Full text
| |
| Author(s): |
Tony Minoru Tamura Lopes
Total Authors: 1
|
| Document type: | Master's Dissertation |
| Press: | Campinas, SP. |
| Institution: | Universidade Estadual de Campinas (UNICAMP). Instituto de Computação |
| Defense date: | 2010-07-29 |
| Examining board members: |
Arnaldo Vieira Moura;
Guilherme Pimentel Telles;
Maria Teresa Moreira Rodrigues
|
| Advisor: | Cid Carvalho de Souza; Arnaldo Vieira Moura |
| Abstract | |
A set of oil derivative distribution depots, including refineries and terminals, have local demands for and productions of different products in a given time horizon. However, there may be not enough local stock of some product to satisfy the corresponding demand, or there may not be enough tank capacity to stock the local production. This brings the need for transportation of oil derivatives between the depots. Among many transportation modes, the network of pipelines is one of the best options when considerying cost and environment risks. In order to adequately operate the pipeline network, a two phase planning strategy is developed. First, a tactical pumping plan is composed monthly and, secondly, a more detailed operational schedule, spanning a few days, is updated daily. Both the tactical and tghe operational plannings must satisfy a large set of operation constraints, involving many restrictions, such as tanks capacities, pipeline flow rates, and stock levels. This dissertation provides a formalization for the problem along with a decomposition of it in two stages, representing the monthly planning and operational schedule. The tactical stage is solved by applying a heuristic and then with a network flow model, while the operational schedule uses constraing programming. Our model treats the oil pipeline network that is operated by the Brazilian oil company Petrobras. This is one of the most complex and large topologies when compared to other networks treated in the open literature. The model was tested with real-world instances and showed significant improvements over human planning (AU) | |