Dynamic Memoryless Demand-Supply Pegging for Parallel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional demand-supply pegging methods, such as FIFO, rely on ordered calculation, making it difficult to uniquely peg a demand to a supply without considering previous demands, leading to inefficient flow plans and memory constraints.
Innovation Solution
A dynamic memoryless approach that allows independent demand pegging at any point in the supply chain, enabling parallel processing and reducing memory usage by calculating demand pegging independently and flushing to disk storage when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional FIFO pegging is used, then demands are pegged to supplies in sequence, but it reduces flow plans and increases memory usage due to dependency on previous demands
Solution Approach 1:
The patent segments the pegging process into independent demand-level operations. Each demand is pegged independently to available supplies without requiring sequential processing of other demands. This segmentation eliminates the chain dependency inherent in FIFO pegging, allowing parallel processing while maintaining accurate demand-supply matching.
Solution Approach 2:
The patent creates a virtual copy of the supply chain state for each demand pegging operation. Instead of modifying the actual flow plans sequentially, each demand is pegged against a copied view of available supplies, ensuring that the original supply chain state remains intact and can be reused for multiple demand pegging operations simultaneously.
2Reliability
If traditional FIFO pegging is used, then demands are processed in order, but it increases memory constraints due to need to track previous demands
Solution Approach 1:
Each demand pegging operation is made self-sufficient by providing it with the complete supply chain state information it needs independently. The system calculates available supplies for each demand without requiring memory of previous demand processing, as each operation starts with a fresh view of supply availability. This eliminates the need to maintain state information across sequential operations.
3Productivity
If dynamic memoryless pegging is implemented, then parallel processing is enabled, but calculation complexity increases
Solution Approach 1:
The patent performs preliminary calculation of available supplies at each supply chain node before demand pegging operations. By pre-calculating and caching the supply state information, the system enables multiple demands to be pegged in parallel without each demand requiring complex real-time calculations. This preliminary action reduces the computational burden during actual pegging operations.
Data Source
AI summary
Systems and techniques for dynamic memoryless demand-supply pegging are described herein. An identification may be received of a demand to be pegged. An inventory may be determined for a material node at a time bucket. A set of other demands may be identified for the material node at the time bucket. A quantity to be pegged may be calculated. A quantity to skip may be calculated based on a sum of demand units corresponding to members of the set of other demands having a lower priority than the demand. A supply set of units may be identified. A pegged subset of the supply set of units may be generated by skipping members of the supply units equal to the quantity to skip. A graphical representation of the supply set of units for the time bucket may be output including identification of members of the pegged subset.