Demand Forecasting

An accurate demand forecast is the basis for a correct purchasing decision ensuring the perfect stock balance. EazyStock’s pattern-driven demand forecasting significantly improves planning accuracy across thousands of inventory items, eliminating the costly inaccuracies that result in both over and under-estimations of customer demand.


Demand Classification

Inventory managers now have the ability to dynamically track monthly or weekly demand patterns across all their inventory items. EazyStock automatically detects and classifies products into different demand patterns. It dynamically tracks the product’s demand pattern including trends as they changes over the life cycle, ensuring a high precision forecast and that purchasing and reordering accuracy is optimized.

Demand Forecasting

EazyStock applies a statistical forecasting algorithm based on the respective product’s demand classification. A library of advanced statistical methodologies is utilized in order to ensure that the most effective calculation is adopted that matches your product lifecycle—from birth and growth to maturity and decline. The demand classification, and subsequent automatic selection of the optimal forecasting algorithm, is revised continuously and updated automatically by EazyStock improving forecast accuracy.

In cases when demand history cannot predict accurate future forecasts (like new product introductions) EazyStock allows users to set or adjust forecasts manually to ensure overall accuracy.


White Paper - 4 Ways to Improve Demand Forecasting Accuracy

Part Supersessions

When a new product is introduced as a replacement for an existing product, the demand history for the existing product is very relevant for creating a forecast for the new product. Using the item supersession feature, demand and forecast history are mapped for one or more old products into the product.

With EazyStock you have support any type of supersession chain; One-to-one substitution (1:1), One-to-many substitution (1:n), Many-to-one substitution (n:1), and Many-to-many substitution (n:n).


  • Improve demand forecast accuracy
  • Reduce safety stock requirements
  • Eliminate obsolescent inventory risks
  • Minimize stock-outs and back orders
  • Reduce holding and carrying costs of inventory
  • Automate the process of accurately predicting market demand


  • Dynamic demand type classification across the product life cycle
  • Calculates accurate statistical forecast that also can incorporate market intelligence
  • Automatic outlier detection and automatic cleansing according to business rules
  • Automatically assign the parameter settings based on the product’s demand behavior
  • N:N supersession for demand history inherent
  • Use advanced exception management to automatically detect and self-correct problems

Questions? We have answers!Man with Computer

If you have questions about product features, implementation, optimization, or anything else, please connect with us. We will help you immediately.