Predicción de la generación de residuos sólidos domiciliarios con machine learning en una zona rural de Puno

Abstract

La gestión de residuos sólidos es uno de los principales desafíos ambientales en todas las ciudades del mundo debido a factores como el crecimiento poblacional y los hábitos de consumo. Una de las principales herramientas para el diseño de proyectos de gestión de residuos, es la estimación de la generación per cápita, sin embargo, el método tradicional para obtener esta información demanda mucho esfuerzo y tiempo, por ello esta investigación plantea un enfoque alternativo de la estimación de la generación per cápita a partir de factores socioeconómicos. Para ello se recogió información socio económica demográfica e información sobre la generación per cápita de residuos sólidos de 50 familias, posteriormente se determinaron las variables que tienen influencia significativa a partir del coeficiente de correlación ρ de Spearman para las variables numéricas y un ANOVA para las variables categóricas con un umbral de aceptación de 0.4 y 0.05 respectivamente. Las variables seleccionadas se utilizaron para entrenar los modelos de redes neuronales, regresión lineal múltiple, support vector machine, procesos gaussianos y random forest, cuyos desempeños fueron de R2 = 0.986, 0.982, 0.959, 0.837, 0.832; respectivamente. Para la validación se utilizó validación cruzada y partición de datos. Los resultados indican que las variables influyentes son el ingreso per cápita, el gasto en insumos y productos, el tamaño familiar y los servicios del hogar. Se concluye que las predicciones de los modelos son confiables (RMSE desde 8g a 27g) y a partir de ellas se pueden diseñar proyectos.

An Artificial Intelligence Approach for Estimating the Turbidity of Artisanal Wine and Dosage of Clarifying Agents

“…In contrast, between the models of Figure 3, the one with the lowest performance was the fine Gaussian support vector machine (SVM) with an R 2 of 0.770, an MSE of 0.026, and higher values of NRSME and AIC. In the exponential GPR model, the performance values obtained resemble those found by Cerna et al [49], who, in their research, trained Gaussian process models to predict solid residues, finding values of 0.982 and 0.007 for R 2 and MSE, respectively. Those results were obtained because they established non-linear relationships between the data used in the training (dose and type of clarifier) using a probability distribution in a In the exponential GPR model, the performance values obtained resemble those found by Cerna et al [49], who, in their research, trained Gaussian process models to predict solid residues, finding values of 0.982 and 0.007 for R 2 and MSE, respectively.…”

“…In the exponential GPR model, the performance values obtained resemble those found by Cerna et al [49], who, in their research, trained Gaussian process models to predict solid residues, finding values of 0.982 and 0.007 for R 2 and MSE, respectively. Those results were obtained because they established non-linear relationships between the data used in the training (dose and type of clarifier) using a probability distribution in a In the exponential GPR model, the performance values obtained resemble those found by Cerna et al [49], who, in their research, trained Gaussian process models to predict solid residues, finding values of 0.982 and 0.007 for R 2 and MSE, respectively. Those results were obtained because they established non-linear relationships between the data used in the training (dose and type of clarifier) using a probability distribution in a function space.…”

Waste Management and the Economic Value of Recycling in Households in the Santa Mónica - Chota Sector, 2023

“…The generation of SR is the dimension that comes from human activities, from the acquisition, preparation, consumption of food, productive sequences of inputs and finished goods. In his paper [21], point out that the SWM is initiated by the peSWon, who must select and place it according to the type of biodigester from the home or common good spaces. It is the act of reducing the generation of SR in the home as a result of the acquisition of food products, work or family activities, whose waste is placed in a container.…”