Probabilistic Allocation of Indivisible Resources

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods struggle to efficiently allocate indivisible resources to agents over a finite horizon while balancing resource utilization and fairness.

Innovation Solution

A method that involves receiving information about agents' requests for indivisible resources, assigning weights to each agent based on their importance, calculating agent-specific allocation probabilities, and determining resource allocations for each round based on these probabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If resources are allocated to maximize resource utilization, then resource efficiency improves, but fairness among agents deteriorates

Engineering Contradiction:
Improveresource utilizationVSAvoidfairness
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent transforms the allocation decision from a deterministic choice to a probabilistic one by introducing allocation probabilities as a new parameter. Each agent is assigned a probability of receiving the resource based on their historical utilization rate, transforming the binary allocation problem into a continuous probability space that balances efficiency and fairness

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The allocation probability is dynamically adjusted based on each agent's historical resource utilization. Agents who have not yet utilized the resource receive higher probabilities in subsequent rounds, creating a dynamic feedback mechanism that adapts to past behavior and promotes fairness over time

Inventive Principle:
Principle #15Dynamics

2Reliability

If resources are allocated to ensure fairness among all agents, then fairness improves, but resource utilization efficiency deteriorates

Engineering Contradiction:
ImprovefairnessVSAvoidresource utilization
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system implements a feedback mechanism where each agent's historical resource utilization is tracked and fed back into the allocation probability calculation. This feedback loop ensures that agents who have not yet received or utilized resources are given higher priority in future allocations, promoting fairness while maintaining overall system efficiency

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary calculations of allocation probabilities before actual resource distribution occurs. By pre-computing probabilities based on historical data and current state, the system prepares a fairness-oriented allocation strategy in advance that can be executed efficiently without compromising real-time performance

Inventive Principle:
Principle #10Preliminary action

3Reliability

If allocation decisions are made based on complex fairness criteria, then fairness improves, but computational complexity increases

Engineering Contradiction:
ImprovefairnessVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent simplifies the fairness criterion by reducing it to a single parameter: the allocation probability based on historical utilization. This transforms a potentially complex multi-dimensional fairness problem into a straightforward probabilistic calculation that is computationally efficient while still capturing the essence of fairness

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system extracts only the essential information needed for fair allocation - the historical utilization rate of each agent - and uses this extracted data to compute allocation probabilities. By focusing on the most critical fairness indicator and ignoring less important factors, the system achieves fairness with minimal computational overhead

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS20250130860A1Method and system for probabilistic allocation of indivisible resources
Publication Date: 2025.04.24 JPMORGAN CHASE BANK NA
  • US20250130860A1 patent drawing
  • US20250130860A1 patent drawing
  • US20250130860A1 patent drawing

AI summary

A system and a method for allocating indivisible resources to agents over a finite horizon in a manner that balances resource utilization and fairness are provided. The method includes: receiving first information that relates to a number of agents included in a set of agents; receiving second information that relates to whether each respective agent is requesting an allocation of an indivisible resource during a particular allocation round; assigning, to each respective agent, a respective weight that relates to an importance of the respective agent; calculating, for each respective agent, a respective agent-specific allocation probability; and determining an allocation of the indivisible resource for the particular allocation round based on the agent-specific allocation probabilities. The determination of the allocation may be adjusted based on a budget that indicates an availability of the resource during the particular allocation round.