Supply Chain Analytics via Automatic Differentiation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Complex supply chain management systems lack the ability to isolate the direct impact of each element and identify how changes contribute to the overall supply chain picture, making it difficult to analyze bottlenecks and high-risk areas effectively.

Innovation Solution

The implementation of an automatic differentiation engine and analytical solver that instruments supply chain logic to provide transparent analytics, allowing for the extraction of low-level data and the determination of the impact of each variable on Key Performance Indicators (KPIs) through the use of path-integrated gradients and directed acyclic graphs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If complex supply chain analytics algorithms are implemented to optimize supply chain management, then productivity and optimization capability are improved, but the ability to isolate the cause of specific issues and identify direct impact of each element deteriorates

Engineering Contradiction:

Improvesupply chain optimization capabilityVSAvoidability to isolate cause and identify direct impact

Core Design Contradiction:

ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the complex supply chain analytics into multiple layers of abstraction. The system divides the analytics code into discrete computational steps and tracks variable interactions at each layer, enabling granular analysis of individual elements' impacts while preserving overall optimization capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary analysis layer that sits between the complex analytics algorithms and the user interface. This intermediary layer captures variable interactions, computes impact metrics, and presents simplified causal relationships to users, bridging the gap between complex computations and interpretable results.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If additional code is written to analyze bottlenecks and high-risk areas after supply chain execution, then measurement precision is improved, but loss of time increases

Engineering Contradiction:

Improveanalysis precision of bottlenecks and risk areasVSAvoidruntime for separate analysis code

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by instrumenting the supply chain analytics code during development to automatically capture variable interactions and compute impact metrics during normal execution. This eliminates the need for separate post-execution analysis code, as the analytical data is collected concurrently with the primary supply chain optimization operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent merges the primary supply chain optimization functionality with the analytical analysis functionality into a single integrated system. The same code execution that performs supply chain optimization also simultaneously captures variable interactions and computes impact metrics, eliminating redundant execution time.

Inventive Principle:

Principle #5Merging (Combining)

3Measurement precision

If separate stand-alone software is built to analyze specific supply chain issues, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improveanalysis capability precisionVSAvoidsoftware architecture complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements a universal analysis framework that can analyze multiple types of supply chain issues (bottlenecks, high-risk areas, variable impacts) through a single integrated system. The same instrumentation and analysis infrastructure handles diverse analytical queries, eliminating the need for separate specialized software tools for different analysis types.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20230095571A1Methods and systems of sequential latent inference of complex knowledge

Publication Date: 2023.03.30 KINAXIS INC

AI summary

Methods and systems that allow for supply chain logic to be instrumented in such a way that a supply planner can see the major factors driving KPIs, as well as drill down to the lower level to see the impact of each item at the smallest possible level.