An explainable artificial intelligence approach to understanding drivers of economic energy consumption and sustainability

Praveen Ranjan Srivastava, Sachin Kumar Mangla, Prajwal Eachempati and Aviral Tiwari

Energy Economics, 2023, vol. 125, issue C

Abstract: This study aims to optimize energy consumption to achieve sustainability, as there is a necessary foray into the green energy market by switching to net-zero carbon-emitting fuel alternatives. There is a need to decide when to switch to these net-zero fuels from conventional energy sources, which calls for a deeper investigation into the driving factors that lead to harmful energy emissions. This investigation will help monitor and curb such indicators to minimize harmful emissions or determine the opportune time to switch to green energy alternatives if the indicator levels cannot be controlled. This research is motivated by recent studies that consider factors such as per capita carbon intensity and other sector-specific factors such as per dwelling carbon intensity and per value-added carbon intensity. However, these studies do not scientifically quantify the extent to which each element contributes to the final conventional energy consumption. Furthermore, sector-specific consumption indicators also need to be estimated to undertake significant design modifications for developing energy-efficient systems. The consumption is predicted as “total consumption” for the four sectors considered: residential, industry, services, and transportation. Advanced machine learning algorithms such as random forest, gradient boosting, and deep neural network are used for this purpose. The essential factors/drivers for each sector are derived through the explainable artificial intelligence Shapley framework, which scientifically measures the contribution of each factor to energy consumption. Consequently, the major key indicators for each of the four sectors under consideration are identified. Alternatively, the insights gained from the study may prompt a complete switch to alternative energy-efficient sources. The study findings provide valuable insights for both communities and businesses in achieving the goals of net-zero and energy conservation. Additionally, some examples are highlighted to demonstrate how practitioners can implement customized designs under each sector to reduce harmful emissions and promote energy efficiency. Suitable steps for stakeholder engagement and community participation in this regard are also suggested.

Keywords: Random forest; Gradient boosting; Deep neural network; XAI; Shapley; Energy consumption; Non-conventional energy; Net-zero emissions; Sustainability; Energy economics (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations: View citations in EconPapers (7)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0140988323003663
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:eneeco:v:125:y:2023:i:c:s0140988323003663

DOI: 10.1016/j.eneco.2023.106868

Access Statistics for this article

Energy Economics is currently edited by R. S. J. Tol, Beng Ang, Lance Bachmeier, Perry Sadorsky, Ugur Soytas and J. P. Weyant

More articles in Energy Economics from Elsevier
Bibliographic data for series maintained by Catherine Liu ().