Current challenges and forecasts of grain corn production and con-sumption in Mexico

Abstract

ABSTRACT

Objective: Characterize the production and consumption of grain corn in Mexico, through time series and artificial intelligence models to determine present and future challenges.

Design/methodology/approach: Key variables were analyzed in graphs and maps created in Excel^® and SCImago Graphica^®, respectively. Forecasts for the year 2050 were obtained in Python^® with Recurrent Neural Networks (RNN) of the Long Short-Term Memory (LSTM) type and compared with the years 1980 and 2020.

Results: The largest production of white and yellow grain corn was obtained by the United States and China; Mexico ranks seventh, is not competitive in exports and depends on imports of yellow grain corn from the United States to supply demand. The states that implemented technological packages showed the highest yields and production. By 2050, grain corn production in Mexico will increase thanks to the technological advances of Agriculture 5.0, although it will not be enough to supply the apparent consumption of the population with a growing trend, so imports will increase.

Limitations on study/implications: Analysis of the possible future, created from time series and RNN-LSTM, helps guide decision making in the present.

Findings/conclusions: New agricultural public policies are needed that guide, in the long term, the challenges of the production and consumption of grain corn in Mexico to guarantee food sovereignty.

Keywords: Yields, imports, exports, time series, deep neural networks.