AI driven automation for enhancing sustainability efforts in CDP report analysis

Area of Science:

• Supply Chain Management
• Environmental Science
• Artificial Intelligence

Background:

• Increasing pressure on businesses to adopt sustainable practices and reduce carbon footprints.
• Need for efficient operational strategies within supply chains to balance environmental and economic goals.

Purpose of the Study:

• To develop a novel hybrid approach combining Genetic Algorithms (GA) and Long Short-Term Memory (LSTM) networks for optimizing supply chain sustainability.
• To create a cost-effective, scalable solution for reducing emissions, improving efficiency, and ensuring regulatory compliance.

Main Methods:

• Utilized publicly available Carbon Disclosure Project (CDP)-reported data for emissions prediction and resource allocation.
• Employed LSTM networks for forecasting emission trends based on historical data.
• Applied GA for optimizing supply chain processes, including transportation and energy consumption, through multi-objective optimization.

Main Results:

• Achieved a 23.67% reduction in total supply chain emissions, with notable improvements in indirect emissions.
• Enhanced operational efficiency by 10.98%.
• Ensured 100% compliance with environmental regulations, eliminating penalties.

Conclusions:

• The hybrid GA-LSTM framework effectively optimizes supply chain sustainability, offering a practical, data-driven solution.
• The approach provides valuable insights for businesses aiming to meet sustainability targets and improve performance.
• The system is scalable for both large corporations and SMEs, promoting widespread adoption of sustainable practices.