Waste Reduction

Identify the causes of waste across different levers and build predictive alerts across supply chains to reduce finished goods and resource/material waste as stales, damages, returns, disappearances, etc.

How C5i’s Waste Reduction Solution Helps

Data

  • Demand Planning
  • Damages & Stales by site
  • Ending Inventory by site
  • Reconciliation (Order vs. Shipped vs. Delivered)
  • Sales
  • Shipment History
  • Product details
  • Bin-wise data

Analytics

  • Descriptive analytics with Stales, Damages, and Returns/Chargeback data that tells us WHAT & HOW by category
  • View into anomalies and patterns that drive key waste metrics
  • Causal insights based on machine learning to estimate waste

Insights

  • Visibility on volume and value of finished goods waste (FGW) by location, SKU, and destination along with benchmarking between facilities
  • Insights on relationships between processes (sales vs. planning, shipment, inventory count, invoice adjustment) and their drivers
  • Waste predictions based on causal factors

Actions

  • Identify and prioritize specific products intelligently and at speed with no manual data analysis required
  • Focus on the current week and weeks ahead with the potential to take pre-emptive action based on predictions

Outcome
Reduce finished goods waste by 30%-40% across sites with
predictives alerts across the supply chain.

Customers Seeing Success with C5i

“The insight provided is something our team is very excited about. We appreciate the work you’ve been doing in helping to get this up and running as well as your patience with the information we’ve been working to supply. The option you’ve helped to add with the alternative to review the position of the cases, as well as demand and how it compares with the ending inventory for the week, will be a big win for us.”

Site Manager,, Multinational Fortune 50 F&B Company

Case Study

‘Best Data Science Project of the Year’
The Data Science Excellence Awards 2022 by AIM

Finished Goods Waste Reduction with Augmented Analytics platform, Discovery

Client – Multinational Fortune 50 F&B Company

  • Connected supply chain datasets across Warehouse, Planning, Transportation, Sales, and Shipment (scaled to 170+ client locations across North America)
  • Reduced inventory waste through factual, diagnostic, causal, and predictive insights with real-time alerts, anomaly detection, and actionable recommendations for proactive and corrective action

Led to 30% reduction in inventory waste with business impact worth millions of dollars

Deployed on Microsoft Azure

The Finished Goods Waste Reduction (FGWR) Model is integrated into the client’s Microsoft Azure infrastructure. Source data is first integrated with the client’s Azure delta lake using Azure Data Factory. The data is further cleansed and transformed using Azure Databricks and then fed into the FGWR Model. The output from the FGWR Model is used to generate actionable insights that are consumed by business users using an interactive Power BI dashboard.

Connected Intelligence with AI-powered Augmented Analytics Platform
Discovery

Powered by Generative AI

Deliver relevant, actionable, and human-friendly insights across multiple consumption mediums and personas to create an insights-first culture that rewards data-driven decision-making

Automated insights generation from connected enterprise and external data

Descriptive, Diagnostic, Predictive, and Prescriptive Analytics driving actionable insights

Persona-based approach to provide contextual insights on near real-time basis

High adoption with curated natural language insights available on chat, voice, enterprise BI platforms, executive presentations, emails, Teams, Slack, etc.

Tracking of impact of decision-making on key performance indicators (KPIs)

Important Metrics

Speed-to-actionable insights reduced from days to seconds
45% increase in analytics adoption by use of generative and conversational AI
30% time savings with a single source of truth and Natural Language querying
~20% revenue impact with timely, data-driven decision-making

Recognition for Discovery

Next Steps

Get Started