Application of Bayesian Networks to Predict the Cascading Effects of COVID-19 Restrictions on the Planting Activities of Smallholder Farmers in Uganda

Context: There are rising concerns over the cascading effects induced by COVID-19 restrictions on the planting activities of smallholder farmers in low- and middle-income countries, which may become a non-negligible threat to the long-term food supply and food security. Studies that utilize probability based models to examine the effects of COVID-19 restrictions on planting activities of smallholder farmers in Uganda are still limited.Objectives: To develop a Bayesian network (BN) model based on expert knowledge, existing literature and Uganda’s High Frequency Phone Survey datasets on COVID-19 to bridge this gap..Methods: A comprehensive survey of relevant literature on the effects of COVID-19 restrictions on the planting activities of smallholder farmers was conducted. In total, 16 relevant publications were obtained and imported in Mendeley referencing software. A systematic literature review was later carried out on the 16 publications to identify the documented effects of COVID-19 restrictions on the planting activities of farmers. A total of 12 independent explanatory variables were extracted and used to generate an influence diagram. The influence diagram was used to develop the BN model. A data file containing 6,313 households aggregated from Round 1, 4 and 7 of the Uganda’s High Frequency Phone Survey datasets on COVID-19 was used to calibrate and validate the model using Netica software.Results and Conclusions: The model's error rate was 17.9%% implying that the model had the majority of its predictions correct (82.1%) for the cascading effects of COVID-19 restriction on the planting activities of smallholder farmers in Uganda. The model's spherical payoff was 0.84 with the logarithmic and quadratic losses of 0.45 and 0.29 respectively, indicating a strong predictive power. Model results indicated that the variables of ‘abandoned crop farming’, ‘advised to stay home’ and ‘planted more crop varieties’ were the top 3 factors causing the largest entropy reduction on the planting activities of smallholder farmers. Additionally, lack of access to seeds (2.6 percentage points), fertilizers (1.3 percentage points), travel restrictions (11 percentage points) as well as reduced labour availability (1 percentage point) affected greatly the planting activities of small holder farmers during COVID-19.Significance: The BN model developed was highly accurate and enabled the complex COVID-19 cascading effects to be conceptualized and integrated in order to identify key specific factors that had effect on the planting activities. The study lays a foundation for the future development of advanced dynamic models on the cascading effects of COVID-19 on agriculture.