Boundary Value Analysis

We already discussed that to test any application we need the definite set of input data to use as a input in out application. By using that input, system generate some output and we match that actual output with our expected output ( as per the defined Requirement).
Now we know that it is not possible to test the application with each and every set of input data because Exhaustive Testing is not possible so to avoid this Exhaustive Testing we set the boundary on input data for our application and check the application in that boundary. By using this boundary values, we can test our application with each and every set of input data and find bugs in that boundaries.

We know that maximum errors found at the boundaries of the input set of data so Boundary value analysis technique is defined to finds the errors on boundary rather then finding those exists in the center of input domain.  Normally Boundary Value Analysis is the part of Stress Testing and Negative Testing.

Now let me provide you an example to understand the Boundary Value Analysis. Suppose there is a text box who accept the value between 1 to 100 now as per the Boundary value analysis following will be the scenario.

1. First Test case we will use the input data which is on the boundary means 1 and 100.
2. Second Test case we will use the test data just below the boundary on both side means 0 and 99.
3. Third Test case we will use the test data just above the boundary on both side means 2 and 101.

So the test cases that you need to test the input is L, L-1, L+1 for lower value and U, U-1, U+1 for upper value so as per above example the test cases will be 0,1,2,99,100,101 but we need only 4 test cases which is 0,1,100,101 to test the input value. You will learn this in Equivalent Partitioning. I will explain Equivalent Partitioning in my next post.