Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/499396Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Ahmad Mahir Razali, Associate Professor Dr. | - |
| dc.contributor.author | Ali Abdul Hussain Salih Al-Wakeel (P38442) | - |
| dc.date.accessioned | 2023-10-13T09:31:46Z | - |
| dc.date.available | 2023-10-13T09:31:46Z | - |
| dc.date.issued | 2013-08-07 | - |
| dc.identifier.other | ukmvital:71566 | - |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/499396 | - |
| dc.description | Mixture Weibull distribution is used to represent times to failure for a number of units when there are more than one cause for failure. It is also used to model data made of a number of sub-populations, each sub-population has its own unique failure mode and distribution. Mixture Weibull distribution is produced from combining k Weibull distributions by including a mixing parameter wi where i=1,2,...,k is the number of component Weibull distributions. The mixing parameter w represents the proportion of contribution of each of the component distributions in the mixture Weibull distribution. The number of parameters in the mixture Weibull distribution is the total number of parameters (shape and scale) in each of the component distributions in addition to the mixing parameters. This study introduced different cases of twocomponent mixture Weibull distribution, with different values of parameters using different sample sizes and different values of the mixing parameters. A number of approaches for parameter estimation were discussed including graphical and analytical methods. A new approach for determining the mixing parameter was introduced in addition a technique was proposed for setting initial values for the iterative procedure. A simulation study was included for different sets of parameters representing uni-modal and bi-modal mixture Weibull distribution. The study was based on 5000 replicates of generated data and the parameters were estimated using maximum likelihood estimation. The effects of sample size, mixing parameter and outliers on parameter estimates and bias were investigated. It was found that increasing the value of the mixing parameter caused a decrease in the values of mean, mode and median. Also, increasing the sample size caused decrease in the bias, mean square error and standard deviation. On the other hand, it was found that by removing the outliers caused a good improvement in the accuracy of the parameter estimates and decreased the values of the bias and mean square error. Real data of wind speed in eleven stations in Peninsular Malaysia were modeled using single two-parameter, single three-parameter and mixture Weibull distributions. The frequency histograms of the data showed uni-modal for some stations and bi-modal for other stations. It was found that mixture Weibull distribution provided higher degree of fit for the data compared to single two- and three-parameter Weibull distributions for both uni- and bi-modal represented by higher coefficient of determination, and lower mean square error and relative error for most stations under study.,PhD | - |
| dc.language.iso | eng | - |
| dc.publisher | UKM, Bangi | - |
| dc.relation | Faculty of Science and Technology / Fakulti Sains dan Teknologi | - |
| dc.rights | UKM | - |
| dc.subject | Mixture Weibull Distribution | - |
| dc.subject | Bi-Modal Mixture Weibull Distribution | - |
| dc.subject | Uni-Modal Mixture Weibull Distribution | - |
| dc.subject | Weibull distribution | - |
| dc.title | Mixture Weibull Distribution; Its Parameters Estimation And Application On Wind Speed | - |
| dc.type | Theses | - |
| dc.format.pages | 202 | - |
| dc.identifier.callno | QA273.6 .W335 2013 | - |
| dc.identifier.barcode | 000612 | - |
| Appears in Collections: | Faculty of Science and Technology / Fakulti Sains dan Teknologi | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_71566+Source01+Source010.PDF Restricted Access | 8.13 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.