I am reading Jayant Deshpande's "Life Time Data: Statistical Models and Methods" book and while reading about Right Random Censoring, I read about this example.

Example of Random (right) Censoring

A mining company owns a 1,400 car fleet of 80 - ton high-side, rotary dump gondolas. A car will accumulate about 100,000 miles per year. In their travels from mines to a power plant, the cars are subjected to vibrations due to track input in addition to the dynamic effects of the longitudinal shocks coming through the couplers. As a consequence the couplers encounter high dynamic impacts and experience fatigue failure and wear. Twenty-eight cars are observed, and the miles driven until the coupler is broken are recorded. The remaining six cars left service after 151000, 155000, 160000, 168000, 175000 and 178000 miles.

"None of them experienced a broken coupler."

Thus giving randomly right censored data. (Example ends)

Can you explain if the quoted text meant that the 6 cars could have traveled more (since the coupler didn't break)?

  • $\begingroup$ One wonders why the remaining six cars "left service:" perhaps some other part wore out in some cases, for instance, or maybe management decided to sell them for scrap because recycled steel prices went up (this sort of thing happened a lot around 15 years ago). In light of these possibilities, in what sense is the property "could have traveled more" relevant to analyzing coupler lifetimes? $\endgroup$ – whuber Apr 18 at 13:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.