Machine Learning

ToolsGroup is the first SCP software vendor to integrate Machine Learning into supply chain planning, way back in 2010. Reduce demand volatility and risk with our self-tuning machine learning engine—built into ToolsGroup Service Optimizer 99+. Our probability forecasting and machine learning engines crunch multiple demand variables to automatically generate a reliable demand forecast. This self-tuning approach allows you to predict demand behavior much more accurately than considering demand history alone.

ToolsGroup used machine learning to identify and track seasonality patterns and trends for a leading HVAC company. The system recognizes more than 200 “micro-climates” within the United States and their seasonal timing variations. Machine learning sifted through the SKU-Locations to identify “clusters” with similar seasonality profiles. They improved service levels by 16 percent while simultaneously increasing inventory turns by 25 percent.

ToolsGroup used its probabilistic forecast combined with machine learning to help a global cosmetics company apply daily point-of-sale demand sensing data to improve their forecast accuracy. Integrating their sell-out demand patterns with their sell-in forecast quickly improved their WMAPE (weighted mean absolute percent error).

ToolsGroup used machine learning for promotions to help a multinational food company lower forecast error 20 percent and lost sales by 30 percent. They increased their service level to 98.6 percent, and realized a 30 percent reduction in product obsolescence

ToolsGroup used machine learning (new product introduction and launch profiles) to help a global leader in eye-wear cluster the behaviors of past launches, select the most probable performance for the new product, then “learn” common demand behaviors in the first launch period through detailed demand profiles. They improved global WMAPE by 10 percent and reduced the forecast baseline on new launches by about 30 percent.

Create groups of products and groups of markets to model periodic, repetitive and

generally regular and predictable patterns at the daily level. The groups are then used for the calculation of the daily sales profiles

How to use weather, social media, IoT, market trends, indicators and other

external data to improve the forecast.

Creates the Life Cycle groups. A Life Cycle Profile is calculated for each Life Cycle

Group.