Artificial Intelligence-based Recommender for Model-Based Systems Engineering (ARMS)

Recommender systems provide information filtering and retrieval capabilities that utilize various sources of data to infer end-user interests and predict their preferences for a given set of items, with the purpose of offering a typically prioritized list of potentially interesting items. Recommender systems are widely used by commercial applications such as music and video broadcasting platforms, e-commerce sites and social networks, and they have started finding applications in assisting engineering efforts. Specifically, recommender systems can be used in Model Based Systems Engineering (MBSE) applications that focus on creating and exploiting domain models as the primary means of information exchange between engineers, rather than on document-based information exchange. As systems are becoming increasingly complex and interconnected, the document-based artifacts are becoming more challenging to manage. Systems Engineering typically depends on the use of models to address concerns from various domains (e.g., mechanical, optical, and control systems). Due to the large amounts of artifacts and data (e.g., engineering analysis, test, Modeling and Simulation (M&S), and production and fielding data), it is almost infeasible for system engineers to continuously monitor and access all the artifacts, tools, and software components. Thus, an automated software solution to assist fast and accurate information retrieval is much needed for MBSE applications. To address this critical need, we propose Artificial Intelligence-based Recommender System (ARMS) for MBSE assistance. Approved for Public Release | 24-MDA-11673 (8 Jan 24)