Demand Forecasting Seasonality Curve Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current demand forecasting systems for retail items face challenges in producing accurate week-by-week forecasts due to insufficient and unreliable historical data, leading to over/understocking and incorrect inventory management, especially when dealing with multiple demand variables like promotions, seasonality, and weather effects.
Innovation Solution
The system automatically determines and verifies the reliability of seasonality and promotion effects by analyzing historical sales data, discarding unreliable parameters and replacing them with optimized ones, using a demand model that focuses on seasonality and promotion impacts to generate accurate demand forecasts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If historical sales data is used to determine demand parameters, then the demand forecast can be generated, but the quality of the forecast is poor due to insufficient and unreliable historical data
Solution Approach 1:
The patent combines multiple seasonality curves for the same item across different years to create a more reliable aggregated seasonality curve. By merging data from multiple time periods, the system overcomes the insufficiency of individual year data and produces more accurate demand forecasts even when historical data is limited
Solution Approach 2:
The system performs preliminary verification of seasonality curve reliability before using them for demand forecasting. It calculates repeatability metrics and smoothness parameters in advance to identify and retain only reliable curves, ensuring that the forecast is based on validated historical patterns rather than unreliable data
2Adaptability or versatility
If multiple demand variables (promotions, seasonality, weather) are considered in the demand model, then the forecast comprehensiveness is improved, but the complexity of determining demand parameters increases
Solution Approach 1:
The patent segments the complex demand modeling process into distinct components: seasonality analysis, promotion effect analysis, and other demand variables. Each component is analyzed separately with specific verification methods, making the overall complex process more manageable and systematic
Solution Approach 2:
The system implements feedback mechanisms to verify the reliability of determined demand parameters. It calculates repeatability metrics and smoothness parameters, then uses this feedback to retain or discard specific seasonality curves and demand parameters, ensuring that only reliable parameters are used in the final forecast
3Measurement precision
If regression is performed on the demand model to separate effects of several demand variables, then the accuracy of individual parameter estimation is improved, but the computational time and resources increase
Solution Approach 1:
The patent applies partial verification to seasonality curves by checking specific criteria (repeatability threshold, smoothness parameter, sparsity parameter) rather than performing exhaustive analysis on all possible curves. This selective approach maintains accuracy by verifying only the most relevant aspects while reducing computational burden
Data Source
AI summary
Embodiments select demand forecast parameters for a demand model for one or more items, receive historical sales data for the items on a per store basis and receive a plurality of seasonality curves for a first item. Embodiments determine a repeatability of each of the seasonality curves using a correlation of each seasonality curve over year to year demand and retain a first seasonality curve based on the repeatability. Embodiments determine a smoothness of the first seasonality curve and determine a sparsity of the first seasonality curve. Based on the determined repeatability, smoothness and sparsity, embodiments determine that the first seasonality curve is reliable and repeat the receiving the plurality of seasonality curves, determining the repeatability, determining the smoothness, and determining the sparsity to determine a plurality of reliable seasonality curves. Embodiments use the demand model and the reliable seasonality curves and determine a demand forecast for the first item.