Orthogonal Array Test Strategy is a systematic and statistical way of testing pair-wise interactions. In simple terms, Orthogonal Arrays are special set of Latin squares, constructed by Taguchi to design the test sets. It provides representative (uniformly distributed) coverage of all variable pair combinations. OATS provides a test set that guarantees, testing the pair-wise combinations of all the variables. It creates an efficient and concise test set with very few test cases rather testing all the combinations of all the variables. It also exercises some of the complex combinations of all the variables. Following are some of the key concepts used in Orthogonal Analysis.
Factors: Number of variables whose pair-wise combinations to be arrived at.
Levels: Maximum of the number of values, each of the factors can take.
Strength: Number of columns it requires to visit each of the levels possibilities equally often.
Runs: Number of rows in the Orthogonal Array.
Usually an Orthogonal Array is represented as Lr (f ^ l) where, r is the number of runs, f is the number of factors and l is the number of levels. So L9 (4^3) represents an Orthogonal Array with 4 factors, 3 levels and 9 runs.