This prerogative guides an experimental research based on real data from an Brazilian open pit mine operation. This research presents a continuity and assertion of experiments already performed with modular Petri nets, which presented significant and better results than tools already available in the industry (SIMAN). Firstly, we present an experimental design optimization methodology, focusing on performance and response quality in the modeling and simulation of the open pit mine truck dispatch process. Then, this research introduces and compares two fast heuristics for dispatch strategies in a mine simulated using Stochastic Timed Petri nets. Issues such as productivity and equipment positioning are evaluated and discussed in order to allow synthesizing sub-optimal dispatch policies. As results of the experiment, we derived a valid simulation structure for the open-pit mining process using Petri nets. It was obtained a behavioral evaluation of the efficiency of the structure according to variations in the probability distribution function. The simulation structure demonstrated a modular, scalable, transparent modeling, besides providing mathematical support, power analysis and computational efficiency.