The Development of Random Generators of Weather and Industrial Pipelines Data using Parametric and Non-Parametric ApproachesReport as inadecuate




The Development of Random Generators of Weather and Industrial Pipelines Data using Parametric and Non-Parametric Approaches - Download this document for free, or read online. Document in PDF available to download.

Simulation, Industrial pipelines, Random, Construction operation, Data generator, Weather

AL-Alawi, Mubarak K

Supervisor and department: Mohamed, Yasser Civil and Environmental Engineering Bouferguene, Ahmed Civil and Environmental Engineering

Examining committee member and department: Robinson, Aminah Civil and Environmental Engineering Moselhi, Osama Civil and Environmental Engineering Bindiganavile, Vivek Civil and Environmental Engineering

Department: Department of Civil and Environmental Engineering

Specialization: Construction Engineering and Management

Date accepted: 2017-01-20T14:52:26Z

Graduation date: 2017-06:Spring 2017

Degree: Doctor of Philosophy

Degree level: Doctoral

Abstract: Construction projects are unique and complex in nature. They are associated with many challenges regarding the randomness, complexity, and interdependency related to the operation-process, the environment hosting the operation, and the product being constructed. These challenges are also common in the area of simulation and modeling of a construction operation. Research in this field demands real life data but unfortunately the availability of such data is one of the major challenges. Also, the random generation of complex construction data structures that contain correlated attributes make it difficult to replicate real systems behaviors. The objective of this research is to investigate alternative techniques that can be used to randomly generate complex construction data structures while preserving the correlation between their formations’ attributes. This research focuses on two different types of construction-related data: weather data, and industrial pipelines data. A non-parametric approach in the form of bootstrapping technique was applied in the generation of weather data, and its performance was measured against a parametric weather generator constructed in the field of modelling construction operations. The validation results showed that the proposed technique performed in a manner similar to that of the parametric weather generator and outperformed it in some cases. A parametric approach in the form of Markov chain technique was applied to randomly generate industrial pipeline data structures, and its performance was tested against real pipeline data. The validation results showed that the proposed Markov chain model was able to generate an industrial pipeline data structure similar to those in reality. The majority 89% of generated pipelines shared characteristics with 85.5 % of the original pipelines. This research demonstrates the application of the developed generators in two areas. The first application modelled an earthmoving operation in oil sand mining and used the weather generator to analyse the effect of temperature on breakdown and repair durations. The second application involved building a pipe-spooling optimization model and used the industrial pipelines data generator to randomly generate instance problems to test the computational efficiency of the optimization’s solution algorithm.

Language: English

DOI: doi:10.7939-R3QJ78959

Rights: This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for the purpose of private, scholarly or scientific research. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.





Author: AL-Alawi, Mubarak K

Source: https://era.library.ualberta.ca/


Teaser



The Development of Random Generators of Weather and Industrial Pipelines Data using Parametric and Non-Parametric Approaches by Mubarak Khamis AL-Alawi A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Construction Engineering and Management Department of Civil and Environmental Engineering University of Alberta © Mubarak Khamis AL-Alawi, 2017 ABSTRACT Construction projects are unique and complex in nature.
They are associated with many challenges regarding the randomness, complexity, and interdependency related to the operation-process, the environment hosting the operation, and the product being constructed.
These challenges are also common in the area of simulation and modeling of a construction operation.
Research in this field demands real life data but unfortunately the availability of such data is one of the major challenges.
Also, the random generation of complex construction data structures that contain correlated attributes make it difficult to replicate real systems behaviors.
The objective of this research is to investigate alternative techniques that can be used to randomly generate complex construction data structures while preserving the correlation between their formations’ attributes. This research focuses on two different types of construction-related data: weather data, and industrial pipelines data.
A non-parametric approach in the form of bootstrapping technique was applied in the generation of weather data, and its performance was measured against a parametric weather generator constructed in the field of modelling construction operations.
The validation results showed that the proposed technique performed in a manner similar to that of the parametric weather generator and outperformed it in some cases.
A parametric approach in the form of Markov chain technique was applied to randomly generate industrial pipeline data structures, and its performance was tested against real pipeline ...





Related documents