Comparison between maintenance policies based on q-Weibull and Weibull models

Abstract

PurposeThe purpose of this article is to compare maintenance policies based on Weibull and q-Weibull models.Design/methodology/approachThis paper uses analytical developments, several figures and tables for graphical and numerical comparison. Previously published hydropower equipment data are used as examples.FindingsModels for optimal maintenance interval determination based on q-Weibull distribution were defined. Closed-form expressions were found, and this allows the application of the method with small computational effort.Practical implicationsThe use of the q-Weibull model to guide the definition of maintenance strategy allows decision-making to be more consistent with sample data. The flexibility of the q-Weibull model is able to produce failure rate modeling with five different formats: decreasing, constant, increasing, unimodal and U-shaped. In this way, the maintenance strategies resulting from this model should be more assertive.Originality/valueExpressions for determining the optimal interval of preventive maintenance were deduced from q-Weibull distribution. Expected costs per maintenance cycle of Brazilian hydropower equipment were calculated with q-Weibull and Weibull distributions. These results were compared in terms of absolute values and trends. Although a large number of works on corrective and preventive maintenance have been proposed, no applications of the q-Weibull distribution were found in literature.

Study on the Generation of Carbon Particles in Oil and its Effect on the Breakdown Characteristics of Oil‐Paper Insulation

“…where y is the breakdown voltage obtained from the test, α is the scale parameter, β is the shape parameter, and F (y) is the probability of failure corresponding to parameter y. The value of the empirical distribution function F (i , m) [35] is obtained from the calculation of (7), where i denotes the order of the test samples and m denotes the number of samples. The two characteristic parameters in the two-parameter Weibull distribution were estimated by the least squares method, as shown in Table III.…”

“… $$$$ F(y)=1-\exp \left({\left(-\frac{y}{\alpha}\right)}^{\beta}\right) $$$$ where y is the breakdown voltage obtained from the test, $$$ \alpha $$$ is the scale parameter, $$$ \beta $$$ is the shape parameter, and F ( y ) is the probability of failure corresponding to parameter y . The value of the empirical distribution function F ( i , m ) [35] is obtained from the calculation of (7), where i denotes the order of the test samples and m denotes the number of samples. $$$$ F\left(i,m\right)=\frac{i-0.44}{m+0.25}\times 100\% $$$$ …”

Maintenance Strategy Choice Supported by the Failure Rate Function: Application in a Serial Manufacturing Line

“…This is not the case of our study, as the involved equipment is composed of independent, autonomous machines. Assis et al (2021) compared results obtained with the Weibull and q-Weibull models for the same field data. The q-Weibull model allows new formats for the bath-tube curve, such as the unimodal and U-shaped formats.…”