Continuous monitoring of changes to utility services and products in a distributed information system is an interesting issue in software engineering. These changes affect the semantics and structural complexity of the system, as a change to one part will in most cases, result in changes to other parts. Therefore, in design and redesign for customization, predicting this change presents a significant challenge. Changes are intended to fix faults, improve or update products and services. Lack of validated, widely accepted, and adopted tools for planning, estimating, and performing maintenance contributes to the problem. One effective way of assessing changeability effect is to assess the impact of changes through a well validated model and framework. This research paper is an extended report on the implementation of a change propagation framework, together with it's associated change impact analysis factor adaptation model, and a fault and failure assumption model to predict the effect of a change of a service in a grid environment. While implementing the framework, data was collected for three hypothetical years, thus helping to predict the next two (2) years consecutively. Significant results corresponding to the impact analysis factor were obtained showing the viable practicality of the use of Bayesian statistics (as against unreported regression method) satisfying best-fit prediction. We conclude that, the higher the number of dependent services on a faulty service requiring a change, the higher the impact due to fault propagation.