The Inductive Monitoring System (IMS) software uses data mining techniques to automatically characterize nominal system operation by analyzing archived system data. These nominal characterizations are then used to perform near-realtime system health monitoring or analyze archived system data to detect anomalies in system behavior as compared with previous nominal behavior. To operate most effectively for system monitoring, the archived system data used to build the system models for IMS should contain only nominal operations data. Most available data sets contain contaminated data, data transients, or other data that does not represent nominal operations. Finding and removing these undesirable off-nominal data points manually is an error-prone and time-consuming task. Along with a variety of data extraction and program controls, the IMS graphical user interface (IMS GUI) allows the user to visually examine results of IMS monitoring analyses, and graphically select segments of data and outlier data points using a mouse or similar input device. The GUI will then automatically process the candidate dataset to remove undesired data points, leaving a clean dataset containing nominal data to use for building IMS system monitoring models.

The IMS GUI is intended to facilitate the creation and updating of IMS monitoring knowledge bases (aka system models) and the analysis of system data using those knowledge bases. In addition to extracting and formatting data for use with IMS and running the IMS algorithms and utilities, the IMS GUI provides graphing capabilities for IMS results (both IMS composite deviation scores and individual parameter contributions), and the ability to select and remove outlier data points and off-nominal data segments using the graphical display and an input device such as a computer mouse. This provides a convenient, intuitive, and effective technique for the user to analyze and clean datasets (remove undesired off-nominal points).

IMS does not require examples of anomalous behavior. IMS automatically analyzes nominal system data to form general classes of expected system sensor values. This process enables the software to inductively learn and model nominal system behavior. The generated data classes are then used to build a monitoring knowledge base. In real time, IMS performs monitoring functions, determining and displaying the degree of deviation from nominal performance. IMS trend analyses can detect conditions that may indicate a failure or required system maintenance.

This technology is applicable to most system monitoring applications where IMS or other outlier detection algorithms are applied, including aerospace, transportation, manufacturing, energy, and process monitoring applications.

This work was done by David L. Iverson, William M. Taylor, and Liljana Spirkovska of Ames Research Center. NASA is seeking partners to further develop this technology through joint cooperative research and development. For more information about this technology and to explore opportunities, please contact Antoinette McCoy at This email address is being protected from spambots. You need JavaScript enabled to view it. or 650-604-4270. ARC-16531-1