Abstract

In industry product testing can be an expensive and time-consuming process. Testing design changes in long-lived products could cause lengthy delays in product introduction or improvement. As an alternative, accelerated life testing can quickly yield information on product life by exposing the product to conditions beyond those of normal design stress. To further streamline this process a two step-stress test will take all elements to failure in a relatively short time. Variables within the sample other than the one that we are controlling in the step-stress testing are uncontrolled but observed and are called covariates. A statistical relationship between the mean lifetime of the test unit and the covariate will allow a prediction of mean lifetime based on the covariate.

In agriculture, animals, or plants may be the test items and dosage of a chemical, amount of fertilizer, temperature, etc may be the stress variable. The breed of the animal or the variety of the plant may be the covariate. In this paper we suggest several potential applications of stepstress testing in agriculture and present inferential procedures for observations that are distributed exponentially.

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 27th, 2:00 PM

SIMPLE STEP-STRESS TESTING WITH COVARIATE IN AGRICULTURE

In industry product testing can be an expensive and time-consuming process. Testing design changes in long-lived products could cause lengthy delays in product introduction or improvement. As an alternative, accelerated life testing can quickly yield information on product life by exposing the product to conditions beyond those of normal design stress. To further streamline this process a two step-stress test will take all elements to failure in a relatively short time. Variables within the sample other than the one that we are controlling in the step-stress testing are uncontrolled but observed and are called covariates. A statistical relationship between the mean lifetime of the test unit and the covariate will allow a prediction of mean lifetime based on the covariate.

In agriculture, animals, or plants may be the test items and dosage of a chemical, amount of fertilizer, temperature, etc may be the stress variable. The breed of the animal or the variety of the plant may be the covariate. In this paper we suggest several potential applications of stepstress testing in agriculture and present inferential procedures for observations that are distributed exponentially.