INTRODUCTION

The purpose of this document is to provide information on the applicability of the results produced by the MOSES software package when it is being applied to "real world" problems. We begin by considering some simple problems where exact solutions are available. However, many problems of interest are not amenable to such checks. For these cases, we turn to comparisons with model tests and/or other computed results.

Although not strictly applicable, it should be mentioned that MOSES has been around in some incarnation since the late 1970s and it has been used on hundreds of projects. This is, perhaps, the best statement regarding the accuracy and reliability of the results it produces.

Because of the range of problems that the MOSES software can solve, it is impossible to compare everything. What we have here are a sufficient number of comparisons to establish that the "basic ingredients" function correctly. Thus, they give comfort that MOSES functions properly for a much wider class of problems than covered here.

The document finishes with a description of our Quality Assurance procedures. In essence, these procedures are designed to minimize the chance of errors creeping into the software during the maintenance and development process.