A Bayesian Network for predicting yield response of winter wheat to fungicide programmes

Bayesian Probability Networks (BNs) present a powerful tool for incorporating uncertainty in decision support systems. They provide a basis for probabilistic inference, to calculate the changes in probabilistic belief as new evidence is entered into the model. In practice, for agricultural systems, the construction of a BN is challenging, as there is often insufficient data for computing the prior and conditional probabilities required for the network. This paper illustrates the possible applications of BNs by describing a BN for yield response of winter wheat to spray programmes. It uses experimental data from ADAS trials and makes some assumptions in order to complete these data.