- GIM research introduces AI-based forecasting model to boost retail profits while cutting carbon emissions
The Goa Institute of Management (GIM) research team has developed an AI-based forecasting framework that enables retailers to improve profitability and reduce carbon emissions. Addressing one of the most pressing challenges facing modern retail organisations, the study demonstrates how advanced AI can support sustainable retail management without compromising commercial performance.
Published by the prestigious Springer Nature in the journal Circular Economy and Sustainability, the findings of this research introduce an explainable and sustainability-aligned forecasting system that goes beyond traditional demand prediction models. The research team includes Prof. Sumit Tripathi, Research Lead from Goa Institute of Management, along with Ms. Roma Trigunait from Babasaheb Bhimrao Ambedkar University, and Mr. Dinesh Chandra Pandey from Graphic Era Deemed to be University.
In the present market landscape, retailers operate under pressure to meet environmental sustainability goals. Most existing forecasting systems focus on short-term profit, often ignoring carbon intensity, managerial risk preferences, and the need for transparency. Such gaps often lead to excess inventory, waste, and higher emissions.
To bridge these gaps, the GIM research team has combined behavioural decision theory and carbon intensity weighting into an AI-based framework. A unique feature of the developed framework is that sustainability is embedded directly into the model through carbon-weighted cost functions, ensuring that it provides accurate inputs.
Based on the tests conducted, the research team found that the developed framework, enables practical managerial decision-making by maintaining narrow uncertainty ranges, provides accurate demand forecasts with reliable uncertainty estimates and distinguishes between customer segments such as Eco Advocates and Sceptical Buyers, supporting targeted customer retention forecasting
Additionally, through the integration of the Shapley-value attribution AI tool, the system allows teams from marketing, operations, and sustainability to work effectively and ensures robustness under changing market conditions and policy environments.
Speaking about the developed framework, Sumit Tripathi, lead author of the research and Associate Professor at GIM, said, “Sustainability cannot be achieved through intuition or slogans alone; it requires scientifically rigorous, transparent, and trustworthy decision systems. Our research demonstrates that businesses do not have to choose between profitability and environmental responsibility. With the right forecasting framework, organizations can achieve both simultaneously, making decisions that are economically sound, environmentally conscious, and ethically accountable.”
The developed framework can be directly utilised by retail organisations for inventory and demand planning, promotion and pricing strategy development, sustainable logistics and supply chain decisions, and environmental impact assessment.
Beyond retail, the model can be adapted to other sustainability-sensitive sectors such as manufacturing, healthcare, energy systems, and supply chains, where balancing economic performance with environmental responsibility is critical.
