Netting Proposal System Using ML for Approval Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing netting systems for accounts payable and receivable between multiple trading partners face complexity and low approval rates due to the difficulty in predicting which proposals will be accepted, especially in large trading groups with thousands of transactions, leading to increased processing costs and resource utilization.

Innovation Solution

A netting proposal system utilizing a machine learning model that identifies characteristics of previously accepted and rejected proposals to generate new proposals likely to be approved, selecting attributes associated with successful proposals while avoiding those linked to rejections, thereby optimizing proposal content and improving approval likelihood.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional netting systems process thousands of transactions manually, then complete analysis of all transactions is achieved, but processing costs and time consumption increase significantly

Engineering Contradiction:

Improveprocessing speedVSAvoidprocessing costs

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The machine learning model automatically analyzes historical netting data and generates optimized proposals without requiring manual intervention for each proposal, enabling the system to serve itself in identifying patterns and making recommendations

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical review processes with an automated machine learning system that uses algorithms to analyze transactions and generate netting proposals, substituting human effort with computational automation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Quantity of substance

If netting proposals include more transactions to maximize offsetting, then greater financial benefit is achieved, but approval likelihood decreases due to increased complexity

Engineering Contradiction:

Improvenumber of transactions offsettedVSAvoidproposal approval likelihood

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The machine learning model analyzes historical data to identify optimal parameters for netting proposals, such as the ideal number of transactions to include and the appropriate offsetting amounts, adjusting these parameters to maximize both financial benefit and approval likelihood

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system uses historical approval and rejection data as feedback to continuously improve proposal generation, learning from past outcomes to optimize future proposal compositions and increase approval rates

Inventive Principle:

Principle #23Feedback

3Measurement precision

If machine learning model analyzes all historical data thoroughly, then prediction accuracy improves, but computational resources and processing time increase

Engineering Contradiction:

Improveprediction accuracyVSAvoidmodel training time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The machine learning model extracts only the most relevant features and patterns from historical netting data that are predictive of approval outcomes, rather than processing all raw data equally, thereby reducing computational burden while maintaining accuracy

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS11276117B2Generating payables and receivables netting proposals based on historical information

Publication Date: 2022.03.15 ORACLE INT CORP

AI summary

Operations include generating proposals for accounts payable and accounts receivable netting across multiple trading partners. A netting proposal system uses a machine learning model to generate netting proposals that are likely to be approved by the respective trading partners. The netting proposal system identifies characteristics of attributes of previously accepted netting proposals. Candidate attributes, with the same characteristics of attributes of previously accepted netting proposals, are selected for a new netting proposal. The netting proposal system further identifies characteristics of attributes of previously rejected netting proposals. Candidate attributes, without the same characteristics of attributes of previously rejected netting proposals, are selected for the new netting proposal. Furthermore, candidate attributes, with the same characteristics of attributes of previously rejected netting proposals, are not selected for the new netting proposal.