Authors: Dr. Ryan Dougherty, CDT Dylan Green, CDT Hyunmook Kang, CDT Grace Kim, Dr. Stephanie Forrest
The purpose of this program is to generate a locating array. We use a two-stage approach, where the second uses a genetic algorithm.
The program will append rows for non locating pairs from a given covering array using a two-stage genetic algorithm until a complete Locating Array is created. The total number of rows needed to make it locating is returned at the end of the program.
To compile the program use the command navigate to the location of :
g++ -std=c++20 -O3 LocAG.cpp -o ./LocAG
Run the ./LoCAG program after compilation by running the command: ./LocAG <name of config>. The names of real-world systems we have modeled are in the LocAG.cpp file.
Combinatorial Interaction Testing (CIT) refers to the creation of test suites that detect or locate desired interactions. Past research has shown success in faster location of specific interactions using a two-stage creation of uniform Locating Arrays (LAs). Given that LAs guarantee unique location for every such set of interactions, we contribute to past research by implementing the ability to generate non-uniform arrays using a multistage genetic algorithm that splits up the second stage of generating locating pairs.
Covering Arrays (CAs) A formal model that guarantees all interactions of size at most a specified number,
Locating Arrays (LAs) - Test suites that can detect all covered interactions between test variables of size at most