Anomaly detection applied to a materials control and accounting database
- Los Alamos National Lab., NM (United States); and others
An important component of the national mission of reducing the nuclear danger includes accurate recording of the processing and transportation of nuclear materials. Nuclear material storage facilities, nuclear chemical processing plants, and nuclear fuel fabrication facilities collect and store large amounts of data describing transactions that involve nuclear materials. To maintain confidence in the integrity of these data, it is essential to identify anomalies in the databases. Anomalous data could indicate error, theft, or diversion of material. Yet, because of the complex and diverse nature of the data, analysis and evaluation are extremely tedious. This paper describes the authors work in the development of analysis tools to automate the anomaly detection process for the Material Accountability and Safeguards System (MASS) that tracks and records the activities associated with accountable quantities of nuclear material at Los Alamos National Laboratory. Using existing guidelines that describe valid transactions, the authors have created an expert system that identifies transactions that do not conform to the guidelines. Thus, this expert system can be used to focus the attention of the expert or inspector directly on significant phenomena.
- Research Organization:
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- W-7405-ENG-36
- OSTI ID:
- 102386
- Report Number(s):
- LA-UR-95-2122; CONF-950787-60; ON: DE95016871; TRN: 95:006763
- Resource Relation:
- Conference: 36. annual meeting of the Institute for Nuclear Materials Management, Palm Desert, CA (United States), 9-12 Jul 1995; Other Information: PBD: 1995
- Country of Publication:
- United States
- Language:
- English
Similar Records
Anomaly and error detection in computerized materials control & accountability databases
Detecting errors and anomalies in computerized materials control and accountability databases