To make qualified decisions when extrapolating results from a survey sample with imprecise tests requires careful handling of uncertainty. Both the imprecise test and uncertainty introduced by the sampling have to be taken into account in order to act optimally. This paper formulates an influence diagram with discrete and continuous nodes to handle an example typical for animal production: a veterinarian who as part of a biosecurity program - has to decide whether to treat a herd of animals after inspecting a small fraction of them. Our aim is to investigate the robustness of the obtained strategy by performing a two-way sensitivity analysis with respect to the proportion of false positives and false negatives of the test. Output of the analysis is a treatment map illustrating how the chosen strategy varies according to variation in these proportions. The map helps to investigate whether a certain variation is acceptable or if the test procedure has to be standardized in order to reduce variation. Objective of the paper is to be an appetizer to work more with the issues raised in obtaining a practical solution.
CITATION STYLE
Höhle, M., & Jørgensen, E. (2003). Decision making based on sampled disease occurrence in animal herds. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 2711, pp. 220–229). Springer Verlag. https://doi.org/10.1007/978-3-540-45062-7_18
Mendeley helps you to discover research relevant for your work.