Dynamic File Generation for New Product Demand Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Demand planning for new products is challenging due to the lack of historical data, as existing methods rely on past values that do not exist for new products, making it difficult to forecast future demand accurately.
Innovation Solution
The system allows users to define references to existing products with similar characteristics, dynamically generating phase-in and phase-out curves by analyzing historical data from these products, creating a CSV file that can be edited and used for forecasting new product demand.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If statistical algorithms use past values for demand forecasting, then forecasting accuracy is improved, but new products cannot be forecasted because they have no historical data
Solution Approach 1:
The patent applies the copying principle by creating synthetic historical data for new products through copying and adapting patterns from similar existing products. The system identifies comparable products with established sales histories and replicates their phase-in curve patterns to generate forecasted demand data for the new product, enabling statistical algorithms to operate on new products as if they had historical data.
Solution Approach 2:
The patent implements preliminary action by pre-calculating and storing phase-in curves for existing products before they are needed for forecasting. When a new product is introduced, the system has already prepared reference data from similar products, allowing immediate generation of forecasted historical data without waiting for actual sales data to accumulate.
2Measurement precision
If phase-in and phase-out curves are manually defined for each product, then customization and accuracy are improved, but time consumption and complexity increase significantly
Solution Approach 1:
The patent applies self-service by enabling the system to automatically generate phase-in and phase-out curves for new products without requiring manual user input. The system autonomously identifies similar existing products, extracts their curve characteristics, and generates forecasted curves for the new product, eliminating the need for users to manually define each curve while maintaining accuracy through data-driven approaches.
Solution Approach 2:
The patent implements universality by creating a reusable framework where phase-in and phase-out curves from existing products serve multiple purposes. The same curve generation mechanism works across different product types and categories, and the prepared curve data can be reused for various forecasting scenarios, reducing redundant work and accelerating the forecasting process for multiple products.
3Reliability
If historical data from multiple existing products is collected and processed, then forecasting reliability is improved, but data processing complexity and computational resources increase
Solution Approach 1:
The patent applies local quality by selecting and processing data from only the most relevant similar products rather than uniformly processing all available historical data. The system evaluates product similarity metrics and focuses computational resources on the top matching products, ensuring high-quality forecasting input while avoiding the complexity of processing irrelevant data from dissimilar products.
Solution Approach 2:
The patent implements partial action by processing a selective subset of historical data from existing products rather than attempting to analyze all available data. The system identifies and uses only the necessary phase-in and phase-out curve data from the most comparable products, achieving sufficient forecasting reliability without the computational burden of exhaustive data processing.
Data Source
AI summary
Systems and methods are provided for receiving a request for forecasting data to use for new product introduction, the request comprising an indication of a plurality of existing products and accessing data for a plurality of existing curves corresponding to the plurality of existing products. The systems and methods further provide for, based on determining that at least a subset of curves of the plurality of curves comprises a plurality of values, analyzing each value of the plurality of values for each curve of the subset of curves to determine a maximum number of values among all of the curves of the subset of curves, generating a text file comprising a maximum number of value columns corresponding to the maximum number of values, and populating the text file with the data for the plurality of existing curves, including the plurality of values for each curve of the subset of curves.