Artificial Intelligence (AI) applied to waste management: A contingency measure to fill out the lack of information resulting from restrictions on field sampling

Abstract

Waste characterization is essential in planning to reduce waste, create recycling programs and properly use public funds. Traditionally, waste characterization is successfully performed by on-site sampling. This monitoring requires specialized professionals following safety protocols to decrease the health risks associated with thewastehandling. However, on-site sampling is susceptible to restrictions in mandatory quarantine situations. Here we show the possibility of using Artificial Intelligence applied to waste management. We identified an Artificial Neural Network (ANN) model fed by combination of population, Gross Domestic Product, Potable Water Supply and Sanitation System data that could be used to fill gaps related to the pandemic period. The modified model predictions were successful due to the adaptive capacity of the ANN-based models. Our results demonstrate that ANN can be used in contingency plans to predict the gravimetric composition and specific weight of Municipal Solid Waste, based on strategically chosen socioeconomic information. We anticipate that our model will be a starting point for more sophisticated computational models. It will allow not only the filling of gaps but also the use for auditing, because it allows analyzing the consistency of gravimetric composition and specific weight data provided by third parties.