Abstract:
Many comparisons have been made between Chemical detectors (C), between Biological (B) detectors, and between Radiological detectors (R), providing insights to the best C, B and R equipment for a given purpose. However, no comparison has been made between C, B and R systems to appraise how C, B and R detectors perform against each other and where capability gaps lie. The dissertation generates a method to achieve an inter-comparison between C, B and R detection capabilities and identifies where to invest resources to achieve a more effective overall CBR detection architecture. The inter-comparison methodology is based on an operational analysis tool (SMARTS). The overall CBR detection architecture is illustrated through detect to warn and detect to treat mechanisms across the timeline of a realistic scenario. The scenario has been created to be non-prejudicial to C, B or R incidents, deconstructed into four frames to accommodate SMARTS. The most suitable deconstruction is into early warning, personnel security screening, initial response and definitive identification frames. The most suitable detector Key Performance Characteristics (KPCs) are identified for each frame. SMARTS is performed by analysing the current performance of the C, B and R detection systems drawn from the literature and the target requirements determined by defensible logic. The desire to improve each capability from its current state to target requirement is subjectively determined by the author. A sensitivity analysis is applied to mitigate the effect of a limited pool of opinion. Applying the methodology to published CBR detection capability data and the author’s appraisal of the target requirement reveals that B detection requires the greatest development and R the least, and that detection in the security screening and initial response frames falls short of capability compared to early warning and definitive identification frames. Selectivity is a challenge across a broad range of frames and agents. This work provides a methodology that is modular and transparent so that it can be repopulated should new data or alternative perception arises.
