Modelagem baseada em agentes como ferramenta de análise da operação florestal de Baldeio

• Ano de defesa: 2017
• Autor(a) principal: Kelen Karla Reis Oliveira Moreira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Baldeio florestal; Simulações; Modelagem baseada em agentes (mba); Tratores; Madeira Exploração; Modelagem de dados Aspectos ambientais
• Link de acesso: http://hdl.handle.net/1843/IGCM-AV4NVG

Over the last few years, efficient planning of the forest operations involved in the timber production process has enabled the sector to achieve important technical, operational and, above all, productivity gains. However, the increasing costs have awakened the sector to the need to seek new tools, which aim at rationalizing resources and minimizingenvironmental impacts. In this context, this research aims to develop, through spatial modeling tools, an agent - based simulation model (MBA) capable of analyzing the wood recycling activity, present in eucalypt forest harvesting operations, based on Preestablished criteria aiming to optimize the performance of three important components of the extraction activity, linked to the main agent of the process: forest tractors forwarders.These components are: the total distance traveled; The time spent to execute the activity; And fuel consumption. The established criteria gave rise to two scenarios of execution of the activity. The data obtained through the simulations showed that in the first scenario the total distance traveled by the tractors was, on average, 7% lower than in the second one. The mean values of distance traveled by each of the eight fleet configurations tested were applied to a statistical test comparing means, which showed statistical difference between the means of the two scenarios. The time spent to execute the activity was lower in the first scenario for the fleet configurations 1, 2 and 3; For the others the time was lower in the second scenario. It was also observed that the average time of execution of the activity with the largest number of tractors, in relation to the activity of smaller numbers, is on average 62% lower for both scenarios. The fuel consumption data indicated a lower consumption of the tractors of the first scenario. It was also verified thatwith each addition of tractors in the fleet, there was an average increase of 24L in the fuel consumption by the tractors in the first scenario and 25L in the second. The Gini coefficient indicated that the inequality in work performed by each tractor is very small and increases slightly as more tractors are added to the model. The results demonstrate the potential of the tool for the compression of the patterns resulting from the interactionsbetween the agents. It is possible to analyze and understand the behavior of the variables of interest within the model