Statistical distribution fitting for real test data
Tan, Germaine Qiu Wen.
Date of Issue2010
School of Electrical and Electronic Engineering
In this report, the various statistical distributions and types of Goodness-of-Fit tests are presented. Reliability engineers today are required to analyze large amounts of data and it is unwise to assume a specific distribution as it would be time consuming. The main goal of this project is to help analysts reduce the time taken for an analysis, by giving them the specific statistical distribution their data follows. Microsoft Excel workbook is created to help specifically state if the data follows either an Exponential distribution or a Two-Parameter Weibull distribution. If the data does not follow either distribution, it is tested with external add-in software known as EasyFitXL. The Microsoft Excel workbook created is aimed to be user-friendly and comprehendible for analysts, reducing the time spent initially as they assume a distribution of their data. Microsoft Excel is also chosen for this project as it is a widely used application, which is readily available for most users. This project illustrates the simplicity of having the specific form of statistical distribution for analyzing of data.
Final Year Project (FYP)
Nanyang Technological University