Otimização das operações de terraplenagem e pavimentação: uma nova abordagem com a consideração de interferências

• Ano de defesa: 2021
• Autor(a) principal: Lima, Roberto Xavier de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Engenharia rodoviária; Terraplanagem; Pavimentação; Otimização; Programação matemática; Programação Linear
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/61820

An optimization model for planning the distribution of materials in earthmoving operations is presented in this thesis, considering possible interferences between cut sections, fill sections and paving strata such as rivers, vegetation, drainage works or expropriations. The earth allocation problem incorporating blocks was modeled as a linear programming problem, aiming to minimize the total earthmoving cost while considering the constraints related to volume balance, construction completion time and time for release equipment traffic. The proposed model was run by an integrated system using Excel for data entering and IBM CPLEX as the optimizer. The mathematical model was validated by one numerical example and a real-world project of a dam maintenance road in the city of Maranguape, State of Ceará, Brazil. The unit costs and productivity rates used in the fictional examples and real-world application followed the referential cost system created by Ceará’s Secretariat of Infrastructure. The proposed optimization model achieved reasonable processing times for all tested applications, presenting itself as a viable and efficient option for planning earthmoving operations. The linear programming approach provided a 2.12% cost reduction for the real-world case study, when comparing the optimized solution and original budget. The study explored the problem of earth allocation with interferences avoiding complex algorithms found in recent literature. As a result it was proposed a user-friendly optimization system that can be easily utilized by construction companies and public agencies.