6 real-life Business Process Automation examples for Sustainability

6 real-life Business Process Automation examples for Sustainability

Automation in the sustainability function can be effective because the work depends on fragmented data (operational, supplier, and financial) that has to be translated into credible metrics, disclosures, and action plans.

The best automations reduce the manual burden of data collection, emissions calculation, evidence gathering, and reporting while preserving methodological consistency, auditability, and enough transparency for regulators, investors, and customers to trust the numbers.

Industry

Sector

Nestlé and IBM's Packaging Efficiency

This strategic collaboration between Nestlé and IBM Research utilises advanced generative AI and chemical language models to rapidly discover sustainable, high-barrier packaging materials, effectively compressing years of traditional laboratory R&D into digital simulations.

Use Case
Identifying Novel High-Barrier Sustainable Packaging Materials
Tools
IBM Watson
Input
Public and proprietary documents on packaging materials, molecular structure data, and physical-chemical property datasets
Process
IBM AI learns molecular structures from a vast knowledge base while a regression transformer correlates structural features with properties to propose entirely new materials that resist moisture, temperature, and oxygen
Output
In-silico generation of novel packaging candidates evaluated for cost, recyclability, and functionality
Outcome
Drastic compression of R&D timelines, supporting the goal of 100% recyclable or reusable packaging by 2025
Food Processing and Packing
Sustainability
Product

Mondelez's Sustainability Monitoring System

Mondelez International utilises an AI-powered monitoring system within its Harmony Academy Digital Platform to drive sustainable wheat farming. By analysing real-time environmental data from supplier farms, the automation provides actionable insights that optimise agricultural inputs and ensure high-level ESG compliance across the global supply chain.

Use Case
Sustainable Agriculture Monitoring
Tools
Harmony Digital Platform
Input
Farm-level data including greenhouse gas emissions, nitrogen use, pesticide application, soil health metrics, and biodiversity indicators
Process
AI tools analyze farm-level environmental data in real time to generate data-driven recommendations for input optimization and continuous ESG KPI monitoring
Output
Real-time visibility into sustainability performance and improved traceability of raw material origins
Outcome
Advances the commitment to source 100% of wheat through the Harmony programme and achieves measurable improvements in regenerative agriculture practices
Food and Beverage Manufacturing
Sustainability

H&M's AI Demand Forecasting to Reduce Overproduction Waste

H&M deploys AI-driven demand forecasting across its global retail operations to predict customer demand with greater precision, reducing overstock and cutting the excess production that contributes to unsold apparel waste. The system combines machine learning with real-time sales, trend and regional data to balance supply and demand at scale.

Use Case
AI-Powered Demand Forecasting for Waste Reduction
Tools
Internal Tools
Input
Historical sales data, real-time inventory levels, consumer trend signals, regional buying behaviour patterns, seasonal data and external variables such as weather and economic indicators.
Process
Machine learning models process vast datasets to generate granular demand forecasts by product, region, and season, feeding directly into production planning workflows to align supply with actual consumer demand.
Output
Product-level demand forecasts with 85–90% accuracy, used to set production volumes, optimise replenishment orders and reduce markdown requirements across H&M's global store and ecommerce estate.
Outcome
Overstock falls by 25%, sell-through rates improve by 15–20% and forecasting accuracy reaches 85–90%, materially reducing the volume of unsold apparel destined for waste and aligning operational performance with H&M's sustainability targets.
Non Food Retail
Sustainability

Reckitt's AI-Powered Product Carbon Footprint Automation

Reckitt partners with CO2 AI and Quantis to automate the calculation of product-level carbon footprints across its entire 25,000-product portfolio. Within four months, the platform improves the accuracy of Reckitt's emissions footprint by 75 times, enabling targeted Scope 3 reduction across its supply chain in pursuit of a 50% emissions cut.

Use Case
Automated Product-Level Carbon Footprinting
Tools
CO2 AI & Quantis
Input
Over 300,000 operational data points collected from Reckitt's brands, including raw material usage, packaging data, and supplier activity records across its health, hygiene and nutrition product portfolio.
Process
CO2 AI's generative AI engine automatically matches each activity line to the most relevant emission factor from a library of over 110,000 factors, using retrieval-augmented generation and vectorial search to replace manual categorisation.
Output
Precise, product-level carbon footprints for all 25,000 products, delivered in minutes rather than months, with full traceability and emissions hotspot visualisations for use by procurement, R&D and sustainability teams.
Outcome
Reckitt achieves complete product emissions coverage in under four months, improving footprint accuracy by 75 times compared to the previous approach using 333 representative products and identifies new pathways to meet its Scope 3 reduction target.
Direct to Consumer Brands
Sustainability

General Mills' Automated Scope 3 Emissions Tracking with CO2 AI

General Mills partners with CO2 AI to integrate carbon data into its core enterprise systems and workflows, automating Scope 3 emissions reporting across its supply chain and enabling supplier-specific emissions data collection at scale to support its net-zero-by-2050 target.

Use Case
Automated Scope 3 Emissions Reporting and Supply Chain Carbon Tracking
Tools
CO2 AI
Input
Supplier activity data, procurement records, logistics data, and product lifecycle information
Process
CO2 AI's platform maps activity data to emissions sources
Output
Automated emissions reports across Scopes 1, 2, and 3, supplier-level carbon footprints, and data to support decarbonisation planning and regulatory disclosure
Outcome
Eliminates manual emissions reporting, improves data accuracy, and provides a scalable foundation for supplier engagement and Scope 3 tracking.
Food Processing and Packing
Sustainability

Capgemini's AI-Driven Campus Energy Optimisation

Capgemini deploys Schneider Electric's EcoStruxure Command Centre (ECC) across 23 campuses in India to autonomously manage energy consumption. The AI platform reduces energy usage by 25 GWh and saves €3 million while enabling a transition to 100% renewable electricity across the estate.

Use Case
Autonomous Building Energy Management
Tools
Schneider Electric EcoStruxure
Input
Real-time data from building systems across 23 campuses, including occupancy sensors, HVAC controls, lighting systems, weather feeds and electricity consumption data.
Process
Schneider Electric's EcoStruxure Command Centre uses AI to continuously analyse occupancy patterns, weather forecasts, and energy loads, autonomously adjusting heating, cooling and lighting to minimise consumption and carbon output.
Output
Automated energy control commands issued to building systems in real time, with centralised dashboards providing visibility of consumption, carbon metrics and renewable energy utilisation across all campuses.
Outcome
Capgemini reduces energy consumption by 25 GWh and saves €3 million across its 23 Indian campuses, while successfully transitioning the entire estate to 100% renewable electricity.
IT Services and System Integration
Sustainability
See all 121 Business Process Automation examples