(361d) Comparing the Environmental Implications of National Food Supplies Using Data Envelopment Analysis | AIChE

(361d) Comparing the Environmental Implications of National Food Supplies Using Data Envelopment Analysis


Lucas, E. - Presenter, Imperial College London
Guillén-Gosálbez, G., Imperial College London
Guo, M., Imperial College London
Pozo Fernández, C., Imperial College London
Galán Martín, Á., Imperial College of Science, Technology and Medicine
The three most fundamental resources to basic human needs are water, energy, and food. As conceptualised by Maslow’s Hierarchy of Needs, the first tier that must be primarily realized are basic physiological requirements (food, water, warmth, rest). Until these have been met, individuals are unable to progress up the hierarchy pyramid and fulfil needs of a more complex nature, such as safety and security (Maslow, 1943).

Sustainable use and management of natural resources for clean water, energy and food provisioning is therefore required to ensure the physiological requirements are met for the world’s population. The challenges, interdependencies and connections between water, energy and food resource use form the Water-Energy-Food Nexus (WEFN) and achieving WEFN sustainability is inextricably linked to human well-being.

The food production, supply and distribution system has the strongest relationship with human and environmental health, as its sole purpose is to feed and nourish people whilst accounting for 70% of global water withdrawals and 37% of land use. Though the food system’s purpose is seemingly simple, current practices and structure has led to over 820 million people facing chronic food deprivation (FAO, 2018) and a rise in diet-related non-communicable diseases, as well as significant contributions to environmental degradation (Tilman & Clark, 2014, Willett et al., 2019).

The existence of two malnourishment extremes highlights the double-ended shortcomings of the current food system. Chronic food deprivation and undernutrition is a symptom of the food system’s lack of capacity to ensure universal access and supply of calories. On the other hand, the rise in diet-related non-communicable diseases can be attributed to the failure of today’s food system to adequately nourish people with nutrient-dense food supplies (Nelson et al., 2018).

In addition to existing malnutrition challenges, the estimated global population growth to 9 billion by 2050 means the food system must increase its capacity to provide for every person. It must do this whilst also undergoing reform to reduce its contribution to greenhouse gas emissions, freshwater withdrawal demand, soil degradation, land use change, and water eutrophication and acidification. One key strategy is through a shift towards diets consisting of foods with lower environmental footprints (Davis et al., 2016, Bajželj et al., 2014).

In order to facilitate widespread dietary shifts, that alleviate malnourishment and environmental degradation problems, an improved understanding of food supply patterns across countries and regions is crucial. Analysis of global food supply can be combined with an assessment of the food system’s performance at providing calories and nutrition at current levels of environmental impact. A performance indicator suitable for this assessment can take the form of a benefit/cost efficiency, with cost being the environmental footprint and benefit as number of calories or nutritional quality.

The environmental footprint of food production and distribution covers a range of indicators that includes land use, greenhouse gas emissions, acidification, eutrophication and freshwater withdrawals. Multiple ‘costs’ therefore introduce a challenge in calculating a single performance indicator (Chaudhary et al., 2018, Gustafson et al., 2016). This challenge can be overcome through the application of Data Envelopment Analysis (DEA), a non-parametric method to calculate relative efficiency considering entities that manage multiple inputs and outputs, formulated as a linear programming optimisation problem (Charnes et al., 1986).

In the DEA model formulation, a country is considered as a ‘decision-making unit’. The country’s food supply per capita, combined with different food’s environmental impacts (Poore and Nemecek, 2018), is used to estimate the environmental ‘costs’ (inputs) required to produce and supply that particular composition of food. The five environmental ‘costs’ (land use, greenhouse gas emissions, acidification, eutrophication and freshwater withdrawals) are considered as one input through a weighted sum. The assignment of subjective weightings is avoided as weighting values are optimised by the DEA model.

The number of calories and nutritional quality associated with each country’s food supply per capita are considered as outputs. Nutritional quality is based on the food supply’s associated Qualifying Nutrient Balance Score (NBS) (Fern et al., 2015) using estimated supply of nutrients per capita (Smith et al., 2016). The NBS represents the extent to which the per capita food supply can satisfy the daily requirements for a set of nutrients. This includes micronutrients that are of widespread public health concern due to prevalent deficiencies.

In order to address the two extremes of malnourishment caused by current food supplies, this study undertakes two parallel DEA assessments that differ by the type of efficiency it calculates. One type of efficiency measures how a country’s food supply makes efficient use of environmental resource inputs to supply calories, and the other to supply nutritional quality (embodied by the NBS). Two efficiency indicators are generated for 139 countries with a reference year of 2010. Therefore, relative ‘efficient’ and ‘inefficient’ countries are identified, along with improvement targets.

