Probabilistic model evaluation of continuous air monitor response for meeting radiation protection goals.

Effective continuous air monitor (CAM) programs can eliminate or significantly reduce the amount of inhaled radioactive material following an accidental release. Numerous factors impact the levels of protection CAM programs provide to the workers during these releases. These factors range from those related to the capability of the CAM instrument (e.g., CAM alarm set point and length of counting intervals) to those related to CAM placement in the room relative to dispersion rates and patterns of the released material in a room. While the impact of many of these factors on alarm sensitivity has been investigated in isolation, there are no methods for holistic evaluations of CAM programs relative to radiation protection goals (RPGs) or the contribution of the factors, either individually or combined, toward limiting worker dose. In this study, worker exposure was predicted using CAM response models developed to evaluate protection levels for continuous and acute releases. Monte Carlo simulations of 10,000 releases were performed using various combinations of model parameter values, with associated uncertainty distributions, to assess the expected ability of a CAM program to meet RPGs, and, further, to assess the relative influence of each factor toward lowering worker exposure. Results showed that improvements to CAM instrument capability combined with better ventilation and CAM placement improve worker protection nonlinearly and that these improvements are critical to meet RPGs. The sensitivity analysis showed that ventilation-driven dilution had the greatest impact on exposure reduction with the selected counting interval for alarm decisions and the alarm set point as secondarily important.