A Methodology for Quantitative Comparison of Experimental to Predicted Chemical Deposition Data
Final rept. Oct 1989-May 1990
ARMY BALLISTIC RESEARCH LAB ABERDEEN PROVING GROUND MD
Pagination or Media Count:
Evaluation techniques were developed to assess computer simulation of single chemical munition field trials. These techniques were used to quantitatively compare the predicted to experimental deposition data. The methods employed for evaluating single munitions are not appropriate for the multiple munition field trials that are currently being conducted. Hence, the data fitting techniques were re-examined and new methodology developed that could be applied regardless of the number of munitions tested. The statistical tests chosen were the Sign, Mann-Whitney U, and Chi-Square tests. An additional metric was a modified version of the Chi-Square, referred to as the Point-by- Point. Each test was exercised using data from eleven field trials to test its appropriateness for this application. A bet-fit criterion was established based on the evaluation of the test results. Keywords Chemical simulant Chemical field trial Hypothesis tests Area dosage plots Quantitative analysis.
- Chemical, Biological and Radiological Warfare