Quantifying benefit of IoT in agriculture: Optimal irrigation with yield prediction from big data

Authors

รศ.ดร.กาญจ์นภา อมรัชกุล

Published

Proceedings of the 49th Conference on Computers and Industrial Engineering

Abstract

Consider an intelligent irrigation system (IIS), which processes the sensed data, remotely or locally
from agricultural fields, and produces a closed-loop control signal to fully automate irrigation at
different cultivation areas, owned by an agricult
ural firm. We want to quantify the monetary
value of IIS to the agricultural company, who face
s uncertainties in both demand and yields. Yields
are spatial dependence, and they are affected
by both controllable and uncontrollable factors
such as soil moisture, nutrient, weather and climat
e. We assume that the probability distribution
of the yields depends on the irrigation water,
controlled by the IoT-based irrigation system. We
compare the expected profits with and without
the system. Without the system, the profit is
maximized with respect to
the probability distribution of the demand and the yields. With the IIS,
the yields from different areas are predicted from the sensed data, and optimal irrigation water
is determined analytically. In the numerical illustration, we show that the expected profit gain of
the system is significant, when the yield variabilit
y is large, and the error of the yield prediction
model is small.

https://www.dropbox.com/sh/fw5edhyjuq24wai/AABWalzuxR83YIXweIJBpBJwa?dl=0&preview=CIE49_paper_188.pdf

Quantifying benefit of IoT in agriculture: Optimal irrigation with yield prediction from big data. Beijing, the 49th Conference on Computers and Industrial Engineering (81-91).