A blockchain-based dynamic energy pricing model for supply chain resiliency using machine learning

• Published in: Supply Chain Analytics Vol. 6; p. 100066
• Main Authors: Abadi, Moein Qaisari Hasan, Sadeghi, Russell, Hajian, Ava, Shahvari, Omid, Ghasemi, Amirehsan
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.06.2024; Elsevier
• Subjects: Blockchain; Energy consumption management; Machine learning; Power outage; Supply chain resiliency; Supply chain resiliency; Power outage; Blockchain; Machine learning; Energy consumption management
• ISSN: 2949-8635; 2949-8635
• DOI: 10.1016/j.sca.2024.100066

The escalation of energy prices and the pressing environmental concerns associated with excessive energy consumption have compelled consumers to adopt a more optimal approach towards energy usage and an advanced infrastructure such as smart grids. Blockchain technology significantly improves energy management by creating supply chain resiliency in a distributed smart grid. This study proposes a blockchain-based decision-making framework with a dynamic energy pricing model to manage energy distributions, particularly during an energy crisis. Empirical data from U.S. consumers are employed to show the applicability of the proposed model. We include price elasticity to address changes in energy market prices. Findings revealed that the proposed framework reduces total energy costs and performs better when a disruption has occurred. This study provides a post hoc analysis in which four machine learning algorithms are used to predict energy consumption. Results suggest that the autoregressive integrated moving average (ARIMA) algorithm has the highest accuracy compared to other algorithms. •Show blockchain technology can significantly improve energy management during disasters.•Present a blockchain-based decision-making framework.•Propose a dynamic energy pricing model based on blockchain technology.•Demonstrate findings based on empirical North and South Carolina consumer data.•Provide a post hoc analysis based on machine learning analysis.