No abstract

Current food consumption patterns have a substantial impact on our environment and are thus considered unsustainable. In the context of global warming and a rising world population, shifting from meat towards more plant-based products holds potential to reduce the environmental impact of our food consumption. Replacing meat in the diet, however, requires compensation through other products that are able to provide the important nutrients present in meat (protein, iron, zinc and vitamin B12). This study applies linear programming techniques with the aim to compose meat replacers, with equivalent nutritional value to meat (using chicken and beef as a reference), minimizing the environmental impact with regards to climate change, land use, water use and fossil fuel depletion. A life cycle approach was used to quantify the environmental impacts. Particular attention was given to protein quality and the relative quantities of essential amino acids. The results show that soy is a preferred ingredient due to its favorable amino acid profile. Among the different scenarios investigated, the vegan replacers, with reductions of up to 87%, have the largest potential for impact reduction for all indicators except water use. Insect-based replacers have the largest potential for water use reduction (up to 47%), but show relatively high fossil fuel depletion values. The smallest improvement potential is observed with regards to fossil fuel depletion, with some values even 45% higher than the values for meat. Furthermore, it is not possible to obtain equivalent nutritional values to beef without using fortifications.

No abstract

No abstract

MinSum achievement functions can give rise to solutions that are sensitive to weight changes, and that pile all unwanted deviations on a limited number of nutritional constraints. MinMax achievement functions spread the unwanted deviations as evenly as possible, but may create many (small) deviations. EGP comprises both types of achievement functions, as well as compromises between them. It can thus, from one data set, find a range of solutions with various properties.