Soybean varieties portfolio optimisation based on yield prediction

[Display omitted] •A strategy for soybean seed selection is proposed.•Yield is predicted using a novel approach – Weighted Histograms Regression.•Portfolio optimisation of seed varieties is conducted. One of the biggest problems in agriculture is concerned with seed selection. Wrong choice of seed v...

Full description

Saved in:
Bibliographic Details
Published in:Computers and electronics in agriculture 2016-09, Vol.127, p.467-474
Main Authors: Marko, Oskar, Brdar, Sanja, Panic, Marko, Lugonja, Predrag, Crnojevic, Vladimir
Format: Article
Language:English
Subjects:
Agriculture
Algorithms
Convex optimization
Optimization
Portfolio optimization
Regression
Seasons
Seed selection
Seeds
Soybeans
Strategy
Weighted histograms
Yield prediction
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:[Display omitted] •A strategy for soybean seed selection is proposed.•Yield is predicted using a novel approach – Weighted Histograms Regression.•Portfolio optimisation of seed varieties is conducted. One of the biggest problems in agriculture is concerned with seed selection. Wrong choice of seed variety cannot be compensated with fertilisation, spraying or the use of mechanisation later in the season. The purpose of this work was to design the strategy for selecting soybean varieties that should be planted on the test farm in order to maximise yield in the following season, based on the knowledge acquired from heterogeneous historical data. We propose weighted histograms regression to predict the yield of different varieties and compare our method to conventional regression algorithms. Based on the predicted yield, we perform portfolio optimisation to come up with the optimal selection of seed varieties that is to be planted. Presented algorithms and results were produced within the Syngenta Crop Challenge.
ISSN:0168-1699
1872-7107
DOI:10.1016/j.compag.2016.07.009