Creating predictive social impact models of engineered products using synthetic populations

Transformative technologies and products can help solve some of society’s most complex problems. To create these new products and increase the likelihood of desired impact, designers would benefit from understanding and improving a product’s social impacts while still in development. Methods of predicting social impacts have been proposed that estimate a product’s impacts on aggregated groups of people, but these approaches are inadequate at approximating the complexities of a society’s socio-technical systems. In this paper, a new methodology is presented that utilizes information from individuals in a synthetic population to create social impact models for engineered products. Using this method, a product designer can make predictions about a product’s social impacts during the product development process. Social impact predictions are made for individuals in the population, thereby giving product designers an understanding of a product’s impact at the individual level. Once these individual impacts are aggregated, the product designer can also estimate the social impact of a product on sub-populations and communities. While this method can guide product designers in better understanding a product’s social impacts, it is limited by the availability of high-quality individual-level data, the designer’s understanding of the product’s socio-technical system, and the existing complexity of social systems. This methodology is illustrated by predicting the social impact of a new cassava peeling machine being developed with a farming cooperative in the Brazilian Amazon.