Scalable Tuning Engine Parallel Dataflow Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current systems face challenges in efficiently processing large quantities of complex data for price optimization in retail, leading to slow data processing and infrequent updates, which can result in delayed detection of market changes and suboptimal pricing strategies.

Innovation Solution

A computer-implemented method that decomposes dataflows into distinct executable segments across process and data domains, enabling parallel execution across multiple computers, allowing for scalable and frequent data analysis and optimization of prices based on real-time market conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If traditional sequential data processing methods are used for price optimization, then processing accuracy can be maintained, but processing speed becomes too slow to detect market changes frequently

Engineering Contradiction:

Improvedata processing speedVSAvoidsystem complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent segments the data processing flow into distinct executable segments organized along process domains (e.g., data extraction, transformation, loading) and data domains (e.g., product data, sales data, competitor data). This segmentation enables independent parallel execution of segments across multiple computers, dramatically increasing processing speed while maintaining manageable system complexity through modular architecture.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If complex statistical processes are used for demand forecasting, then forecasting accuracy is improved, but the amount of data processing required becomes too cumbersome

Engineering Contradiction:

Improvedemand forecasting accuracyVSAvoiddata processing efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces traditional sequential mechanical data processing with a parallel processing system that distributes computational tasks across multiple computers. Complex statistical processes for demand forecasting are executed as parallel executable segments, maintaining high forecasting accuracy while dramatically improving data processing efficiency through concurrent computation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If price optimization is performed frequently, then responsiveness to market changes is improved, but processing time and computational resources increase

Engineering Contradiction:

Improvepricing strategy update frequencyVSAvoidprocessing time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent establishes a continuous parallel processing system that can execute price optimization analyses frequently without significant increases in processing time. By maintaining parallel execution paths across multiple computers, the system continuously processes sales data, updates demand forecasts, and generates optimized pricing strategies in near-real-time, enabling frequent updates while minimizing processing delays.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS9785951B1Scalable tuning engine

Publication Date: 2017.10.10 DEMANDTEC LLC

AI summary

A computer implemented method for processing data is provided. At least one dataflow comprising transformational and numerical steps is defined. The flow is decomposed into distinct executable segments along process domains. The flow is decomposed into distinct executable segments along data domains. Parallel execution paths are identified across the executable segments. The executable segments are executed across a plurality of execution units.