Parallel Processing Historical Data for Real-Time ERP Simulations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
ERP systems face challenges in performing large-scale 'what-if' scenarios over millions of sales order records in real-time, limiting their ability to provide accurate price and cost analysis and forecasting, especially under changing market conditions.
Innovation Solution
An interactive ERP system with a simulator that allows users to generate on-the-fly revenue and profitability reports, enabling 'what-if' modifications and real-time simulations for cost, price, and sales volume changes, displayed in a graphical format for quick decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ERP systems process large-scale what-if scenarios over millions of sales order records, then processing accuracy is maintained, but processing time becomes unacceptably long and real-time analysis is impossible
Solution Approach 1:
The patent segments the sales order data into different time periods (historical data vs. forecast periods) and processes them separately using different methodologies. Historical data is processed using actual values while forecast periods use simulated what-if scenarios. This segmentation allows efficient processing of large datasets while maintaining accuracy where needed.
Solution Approach 2:
The system performs preliminary calculations and data preparation in advance by pre-processing historical sales data and storing it in optimized formats. This preliminary action enables the what-if simulation to run quickly on forecast periods without re-processing entire historical datasets, significantly reducing processing time while maintaining accuracy.
2Use of energy by moving object
If batch processing is used for revenue and profitability reports, then system resource consumption is reduced, but real-time insights and quick decision-making are compromised
Solution Approach 1:
The system dynamically adjusts its processing mode based on user needs. It can switch between batch processing for routine historical reports (conserving resources) and on-demand what-if simulation processing for forecast analysis (providing real-time insights). This dynamic approach optimizes resource consumption while enabling real-time analysis when required.
Solution Approach 2:
The what-if simulation engine is designed to be self-service capable, automatically managing its own computation resources and data processing without requiring external batch job scheduling. It can independently execute simulations on demand using available system resources, providing real-time results without the overhead of traditional batch processing infrastructure.
3Measurement precision
If custom reports are generated for specific profitability analysis, then analysis precision is improved, but system complexity and implementation difficulty increase
Solution Approach 1:
The patent implements a universal what-if simulation engine that can handle multiple types of profitability analyses through a single integrated interface. The same engine processes revenue analysis, profitability analysis, cost analysis, and scenario comparison, eliminating the need for multiple custom report systems while maintaining high analysis precision across all functions.
Solution Approach 2:
The system achieves different analysis precisions by changing parameters within a unified framework rather than creating separate complex systems. Users can adjust simulation parameters such as time periods, product categories, geographic regions, and scenario variables to get precise analysis for specific needs, all through the same underlying engine that maintains consistency and reduces overall system complexity.
Data Source
AI summary
Disclosed are methods and apparatuses for generating simulations on the fly using large historical data and forecast data. To improve system performance, the large historical data are processed in parallel.