Demand Forecasting for New Products Using Similar Product Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current supply chain management systems rely on human guesses for determining shipment and inventory levels for new products, leading to inefficient use of inventory space due to the lack of demand history for these products.
Innovation Solution
A system and method for forecasting demand data for new products by generating a group phase-in profile based on demand patterns of similar products, which calculates demand for a chosen time period using a process involving filtering, sorting, and categorizing data to create accurate demand profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If human guesses are used to determine shipment and inventory levels for new products, then the system can operate without demand history, but inventory management efficiency deteriorates and inventory space is wasted
Solution Approach 1:
The system creates a synthetic demand history for new products by copying and adapting demand patterns from similar existing products. This allows the system to operate automatically without human guesses while maintaining inventory management efficiency, as the copied patterns provide realistic demand forecasts that optimize inventory levels and reduce waste.
Solution Approach 2:
The system introduces an intermediary mechanism that generates artificial demand history data based on similar product patterns. This intermediary fills the gap between having no demand history and needing accurate forecasts, enabling efficient inventory management without requiring actual historical sales data for new products.
2Measurement precision
If demand forecast data is generated based on past order history, then accurate forecasts can be achieved for existing products, but new products without demand history cannot be forecasted accurately
Solution Approach 1:
The system changes the parameter of demand history generation by creating synthetic historical data parameters based on similar product characteristics rather than using actual past orders. This allows accurate forecast measurements to be applied to new products by transforming their parameters into comparable patterns from existing products, thus achieving both forecast accuracy and adaptability to new products.
3Device complexity
If manual estimation is used for new product demand, then no historical data is required, but inventory space is inefficiently used and shipment quantities are inappropriate
Solution Approach 1:
The system copies demand patterns from similar products to create realistic demand forecasts for new products. This eliminates the need for manual estimation while optimizing inventory space utilization, as the copied patterns provide data-driven quantities that match actual consumer behavior, preventing both over-stock and under-stock situations.
Solution Approach 2:
The system performs preliminary action by pre-calculating demand forecasts using similar product patterns before actual sales occur. This preliminary forecasting based on copied patterns allows appropriate shipment quantities to be determined in advance, optimizing inventory space utilization without requiring actual historical data from the new product itself.
Data Source
AI summary
Embodiments include a system for forecasting demand data for new products. The system and method may include generating a phase-in group profile based on similar products. The phase-in group profile may have multiple demand profiles for low, medium and high demand variations. These demand profiles may be applied to new products in the same phase-in group to generate a forecast profile.