Monitoring the Weibull Scale Parameter Based on Type I Censored Data Using a Modified EWMA Control Chart

Round 1

Reviewer 1 Report

A special procedure is proposed for detecting shifts of the scale parameter in the case of censored Weibull lifetime data and the shape parameter fixed. The submitted manuscript is a typical boring statistical paper. No theorems, no statements. But there are formulas. I checked. In my opinion, the given formulas are correct. However, I cannot say whether the entries in the tables are correct. You need to either reprogram or trust the authors. A few notes on the text.

(1) The title of the article must be shortened. Axioms is not a journal of experimental physics.

(2) Please check lines 106 and 107.Need to clarify.

(3) There are too many tables in the work. If you need so many tables to justify your procedure, then your submitted procedure is useless. It would be best if you demonstrated the advantages of your procedure in theory rather than in tables.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The study is a well presented big project (that is, necessarily, advancing previous work) that is fully explained in a few pages. 

Author Response

Thank you for your highly favorable recommendation. The authors wish to convey their heartfelt thanks to the reviewers for careful reading and providing an excellent summary.

Reviewer 3 Report

Reviewer’s Comments on axioms- 2343455

 

Monitoring the Weibull scale parameter based on Type I censored data using a modified exponentially weighted moving average control chart

 

This paper provides a modified EWMA control chart for type I right-censored lifetimes that are assumed to follow a Weibull distribution. The proposed chart aims to detect shifts in the scale parameter by considering that the shape parameter is fixed. Simulation studies are provided where the proposed approach is compared with other charts and an application in a case study is presented and discussed. I believe that this is an interesting manuscript with interesting contributions. I have the following comments:

 

1.     The authors are encouraged to clearly state and discuss the main contributions of the manuscript in the introduction section, this by denoting the advantages of the proposed approach in comparison with the discussion of other works. Although the authors provide discussion of relevant works the contributions are not emphasized.

 

2.     In line 100, the authors mention that if beta is fixed, then monitoring the decreasing mean is equivalent to detecting a shift in eta. Of course, the mean depends of both parameters, if beta is not fixed then the mean is affected. In this sense, the authors are encouraged to discuss why it is assumed that the shape parameter is fixed and the scale parameter is assumed to vary. Is this scenario appropriate for real life case studies? Please discuss accordingly.

 

 

3.     The authors are encouraged to present a flow chart that summarizes the steps of the proposed approach. This would help to give a clearer vision of the method.

 

4.     As a Weibull distribution is assumed, is it convenient to verify the goodness of fit if this distribution, in certain periods? Please discuss accordingly in the realistic example in section 6.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf