Combinatorial Test Design Methods
One important part of Agile Software Engineering (and the ASE course at SAP) is 'test design methods'. In testing, a usual complaint is that "there are so many combinations - we cannot possibly test them!". And then comes the tedious part of deciding what cases to test and which ones to ignore - hoping for the best. In the ASE course we do not cover the test design part in sufficient depth to be able to actually use the methods practically, therefore we now publish these formerly internal training videos. We hope they are useful to you as well.
The videos below explain an test design method that integrates some well-known and some lesser-known test design techniques into one integrated approach. The mehod combines equivalence partitioning, boundary values, decision tables and All-Pairs into one comprehensive approach to cover virtually any number of parameters / cases with very reasonable effort. The videos listed below describe the approach step by step. The All-Pairs method uses the PICT engine from Microsoft (download link below).
To emphasize that we are dealing with the problem of combinatorial explosion of possible cases (as opposed to many other aspects of test design), we call this approach 'combinatorial test design'.
The videos are (in order):
- Introduction to Combinatorial Test Design
- Equivalence Partitioning
- Boundary Value Analysis
- Decision Tables
- All-Pairs Test Design
Tools: All-Pairs is supported by the PICT tool from Microsoft and a 'covenience wrapper' XLS written by us.
- You can download the All-Pairs tool 'PICT' from Microsoft.
- You can also download the All-Pairs Wrapper XLS that makes working with PICT more convenient and provides some additional analysis data (it needs PICT installed). The tool is provided as-is without any warranty!