GIM develops AI forecasting model to boost retail profits while cutting emissions

The Goa Institute of Management (GIM) research has developed an artificial intelligence–driven (AI-driven) forecasting framework that enables retail organisations to enhance profitability while lowering carbon emissions, addressing the growing challenge of balancing commercial performance with sustainability objectives.

The findings, published by Springer Nature in the journal Circular Economy and Sustainability, introduce an explainable forecasting system that integrates sustainability considerations into decision-making, moving beyond traditional demand prediction approaches.

The study was led by Sumit Tripathi, associate professor and research lead at GIM, along with Roma Trigunait of Babasaheb Bhimrao Ambedkar University and Dinesh Chandra Pandey from Graphic Era Deemed to be University.

In today’s retail landscape, organisations are under increasing pressure to meet environmental targets without compromising financial outcomes. The press note pointed out that many existing forecasting models prioritise short-term profits and often ignore factors such as carbon intensity, risk preferences of managers, and transparency in decision-making. Such limitations can lead to overstocking, higher waste levels, and increased emissions.

Also read IIM Indore to begin IPMAT 2026 registration from February 2; exam on May 4

GIM AI model blends profit with sustainability

The AI-based framework has been designed to overcome these limitations and can be readily used for inventory and demand planning, pricing and promotion decisions, logistics and supply chain management, and environmental impact evaluation.

The model centred on retail applications can also be extended to other sectors sensitive to sustainability concerns, such as manufacturing, healthcare, energy systems, and supply chains, where balancing financial performance with environmental responsibility is increasingly critical.

To achieve this, the GIM research team combined behavioural decision theory with carbon intensity weighting within an AI-driven forecasting model . One of its defining features is the direct integration of sustainability into the forecasting process through carbon-weighted cost functions, enabling organisations to incorporate environmental considerations into routine operational decisions rather than treating them as secondary concerns.

Results from testing the framework indicated that it supports informed managerial decision-making by maintaining narrow ranges of uncertainty, offers accurate demand predictions with dependable uncertainty estimates, and distinguishes between customer segments such as Eco Advocates and Sceptical Buyers. This capability allows retailers to develop more targeted and effective demand forecasting and customer retention strategies.

Also read IIM Mumbai launches executive AI programmes; applications open at iimmumbai.ac.in

The framework also incorporates the Shapley-value attribution tool, enabling marketing, operations, and sustainability teams to work together using transparent and explainable AI insights. This improves robustness under changing market conditions and evolving policy environments.

Commenting on the research, Prof. Tripathi said sustainability requires “scientifically rigorous, transparent, and trustworthy decision systems,” adding that businesses do not need to choose between profitability and environmental responsibility.