A leading global FMCG company faced inefficiencies in concentrate flow planning across LATAM due to discrepancies between Frozen Forecasts and actual bottler orders. These variations caused inventory imbalances, forcing bottlers to hold excess stock at higher costs or rely on expensive air-freight shipments. The demand planning process depended on traditional forecasting with frequent manual revisions, and siloed operations made real-time adjustments difficult.
Sigmoid developed an ML-based forecasting model built on Microsoft Azure to generate data-backed purchase order recommendations, replacing static frozen forecasts with a dynamic real-time planning approach. A React-based web portal was delivered for forecasting and order recommendations, leveraging AI-driven techniques to enhance demand planning accuracy and minimize inventory imbalances.
The solution achieved a 30% improvement in order planning efficiency and a 25% improvement in forecast accuracy, reducing Purchase Order gaps. 1.5x more products became aligned with Days on Hand (DOH) policies, and air-freight costs dropped 15% within six months of deployment.
Have a similar implementation?
Share your customer's AI results and link it to your vendor profile.
Submit a case study →
Your brand here — Reach manufacturing decision-makers researching AI adoption.