Loading...

Multi-Objective Optimization in Cost, Time and Quality Trade-off in Projects with Quality Obtained by Locally Linear Neuro Fuzzy Networks Case Study: Well Drilling Projects

Mohammad Alipour Ahari, Roya | 2014

860 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 46782 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Akhavan Niaki, Taghi
  7. Abstract:
  8. A common decision in project management is the selection of contractors to simultaneously optimize three objectives of the project’s triangle. This issue becomes more important if there is more monetary value involved or there is limited number of capable contractors. In project planning, there is numerous choices for tasks; each with specific time, cost and quality. There is no trade-off problem, if one of the choices has the best time, cost, and quality, simultaneously. However, as these criteria do not generally work in a unique direction, it makes the selection decision difficult. A contractor who performs the projects on time may have higher cost and lower quality. Given that there is a certain number of contractors for each task group and having three measures to evaluate them makes the contractor selection problem equivalent to a discrete multi-objective optimization problem. The time and cost of each task group is known for managers. However, there is ambiguity involved in definition of the quality parameter, where there is a point of discussion about effective parameters and quantification approaches. Considering the importance of this issue and the quality parameter features, it is suitable to use fuzzy arguments and learnable networks to evaluate the quality parameter. In this dissertation, a neuro-fuzzy network named Locally Linear Neuro Fuzzy (LLNF) was employed in some well-drilling projects to evaluate and quantify the quality of the projects. With the quality parameter quantified along with the time and cost for all execution modes of a project’s tasks, a multi-objective problem has to be solved. A multi-objective genetic algorithm is utilized in this research to find the Pareto answers of a well-drilling project as a case study
  9. Keywords:
  10. Locally Linear Neuro-Fuzzy Model ; Multiobjective Optimization ; Multiobjective Genetic Algorithm (MOGA) ; Non-Dominate Sorting Genetic Algorithm (NSGAII) Method ; Quality Evaluation ; Well Driling Projects

 Digital Object List

 Bookmark

...see more