Modeling Farming Systems to Understand Potential Outcomes of Agricultural Interventions


by Jessica E. Raneri, Natalia Estrada Carmona, Carl Timler, Lenora Ditzler and Jeroen Groot | February 18, 2020

Home garden harvesting. Photo: J.Raneri/Bioversity

We often hear that solving the enormous and wicked problems our society faces requires transdisciplinary research, holistic and system-oriented approaches, understanding trade-offs, using robust metrics, and anticipating unintended consequences of interventions. Modelling systems can help bring all these requests together and address these challenges.

One such challenge lies in identifying desirable development pathways for smallholder farmers, for whom producing food of sufficient quantity, quality, and diversity is a key objective. These farms are often managed by multiple household members, who have additional activities and income sources besides farming. More and better tools and methods are needed to address the complex and sizable problems these households face.

Three recent publications show advances in the analysis of smallholder farms and households that depend on farming for their food and nutrition security and livelihoods. We extended and used the whole-farm model FarmDESIGN in combination with other methods such as participatory approaches to explore opportunities to improve productivity, environment, income, nutrition and equity.

  • Bio-economic whole-farm models are a popular tool used by many agricultural systems scientists to understand the current performance of farms and to predict the impacts of development pathways or interventions on farm sustainability indicators, yet many of these models do not sufficiently account for the farm household. In the first study, we expand and improve an already existing whole-farm model, FarmDESIGN, to include new budget, labor, and nutrition modules which together reveal the role of the household in farm management decisions. The new modules more effectively model farming systems in which resources such as cash, labor, and food flow between the farm enterprise, the farm household, and beyond the farm gate. The FarmDESIGN model now reflects the centrality of the household in smallholder farm management, making it a better tool for conducting modeling explorations, optimization routines, and scenario analyses. We illustrate the model’s new capabilities by using it to compare two distinct smallholder farm households in Northwest Vietnam, where we conducted a multi-objective optimization to identify options for improving the farm households’ current performance on key sustainability, nutrition and livelihood indicators. We found that the modeled optimization results reflected the heterogeneity between farm households, highlighting trade-offs between household and farm enterprise objectives on both farms.
  • Agricultural development needs to be more nutrition-sensitive, ensuring new ideas for farming systems changes that aim to achieve sustainable development take household nutrition and dietary diversity into account. Many people in rural communities in South Asia and sub-Saharan Africa suffer from undernutrition, particularly deficiencies in key micronutrients such as vitamin A, zinc, and iron – the result of many years’ focus on cereal-based development interventions. We show in this second article how the FarmDESIGN model can examine changes to farming systems in these communities, taking into account multiple objectives at the same time (Socio-Economic, Productivity, Environmental and Nutritional). The model produces ‘Solution Spaces’ that comprise numerous, different possible configurations of future farming systems, starting from the current configuration. These multiple solutions reveal potential trade-offs and/or synergies between objectives and create an enabling environment for discussion with and between farmers looking to make changes in their farming systems. We show this using four separate farms along a market orientation gradient from subsistence farmers to market gardening in western Kenya and north-western Vietnam, respectively. We present the differences between the modelled diets of the four farms and demonstrate how diversification with novel traditional vegetables can increase the number of people whose vitamin A requirement can be met by on-farm production. We find possible synergies with vitamin A production and greater incomes, but are also trade-offs with greater labour requirements for these solutions. We conclude that the nutrition-sensitive, multi-method approaches demonstrated have the potential to identify solutions that simultaneously improve household income, nutrition and resource management in these vulnerable smallholder farming systems.
  • In the third study, “A model-based exploration of farm-household livelihood and nutrition indicators to guide nutrition-sensitive agriculture interventions,” a diverse team of researchers showed through a pragmatic case study the applicability of FarmDESIGN, a model developed to support the demanded aid. We tested the potential contribution of a nutrition-sensitive intervention targeting the diversification of home gardens, aimed at reducing nutritional gaps and improving livelihoods in rural Vietnam. In addition, we analysed how various nutrition-related metrics can capture how diversification from home-gardens can help satisfy household dietary requirements. From this research, we derived potential development pathways for stakeholder-selected interventions to simultaneously improve nutrition security and livelihoods. It allowed us to establish the need for coupling home garden diversification with other interventions to facilitate changing behaviour and preferences on crop production and consumption, by linking to cooking and producing food. We showed the importance of using various and new nutrition-related metrics in research and development projects, making clear that. the applicability of model-based exploration is therefore beyond the case study and nutrition-sensitive interventions.

Through each of these studies, we show how a modeling tool can be an effective method for decision makers as they seek to understand trade-offs and potential outcomes of agricultural interventions at the household level. The work, driven by A4NH and a focus on improving nutrition, also demonstrated how addressing issues in a way that cuts across CGIAR research programs can complement and enhance commodity-based research done by CGIAR and elsewhere.

This post was developed collaboratively by Jessica E. Raneri, Natalia Estrada Carmona, Carl Timler, Lenora Ditzler and Jeroen Groot.

This approach was applied in case studies within the CGIAR Research Program (CRP) on Agriculture for Nutrition and Health (A4NH), but also in collaboration with CRPs MAIZE; WHEAT; Roots, Tubers, and Bananas; RICE; Water, Land, and Ecosystems; and Humidtropics.


Can employer-provided on-site meals provide an avenue to improve diets and catalyze food system changes?


The authors map national food system transformations in a recently developed typology, using economic, social, dietary, and environmental outcomes to examine whether there are linear patterns as countries move from one categorization to another.


PhD candidates with A4NH's Food Systems for Healthier Diets research flagship reflect on what they learned about engaging in national food system transformation during the course of their study.