In addition to the assessment study using DEA, statistical analysis can be undertaken to identify significant correlations between the inputs, outputs and national income datasets. Insightful trends between calorie supply, NBS and Gross National Income (GNI) per capita across countries can be found by applying Spearman’s rank-order correlation.

There exists a strong positive correlation between number of calories supplied per capita and GNI per capita (ρ = 0.69), whilst a strong negative correlation is found between NBS of per capita food supply and GNI per capita (ρ = -0.69). There is also a strong positive correlation found between environmental impact level per 2000-kcal sample and GNI per capita, namely greenhouse gas emissions, acidification and eutrophication (ρ = 0.62, 0.68, 0.60). In addition to variation across national income, clustering analysis is applied to identify groups of countries with similar food supply compositions.

The resultant set of efficient units identified by the two DEA models are different. When countries’ efficiencies are assessed using the associated NBS of its food supply, efficient countries are from low and lower-middle income groups. The most inefficient units identified are all high-income countries with food supply compositions attributed to the same cluster. Contrastingly, efficiencies assessed using number of calories supplied result in a larger set of efficient countries that span across all income groups.

This study highlights the complexity of the food system by showing the difference between supplying calories and providing nutrition. Through the novel application of DEA, insight is provided into which countries are of interest in the context of improving environmental and nutritional sustainability of food supplies.


Bajželj, B., Richards, K. S., Allwood, J. M., Smith, P., Dennis, J. S., Curmi, E., & Gilligan, C. A. (2014). Importance of food-demand management for climate mitigation. Nature Climate Change, 4(10), 924–929. https://doi.org/10.1038/nclimate2353

Charnes, A., Cooper, W. W., & Thrall, R. M. (1986). Classifying and characterizing efficiencies and inefficiencies in data development analysis. Operations Research Letters, 5(3), 105–110. https://doi.org/10.1016/0167-6377(86)90082-9

Chaudhary, A., Gustafson, D., & Mathys, A. (2018). Multi-indicator sustainability assessment of global food systems. Nature Communications, 9(1), 848. https://doi.org/10.1038/s41467-018-03308-7

Davis, K. F., Gephart, J. A., Emery, K. A., Leach, A. M., Galloway, J. N., & D’Odorico, P. (2016). Meeting future food demand with current agricultural resources. Global Environmental Change, 39, 125–132. https://doi.org/10.1016/J.GLOENVCHA.2016.05.004

Fern, E. B., Watzke, H., Barclay, D. V, Roulin, A., & Drewnowski, A. (2015). The nutrient balance concept: A new quality metric for composite meals and diets. PLoS ONE, 10(7). https://doi.org/10.1371/journal.pone.0130491

Gustafson, D., Gutman, A., Leet, W., Drewnowski, A., Fanzo, J., Ingram, J., ... Ingram, J. (2016). Seven Food System Metrics of Sustainable Nutrition Security. Sustainability, 8(3), 196. https://doi.org/10.3390/su8030196

Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50(4), 370-396. http://dx.doi.org/10.1037/h0054346

Nelson, G., Bogard, J., Lividini, K., Arsenault, J., Riley, M., Sulser, T. B., ... Rosegrant, M. (2018). Income growth and climate change effects on global nutrition security to mid-century. Nature Sustainability, 1(12), 773–781. https://doi.org/10.1038/s41893-018-0192-z

Poore, J., & Nemecek, T. (2018). Reducing food’s environmental impacts through producers and consumers. Science, (360), 987–992. Retrieved from http://science.sciencemag.org/

Smith, M. R., Micha, R., Golden, C. D., Mozaffarian, D., Myers, S. S., & Obukhov, A. G. (2016). Global Expanded Nutrient Supply (GENuS) Model: A New Method for Estimating the Global Dietary Supply of Nutrients. PLoS ONE, 11(1), 146976. https://doi.org/10.1371/journal.pone.0146976

Tilman, D., & Clark, M. (2014). Global diets link environmental sustainability and human health. Nature, 515(7528), 518–522. https://doi.org/10.1038/nature13959

Willett, W., Rockström, J., Loken, B., Springmann, M., Lang, T., Vermeulen, S., ... L Murray, C. J. (2019). The Lancet Commissions Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. The Lancet Commissions. https://doi.org/10.1016/S0140-6736(18)31788-4