Decreased Labor Costs While Producing 95% Accuracy Level On Yield Prediction Using A Multi-Year Sample Set

CHALLENGE

Improve the efficiency and decrease labor costs associated with yield prediction on citrus crops

The yield prediction process has historically had high labor costs due to the complexity of the data points needed to collect and analyze. Our client wanted to improve the accuracy of their yield prediction while lowering their labor costs.

SOLUTION

Developed an offline mobile mapping solution to navigate trees and accurate collect data from the field

Leveraging existing data sets, an on­line mobile mapping solution was developed to allow low skill workers to e­fficiently, and accurately, navigate to specific trees and capture the data from the field directly into the mobile application.

This entered data could then be exported and the calculation of predicted yield automated.

RESULTS

Yield prediction 95% accurate over a multi-year sample set

Using Agerpoint data and Volumetric methodology, the calculated yield prediction was verified to be 95% accurate over a multi-year sample set of data collected.

This allowed our client to leverage lower skilled (and lower wage) workers to collect data in the field and still receive an extremely accurate prediction. The data analysis complexity and associated eff­ort was almost completely eliminated and allowed the data to become actionable.

In addition, the availability of multi-year data allows for the analytics to potentially quantify, measure, and improve field conditions and yield across blocks (block variability).