Browse Prior Art Database

Prioritizing tests for automation

IP.com Disclosure Number: IPCOM000240596D
Publication Date: 2015-Feb-11
Document File: 2 page(s) / 22K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method to prioritize the manual work of automating software or hardware tests, taking into account the specificity of each test step, and the number of additional tests in which this step appears.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Prioritizing tests for automation

Automating manual testing is a way to reduce costs, but not all instructions for manual test execution are equal. Test that are very specific, such as "enter 15.3 to the VAT field" are relatively easy to automate, whereas "fill in the VAT field" is not. Another factor that should be taken onto account is that some test steps are common to many tests, whereas others are unique,

GBS is sometimes confronted with the task of taking responsibility for hundreds of thousands of tests, and deciding which test to automate next is a significant challenge.

There is a need to help GBS in prioritizing automation efforts.

Prior Art

Searches for "concrete values in test automation" and for "test automation order optimization" did not find any work related to optimizing the process of test automation.

A search for "optimizing the process of test automation" did not find anything related to the order in which tests are automated.

We propose to solve the automation prioritization challenge in two steps:

1.

The cost of a test step is an estimate of how much effort is needed to process that step.

The credit of a test step is an estimate of how much effort is saved in processing other tests by processing this step.

There are multiple methods for defining the credit of a test step. For example:

1.


2.


3.

Provide a tool that allows a practitioner to classify test steps as easy or hard to automate (or, optionally, additional values in between) , taking advantage of redundancy in test steps:

Steps can repeat exactly


1.

Steps can belong to the same topic


2.

Steps can be abstract or concrete instances of the same goal (e.g., "add the number of


3.

friends" and "add 3")

Order test cases so that the automation effort is distributed smartly over all the tests.


2.

Prioritizing test cases:

For every test step we define a cost and a credit.

Count the number of occurrences of the test step in the test suite that is being processed . In this case the cost may be 1, and the credit N-1, where N is the number of occurrences of the t...