Waste Generation Rate Prediction by Time-series Data Analysis

To effectively plan and manage waste disposal resources, improve collection schedules, and reduce environmental consequences, governments need to have accurate trash prediction rates. Therefore, waste generation prediction is an important concern for a sustainable city and a healthy life. Existing waste generation rate prediction faces challenges like limited dataset, reliance on synthetic data, and negligence of seasonality and holiday effects. Time series analysis for predicting waste generation rate can be effective for this purpose as it considers temporal patterns, including trends and seasonality. In this paper, a comparative assessment of three time series models is conducted on a real dataset aiming to provide a forecast of municipal solid waste generation rate. The models are Long Short Term Memory (LSTM), Prophet and Seasonal Autoregressive Integrated Moving Average with Exogenous Factors (SARIMAX). Simulation results show that the Prophet model exhibits promising results in terms of various error metrics like Mean Absolute Error(MAE), Mean Absolute Percentage Error(MAPE), Mean Squared Error(MSE), Root Mean Squared Error(RMSE) and coefficient of determination (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) in comparison to LSTM or SARIMAX. This work can give useful insight for decision-makers to get the production rate of solid waste for efficient garbage management.