Overview
In this tutorial, we will discuss two important metrics for error removal effectiveness. Provided below are sample error distribution and sample weights of the different classifications of the errors.
- DERE
- DWERE
Let’s assume a Sample hypothetical project called TestingDocs with 325 KLOC and the development effort hours of 540 man-hours. The Error distribution is given below.
Errors detected Distribution: Table 1
Sample Error Weights: Table 2
The sample defect weights in the project are as below. There are classified into three categories called
Simple, Average, and Complex.
DERE(Development Errors Removal Effectiveness)
DERE =
NDE = Number of Design & Development errors
NYF = Number of software failures in 1 year of Maintenance
Let’s calculate the metric for the Sample project based on the given data. Lookup the above table 1 for the number of design and development errors detected during those phases.
NDE =
NYF = Number of software failures in 1 year of Maintenance =
DERE =
=
DWERE(Development Weighted Errors Removal Effectiveness)
DWERE = 
WDE=The Weighted Development&Design Errors.
WYF = The Weighted number of software failures in 1 year of Maintenance.
Lookup both the tables 1 and 2 for the errors in the classifications and multiply with the corresponding defect weights to get the weighted score.
WDE =
=
WYF =
=
DWERE =
=
=
Error Density Metrics
https://www.testingdocs.com/error-density-metrics/