Developing intelligence in food security: An agent-based modelling approach of Qatar's food system interactions under socio-economic and environmental considerations

Maintaining the consistent provision of food is becoming a persistent challenge in the light of increasing demands for resources and the multiple risks governing food systems. These include population growth, market volatilities, changing climate and the lack of coordination amongst involved stakeholders. Dynamic and interactive simulation models have demonstrated their utility for developing possible scenarios related to the food sector's subsystems, allowing stakeholders to understand their outcomes, should they occur, and make efficient and sustainable decisions that potentially alleviate the risks. This study proposes an Agent-based model (ABM) that supports informed decision-making through allowing the simulation of multiple scenarios that can potentially influence the food system's strategies and actions. The framework is illustrated considering the economic actors within Qatar's food sector, with a special focus on delivering products in high demand, such as the tomato crop. The model was implemented using the Python-basedMesa library, and it simulates interactions between domestic farmers, local importers, and international exporters to assess the performance of the food sector considering economic, environmental and social scenarios. Scenario (1) describes the baseline case wherein local farmers and importers are acting independently with the aim to maximise their individual profits. Scenario's (2) and (3) represent enhancement plans that foster collaboration, water security and improved food quality. Findings of this study assert that from an economic standpoint, the food system can supply 75% of the market with imported tomatoes, as for the remaining 25%, it can be provided through national farms, with a 76% contribution of greenhouses and 24% open fields. This option offers the minimum economic cost and the lowest GreenhouseGas (GHG) emissions. However, this perspective implies massive food losses and relatively high-water consumption. Scenario 2, which advocates a full reliance on imports offers a costly monthly strategy in terms of economic expenses and environmental emissions, however, it results in considerable water savings that relieve the strain on the local water supply system. When implementing a plan driven by quality consideration (scenario 3), the reliance on imports reaches up to 82%, and the remaining local produce is mostly supplied by greenhouses, with approximately 78%, leading to quality improvement and a substantial decrease in food losses yet at the expense of both the economic and environmental performance. These results confirm that aspiring water security and social welfare while provisioning food products requires tremendous investment and might lead to significant GHG emissions. The framework developed in this study provides a generic and interactive decision-making scheme that can assist in undertaking future decisions considering economic and environmental constraints and market-driven specifications. (C) 2022 Institution of Chemical Engineers. Published by Elsevier Ltd. All rights reserved.

Automated summary

This study proposes an Agent-based model that supports decision-making for the food system by simulating multiple scenarios. The model is applied to Qatar's food sector and evaluates its performance under economic, environmental, and social scenarios. The findings suggest that ensuring food supply requires significant investment and may result in substantial greenhouse gas emissions.