Prescriptive Engine Optimizes Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Businesses face complexity in effectively allocating resources and actions across multiple assets and participants with multiple constraints and objectives, making it difficult to achieve desired outcomes.
Innovation Solution
A prescriptive engine that receives information on actions and participants, generates suitability information, allocates actions based on this information, deploys the actions, and updates the suitability information based on results, using methods like Thompson Sampling and multi-armed bandit models to optimize assignments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If businesses use large amounts of data and predictive models to determine optimal resource allocation, then the accuracy of predictions improves, but the complexity of the task increases
Solution Approach 1:
The patent segments the complex resource allocation problem into multiple components: predictive models generate initial predictions, prescriptive models optimize allocations across multiple assets and actions, and iterative refinement processes improve suitability information. This segmentation allows each component to handle specific aspects of the problem, reducing overall complexity while maintaining high prediction accuracy.
Solution Approach 2:
The system implements feedback loops where results from deployed actions are collected and used to update suitability information for future allocations. This iterative feedback process refines the predictive and prescriptive models over time, improving accuracy while the system learns from actual outcomes, making the complex task more manageable through continuous improvement.
2Adaptability or versatility
If multiple actions are allocated across multiple assets with multiple constraints, then the comprehensiveness of resource allocation improves, but the difficulty of optimization increases
Solution Approach 1:
The patent employs dynamic optimization where suitability information is updated iteratively based on action results. The system adapts to changing conditions and learns from outcomes, allowing it to handle multiple constraints and objectives dynamically rather than statically. This dynamic approach makes the comprehensive allocation problem more solvable by continuously improving the optimization based on actual performance data.
Solution Approach 2:
The system changes parameters iteratively by updating suitability information based on observed results. The prescriptive engine adjusts allocation parameters dynamically, modifying how resources are distributed across assets and actions based on learned effectiveness. This parameter adjustment approach reduces optimization difficulty by focusing on improving key suitability metrics rather than solving the entire complex problem at once.
Data Source
AI summary
The present disclosure includes a prescriptive engine system and a method of using the prescriptive engine system. The method includes receiving information on actions and receiving information on participants, the information on the participants including first suitability information of at least one participant for at least one of the actions, generating, based on the first suitability information, second suitability information for a set of participants for at least one action, allocating, based on the second suitability information, the at least one action to the set of participants, deploying the at least one action to the set of participants, receiving, after the at least one action has been performed, results of the at least one action for each participant in the set of participants, and updating, based on the received results, the first suitability information of each participant in the set of participants for the at least one action.