An Intelligent Predictive Model for Carbon Dioxide (Co2) Emission in Kobape/Sagamu Industrial Estate Zone of Ogun State

Abstract

This paper focuses on developing an intelligent predictive model for forecasting carbon dioxide (CO2) emissions in the Kobape-Sagamu Industrial Estate Zone, Ogun State, considering environmental sustainability. The approach involves utilizing machine learning algorithms and data analytics to analyze historical emission data, incorporating factors like industrial activities, energy consumption, and meteorological conditions. The model employs a sophisticated algorithm to enable stakeholders to anticipate and mitigate potential environmental impacts. The significance lies in empowering industries, regulatory bodies, and communities with a proactive environmental management approach. The K-Nearest Neighbor, ARIMA, SARIMAX, Holt-Winters, and LSTM models were evaluated, with varying predictive accuracies. The paper contributes to environmental science by offering a practical solution for addressing challenges related to industrial emissions. The findings aim to inform policy-making, promote eco-friendly industrial practices, and serve as a benchmark for similar studies in other industrial zones.

Keywords: Machine learning, Carbon dioxide, Deep learning, ARIMA, SARIMAX, Holt-winters