The Application of Mathematical Modelling as a Design Solution for Affordable Nutrition

Abstract

Optimal nutrition is critical across all demographic groups and at different stages of life, from infancy to adolescence and adulthood. It is also an important component of sustainable diets. Sustainable diets are defined as diets with four main dimensions: affordability, nutrition, acceptability, and environment. However, access to sustainable diets is often constrained by cost factors and a lack of tools to make informed decisions. The study aimed to investigate the complex challenge of achieving sustainable diets that address affordability, cultural acceptability, nutritional adequacy, and environmental impact across different demographics and within the food product development industry from locally available foods in Ghana. Adolescent nutrition is crucial as it directly influences their health and long-term well-being, thus, the study initially designed affordable and culturally acceptable food baskets for adolescents in Ghana, addressing the economic (cost) constraints that hinder access to nutritious diets. Using a linear programming approach, subject to nutritional and cultural acceptability constraints, the study identified locally available ingredients that met nutritional requirements at minimal costs. The estimated cost of the optimised food baskets for adolescent males and females was GHS 15.70 and 15.57, respectively, for the day. They included fifteen food items each and fulfilled nutrient requirements recommended by the FAO & WHO and the Institute of Medicines. The study also considered the nutritional needs of pregnant and lactating women within the context of environmentally sustainable diets. Linear programming was employed to design nutritionally adequate diets with low greenhouse gas emissions (GHGEs). The optimised diets for pregnant and lactating women, comprising 19 and 21 food items respectively, achieved nutrient adequacy with GHGE values of 1.06 and 1.48 kg CO2 eq./day. The diets highlighted a shift towards plant-based foods, legumes, seeds, and nuts, reducing reliance on staples and red meat. Utilising a bi-objective optimisation method, the epsilon constraint technique was adopted to minimise cost and GHGE simultaneously in recognition of the significant contribution of food systems to global greenhouse gas emissions. The target consumers were adult males and females. The study aimed to simultaneously balance cost and environmental impact while ensuring nutritional and cultural acceptability. Results indicated a strong negative correlation between cost and GHGE, with correlation values of -0.93 and -0.95 for adult males and females respectively. The study also noted that achieving optimal diets requires a mix of nutrient-rich and environmentally friendly food choices, highlighting the need for dietary adjustments to balance economic and environmental goals. To facilitate sustainable food product development, the study developed the "NourishCraft Graphical User Interface (GUI)," a user-friendly interface designed to aid sustainable product formulation. Leveraging Streamlit and Python libraries, the interface allows users to formulate nutritious food products considering affordability, nutrition, cultural preferences, and environmental impact. Users can choose to prioritise minimal cost, minimal greenhouse gas emissions, or a balanced approach, and define constraints for preferred nutritional parameters. The interface provides a list of ingredients that meet the defined nutritional parameters at the minimum objective value (cost, GHGE, and/ or both). Additional outputs include tables and figures, making it a practical tool for product developers to integrate sustainability into their processes. To validate the NourishCraft interface, it was used to formulate a product, using one-third of the recommended dietary allowance (RDA) for adolescents 10 to 18 years, setting energy, carbohydrate, protein, and iron as the parameters. This formulation was set to measure the dimensions of affordability, nutrition, and acceptability because GHGE could not be assessed due to limited data. The ingredients from the formulation were purchased for cost assessment, and acceptability was evaluated by a consumer sensory test. The interface proposed output of cassava, coconut, millet, and Bambara beans in proportions of 27.64%, 14.2%, 24.44% and 32.71%, respectively at GHS 0.67. The drum-dried composite developed from the formulation contained 300 kcal of energy, 45 g carbohydrates, 10 g protein, and 4.29 mg of iron. The model predicted between 61.02% - 123.63% of the nutrient parameters. However, the actual cost of the ingredients purchased from the market amounted to GHS 1.48, showing a 45.27% prediction of the formulated cost by the interface. The product was well accepted with an overall likeness score of 7.82 ± 1.21. The results indicated that the NourishCraft graphical user interface has the potential to support nutritious and affordable food formulations. The findings from these studies hold significant implications for policymakers, individuals, researchers, and the food industry. Policymakers can develop programs that promote affordable, nutritious, and sustainable food practices. Individuals can make informed dietary choices that prioritise health and environmental well-being. Researchers can further explore the complexities of sustainable food systems, and the food industry can utilise the "NourishCraft GUI" to aid in designing sustainable products. By integrating these considerations, stakeholders across the food system can contribute to a future where healthy and sustainable diets are accessible to all.