Previsão de demanda de importação de fertilizantes em tempos de pandemia e guerra

Abstract

Brazil is the fourth largest consumer of fertilizers in the world, as agriculture is one of the essential natural activities in the country, which encompasses small family farmers and large agro-industries. During the pandemic, the demand for inputs increased and the increase in the price of fertilizers was driven by an increase in the costs of agricultural crops in the world. Since the beginning of Russia's attacks on Ukraine, prices have skyrocketed, increasing agricultural production costs across the country, which imports 85% of the fertilizers it uses in agriculture. For this reason, the Brazilian Federal Government has announced the launch of a National Fertilizer Plan (PNF), in order to adapt the control between domestic production and imports, and meet the high demand for products. The present work aimed to analyze the effect of the pandemic and war on the forecast of fertilizer import demand, analyzing historical data from all ports in Brazil, whether public or private. Starting with the identification and analysis of time series, demand forecasting methods are compared, namely, simple exponential smoothing, exponential smoothing with trend (Holt model), exponential smoothing with trend and seasonality (Holt-Winters model), Box model – Jenkins (ARIMA) and Arima Seasonal (SARIMA), and define the most appropriate method for processing the information. The results showed that the Arima Seasonal (SARIMA) model was effective and had less error for forecasting demand (EAM= 2.5%, EQM= 3.0% and EPAM= 9.9%) compared to the other models. methods used. Price escalation was already happening in 2021 with the COVID-19 pandemic and the War between Russia and Ukraine showed an even greater influence on fertilizer prices, that is, in the war prices increased much more than in the pandemic.