skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Monte Carlo simulation of the sequential probability ratio test for radiation monitoring

Conference ·
OSTI ID:6395257

A computer program simulates the Sequential Probability Ratio Test (SPRT) using Monte Carlo techniques. The program, SEQTEST, performs random-number sampling of either a Poisson or normal distribution to simulate radiation monitoring data. The results are in terms of the detection probabilities and the average time required for a trial. The computed SPRT results can be compared with tabulated single interval test (SIT) values to determine the better statistical test for particular monitoring applications. Use of the SPRT in a hand-and-foot alpha monitor shows that the SPRT provides better detection probabilities while generally requiring less counting time. Calculations are also performed for a monitor where the SPRT is not permitted to the take longer than the single interval test. Although the performance of the SPRT is degraded by this restriction, the detection probabilities are still similar to the SIT values, and average counting times are always less than 75% of the SIT time. Some optimal conditions for use of the SPRT are described. The SPRT should be the test of choice in many radiation monitoring situations. 6 references, 8 figures, 1 table.

Research Organization:
Los Alamos National Lab., NM (USA)
DOE Contract Number:
W-7405-ENG-36
OSTI ID:
6395257
Report Number(s):
LA-UR-84-3380; CONF-841007-13; ON: DE85002409
Resource Relation:
Conference: Nuclear science symposium, Orlando, FL, USA, 31 Oct 1984
Country of Publication:
United States
Language:
English