In aerospace industry, the concept of Integrated Vehicle Health Management (IVHM) has gained momentum and is becoming need of the hour for entire value chain in the industry. Identifying and sensing right parameters at right time is the key for success of IVHM. It has opened up challenge to the sensor providers to make sensors smarter, self-contained to compute and communicate efficiently. The expected benefits of lesser time to maintenance, reduced operating cost and very busy airports are motivating aircraft manufacturers to come up with tools, techniques and technologies to enable advanced diagnostic and prognostic systems in aircrafts. These features not only enable detection of failures but also support prediction of tentative failures upfront based on historical data, trend analysis and estimating the future trends. At present, various groups are working on different systems and platforms for health monitoring of an aircraft e.g. SHM (Structural Health Monitoring), PHM (Prognostics Health Monitoring), AHM (Aircraft Health Monitoring), EHM (Engine Health Monitoring) and so on. However, these approaches are mostly restricted to federated architecture where faults and failures for standalone line replaceable units (LRUs) are logged inside the unit in fault storage area and are retrieved explicitly using maintenance based applications for fault and failure diagnostics. With the transformational growth in computing technology, one can easily visualize the possibilities of moving from present federated architecture to integrated architecture in near future. The advanced analytical methods like statistical trend analysers and estimation techniques can be used to process the data from multiple sources to predict the performance of a component, subsystem or a system. The advanced tools help in predicting a possible failure upfront, making critical decisions and scheduling preventive maintenance besides enabling pilot with processed information leading to better decision capabilities during flight. In this paper a qualitative analysis of existing systems is presented and an integrated architecture consisting of data acquisition, data processing, data fusion with capabilities to perform diagnostics, prognostics and decision making is proposed. A framework and a platform required to perform computation intensive prognostics and diagnostic applications which would support decisions for condition based maintenance is presented. This paper also brings out a methodology to run on- board diagnostics and generate inflight warning to the crew.