Modern precision agriculture equipment enables treating different areas of a field differently, i.e., site specific management. This results in the concept of management zones, which are a compromise between treating a field uniformly and treating every plant individually.

This work presents an algorithm for inferring the management zones for a field based on yield data from multiple years. The algorithm uses a hidden Markov random field model (HMRF) to find regions of the field which likely correspond to the same underlying yield distribution (i.e., “management zones”). These regions are assumed to be the same for each year, but their distributions are allowed to vary with time to account for year-to-year variability (from e.g., weather effects, differing crops). The zone assignments and distributions are estimated using Stochastic Expectation Maximization (SEM) and the maximizer of the posterior marginals (MPM). The underlying assumption of the model and algorithm is that the yields corresponding to a given “management zone” will behave similarly, and therefore derive from the same probability distribution.

An advantage of this method is that it is able to run with only the yield data automatically collected during harvest. Also, this method requires no crop specific calibration or configuration.


Hidden Markov models, Markov random fields, Image segmentation, Expectation-maximization algorithms, Monte Carlo methods, Agriculture, Agricultural machinery

Date of this Version