Development of a correlation to estimate per capita municipal solid waste generation rates in Sao Paulo state, Brazil: population, per capita income and electricity consumption influences

Abstract

Forecasting municipal solid waste (MSW) generation is fundamental in choosing and scaling the operations and processes involved in municipal management. The challenge for the financial sustainability of this management is to create indicators that enable municipal solid waste fees to be charged in proportion to the amount of MSW generated by each resident To this end, we sought to develop equations to estimate the per capita waste generation rate (TGP) in the municipalities of the state of sao Paulo, Brazil. It was assumed that the variables of population (P), daily per capita income (RPC) and daily per capita consumption of electric energy (Ce) in the municipalities were determining factors of the TGP. The dataset involved 238 municipalities that routinely weighed their solid wastes in the baseline year of 2013. Several functions were tested in order to adjust the observed TGP to those calculated by the least squares method. The quality of the adjustments was evaluated by comparing the expected and observed values with the bisector of the first quadrant, as well as by the Pearson (r) and determination (R-2) coefficients, root-mean-square error (RMSE), and mean percentage error (E-p) values. The equation that best represented the dataset had an r of 0.49, R-2 of 024, RMSE of 0217 kg.person(-1)d(-1 )and E-p of -14.1%. Despite the relatively low R-2, it was demonstrated by Student's t-test that the proposed equation was able to represent mean values and result in the same variance with probability higher than 99%. This fact allows to consider the equation adequate for predicting the TGP in the municipalities of the state of Sao Paulo.