Modelos de atração de entregas e cargas para a indústria de máquinas agrícolas e rodoviárias

• Ano de defesa: 2019
• Autor(a) principal: José Moreira Gonçalves
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA TRANSPORTES E GEOTECNIA; Programa de Pós-Graduação em Geotecnia e Transportes; UFMG
• Palavras-chave em Português: Transporte de carga; Modelos de geração de viagens de carga; Indústria de máquinas agrícolas e rodoviárias; Transportes; Transporte de cargas; Maquinas agrícolas - Indústria; Máquinas rodoviárias
• Link de acesso: http://hdl.handle.net/1843/30058

The economic development of cities is closely related to the distribution of goods concerning various industrial sectors. However, these activities also generate negative impacts that can be mitigated when the freight flows are acknowledged. Thus, it is important to develop models for freight trips to subsidize city planning. So, in this work, freight generation models for agricultural and road machinery industries were determined and compared conceptually. These models are important for estimating supply flows in this sector and were developed using data from four industries located in Contagem (Minas Gerais), Curitiba (Paraná), Piracicaba and Sorocaba (São Paulo). A systematic review of the literature regarding freight generation models was conducted to better understand the concept and methods used. Information was collected concerning the supply flows in the industrial facilities considered in this work. The data comprised trips performed from January until December 2017, contemplating 24 hours of operation for each observed industry. Statistical analyses of the collected data were then carried out and generalized linear regression models were developed. The models were evaluated regarding their predictive capacity using the leave-one-out cross-validation method and compared qualitatively. Models with better predictive accuracy and balance between simplicity and adjustment were those developed considering the frequency of deliveries as the dependent variable and production, the number of employees allocated to logistics activities and the total built area in each industrial unit as explanatory variables. None of the models that considered the amount of cargo received as a dependent variable presented satisfactory fit.