Verification and validation are necessary efforts to ensure the integrity of any computational model. However, verification and validation alone are not sufficient.

Verification is intended to demonstrate that the methods and algorithms used in the calculational code and mode are correctly implemented. In other words, that no bugs exist in the coding.

Validation is intended to demonstrate that a simulation code has the necessary algorithms and features for providing accurate answers for a class of problems.

A verified and validated code does not guarantee that the results are correct. The user (or analyst) must also be knowledgeable, generate suitable models, and be able to interpret the simulation results properly.

ES3 has verified our methods and algorithms in structural dynamics using two approaches. First, we compare our numerical results with cased form solutions whenever possible. Second, we verify that first principles are satisfied. These include the conservation of momentum, angular momentum, and energy.

ES3 has validated our methods for the range of problems in which we are currently working. Validation is, however, a continuing process whenever new classes of problems are considered. Validation requires:

• The generation of analytical models
• Accurate experimental data from tests that are representative of the range of applications of interest, and
• Development of the necessary analytical skills to detect errors in the model and interpret results correctly.

Hence, a validated and verified code is necessary, but the analyst must still ensure that the model is correct.