An evaluation of aggregate functions in relational database system considerable impact on performance in many application areas like geographic information systems and statistical and scientific databases. The problem with existing systems is inefficient execution of aggregate functions with large data volumes and lack of flexibility. It is not possible to extend the systems with new aggregate (average) functions. In this paper we extend the concept of aggregate functions from relational databases. This paper introduces the concept of average functions which could enhance flexibility and performance considerably. We show how this could be implemented into a database. Our implemented technique easily detects any changing in replicated data and original data with considering Null values.