Multi-Variable Demand Model via Data Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data aggregation methods in enterprise databases are largely manual and inefficient, particularly when combining datasets with different statistical distributions, leading to errors and reduced reliability in multi-variable analysis, especially in Demand Chain Management applications where data scarcity and hierarchical data structures pose challenges.
Innovation Solution
A method involving data normalization and multiplicative regression analysis is employed to combine multiple datasets from relational databases, using a base value such as the average of regular weeks to normalize variables, allowing for the creation of a single aggregated dataset that can be analyzed using a multiplicative demand model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual data aggregation methods are used to combine datasets from different tables, then data can be aggregated, but the process is inefficient and error-prone
Solution Approach 1:
The system automatically performs data aggregation by having the database management system execute SQL queries that self-join multiple tables based on predefined relationships. The normalization process and aggregation operations are performed autonomously without manual intervention, eliminating manual errors while maintaining high efficiency through automated query execution.
Solution Approach 2:
The patent replaces manual mechanical data aggregation processes with automated computational systems. Instead of manual table joining and data combination, the system uses database management systems to automatically execute complex SQL queries that perform normalization, self-joining, and aggregation operations, significantly improving both efficiency and reliability.
2Reliability
If data is aggregated from multiple hierarchical levels to improve reliability, then more data points are available, but conventional manual approaches are inefficient
Solution Approach 1:
The system performs preliminary data normalization and table relationship definition before aggregation is needed. By pre-establishing the normalization rules and self-joining relationships between tables, the system prepares the data structure in advance, enabling rapid aggregation when statistical applications require multi-level data combination, thus improving both reliability and productivity.
3Manufacturing precision
If fields from different tables are associated and mapped for proper aggregation, then accurate aggregation is achieved, but the process becomes complex and manual
Solution Approach 1:
The patent creates a universal data aggregation framework that handles multiple table associations and field mappings through a single standardized SQL query structure. The self-joining mechanism provides a multi-functional approach that can aggregate data across different hierarchical levels and table relationships using the same normalization process, reducing process complexity while maintaining high aggregation precision.
Data Source
AI summary
Techniques for multi-variable analysis at an aggregate level are provided. Two or more datasets having different statistical data distributions and which are not capable of being aggregated are acquired. The values for variables in the two or more datasets are normalized to produce a single integrated dataset of normalized values. The normalized values are then used to produce a demand model that represents and integrates multiple disparate products or services from the two or more datasets into a single demand model.