Modeling Scope 3 Green House Gas Emissions

With the growing urgency to combat climate change, there is an increasing need for accurate assessment and reduction of greenhouse gas emissions. Among these, Scope 3 emissions, which measure indirect emissions along a company’s entire value chain pose a significant challenge due to their complexity and diversity. In this research, machine learning algorithms are built to develop predictive models for estimating Scope 3 greenhouse gas emissions. Models are trained on US organizations’ self-reported Scope 3 emissions obtained through the Carbon Disclosure Project. By harnessing the power of machine learning the models achieve better accuracy in predicting Scope 3 emissions within various sectors compared to a naïve model. This work will support organizations to make informed decisions for emission reduction strategies.