Intelligent Cosourcing in E-Procurement Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic procurement systems face challenges in efficiently distributing sourcing tasks among client devices while maintaining control and oversight, particularly in scenarios requiring specific procurement constraints and optimal pricing or supplier relationships.
Innovation Solution
A system utilizing a trained machine learning model to automatically assign transaction items among distributing client devices based on constraints such as geographical region, previous supplier relationships, and pricing forecasts, generating a user interface to aggregate sourcing information for the designating client device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual distribution of sourcing tasks is used among client devices, then control and oversight can be maintained, but efficiency and effectiveness of task distribution deteriorates
Solution Approach 1:
The patent introduces an intermediary system comprising a server and machine learning model that acts as a mediator between the designating client device and distributing client devices. This intermediary automatically analyzes constraints, predicts outcomes, and makes optimal assignments, thereby improving distribution efficiency without requiring complex manual coordination at each client device level.
Solution Approach 2:
The patent replaces manual mechanical decision-making processes with an automated machine learning-based system. The ML model processes constraints and historical data to automatically determine optimal task assignments, substituting human manual distribution efforts with an intelligent automated system that handles complexity centrally rather than at distributed client levels.
2Productivity
If automated assignment using machine learning is implemented, then distribution efficiency improves, but system complexity increases
Solution Approach 1:
The patent segments the e-procurement system into distinct functional modules: a designating client device for creating events, a server for processing, a machine learning model for prediction and assignment, and distributing client devices for execution. This segmentation allows the complex ML-based automation to be isolated in specific components rather than permeating the entire system, making the overall system more manageable despite the advanced capabilities introduced.
3Adaptability or versatility
If multiple distributing client devices are used, then sourcing capability is enhanced, but coordination difficulty and loss of information increases
Solution Approach 1:
The patent implements feedback mechanisms where the machine learning model continuously receives information about constraint satisfaction, assignment outcomes, and performance metrics from distributing client devices. This feedback loop enables the system to learn from past assignments, refine its predictions, and improve future task distribution decisions, thereby maintaining information coherence across multiple distributed devices while enhancing overall sourcing capability.
Data Source
AI summary
Aspects of the current subject matter are directed to implementing a distribution scenario in a system. In particular, implementations of the current subject matter provide for a designating client device to create a group of distributing client devices for events among and with a plurality of second client devices. Implementations of the current subject matter further relate to automatic assignment of items among the group of distributing client devices, the assignment based on designating client device-established constraints, and to providing an aggregate view of information related to the automatic assignment of the items to allow the designating client device to manage the items.