PATTHI, HAMDAN (2013) DEVELOPMENT OF EXCEL-BASED RELIABILITY MODEL FOR SYSTEMS WITH INCREASING/DECREASING FAILURE RATES. [Final Year Project] (Unpublished)
HAMDAN BIN PATTHI 14007.pdf
Download (1MB)
Abstract
Reliability is the probability that a system, component, or device will perform without failure for a specified period of time under specified operating conditions. The objective of this project is to develop excel-based reliability model for systems with increasing and decreasing failure rates. There are several method used to perform the excel-based reliability model for systems such as the analytical method,. Crow-AMSAA method, Maximum Likelihood method and etc. Microsoft Excel is becoming more applied in the industry sector as it is the most familiar. After all the problems have been analyzed and defined, the available data gathered and model development must be conducted for further research. some more specific models was proposed which assume specific forms for the shape of the curve and the simpler models in most widespread use are based on an exponential „learning curve‟. The first, and well known of these is due to Duane. Plots or Crow/AMSAA Plots. The technique involves plotting most commonly cumulative failures Vs cumulative time on log-log scale with the resulting straight lines‟ slope indicating improving. Ways to solve reliability model for systems with increasing or decreasing failure rates is by focusing on the simulation, graphical and analytical method. Crow AMSAA is often used on systems when the usage is measured on a continuous scale. The main application is also for high reliability, huge number of trials and one-shot items. The Maximum likelihood estimation is a totally an analytic maximization procedure approach. It applies within every form of censored or multicensored data, and it is even possible to be used the technique across several stress cells and estimate acceleration model parameters at the same time as life distribution parameters. Lastly, Weibull distribution is the most popular statistical distribution among reliability engineers it can fit all other distribution by changing the parameters. Both Weibull++ Software and Excel Microsoft Software is used to compare the results obtained ( in Fig 4 and Fig 5). The results include the Plot Failure Vs Cumulative on log-log. The toolkit developed in Excel Microsoft Software can finally be used to predict future failures by estimating the parameters needed that can also be found the same parameters in the Weibull++ that can be a great achievement in software history.
Item Type: | Final Year Project |
---|---|
Subjects: | T Technology > TJ Mechanical engineering and machinery |
Departments / MOR / COE: | Engineering > Mechanical |
Depositing User: | Users 2053 not found. |
Date Deposited: | 13 Nov 2013 17:20 |
Last Modified: | 25 Jan 2017 09:39 |
URI: | http://utpedia.utp.edu.my/id/eprint/10615 |