This paper investigates whether the neural network modeling technique can be confidently relied upon to meaningfully explore variable relationships within acquisition business datasets. The paper uses a direct action methodology to develop and test a prototype Acquisition Decision Support Tool based on a Cognitive Learning Application Framework, and tests the application’s feasibility using acquisition test data sourced from a variety of acquisition business system archives. The test hypotheses use a correlated performance-based analysis approach. The framework and resulting repositories facilitate the fusing of multiple data sources with the goal of creating an Acquisition Decision Support Tool and Business Intelligence repository for rapid dashboard development and an ad-hoc query capability. The primary focus for model development and predictability in this paper will pertain to the area of contract structure and terms relative to vendor performance.
Click here to read our white paper.