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

Title: Variance propagation by simulation (VPSim)

Conference ·
DOI:https://doi.org/10.2172/569135· OSTI ID:628999

The application of propagation of variance (POV) for estimating the variance of a material balance is straightforward but tedious. Several computer codes exist today to help perform POV. Examples include MAWST (`materials accounting with sequential testing,` used by some Department of Energy sites) and VP (`variance propagation,` used for training). Also, some sites have such simple error models that custom `spreadsheet like` calculations are adequate. Any software to perform POV will have its strengths and weaknesses. A main disadvantage of MAWST is probably its limited form of error models. This limited form forces the user to use cryptic pseudo measurements to effectively extend the allowed error models. A common example is to include sampling error in the total random error by dividing the actual measurement into two pseudo measurements. Because POV can be tedious and input files can be presented in multiple ways to MAWST, it is valuable to have an alternative method to compare results. This paper describes a new code, VPSim, that uses Monte Carlo simulation to do POV. VPSim does not need to rely on pseudo measurements. It is written in C++, runs under Windows NT, and has a user friendly interface. VPSim has been tested on several example problems, and in this paper we compare its results to results from MAWST. We also describe its error models and indicate the structure of its input files. A main disadvantage of VPSim is its long run times. If many simulations are required (20,000 or more, repeated two or more times) and if each balance period has many (10,000 or more) measurements, then run times can be one-half hour or more. For small and modest sized problems, run times are a few minutes. The main advantage of VPSim is that its input files are simple to construct, and therefore also are relatively easy to inspect.

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:
628999
Report Number(s):
LA-UR-97-2549; CONF-970744-; ON: DE97008985; TRN: AD-a340 778
Resource Relation:
Conference: 38. annual meeting of the Institute of Nuclear Materials management, Phoenix, AZ (United States), 20-24 Jul 1997; Other Information: PBD: Jul 1997
Country of Publication:
United States
Language:
English