Machine Learning RFP Clustering for Award-Likelihood Prioritization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Hoteliers face inefficiencies in prioritizing Requests for Proposals (RFPs) due to the time-consuming process of identifying which RFPs have a high likelihood of being awarded, leading to suboptimal use of resources.

Innovation Solution

A machine learning-driven system for clustering RFPs based on attributes such as event size, duration, and organizational type, enabling identification of high award likelihood clusters and generating alerts for targeted responses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If hoteliers manually review and prioritize RFPs based on their judgment, then they can identify high award likelihood RFPs, but the process consumes excessive time and effort

Engineering Contradiction:

Improveaccuracy of RFP prioritizationVSAvoidtime spent on RFP review

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical review process with an automated machine learning system that uses natural language processing and clustering algorithms to analyze RFP attributes, historical data, and award likelihood predictions, thereby eliminating time-consuming manual evaluation while maintaining or improving accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables RFP prioritization to serve itself by automatically processing RFPs through the machine learning model, which autonomously evaluates attributes, compares historical outcomes, and generates prioritization rankings without requiring human intervention for each RFP review

Inventive Principle:

Principle #25Self-service

2Productivity

If hoteliers respond to all RFPs to maximize opportunities, then they may win more business, but resources are wasted on low award likelihood RFPs

Engineering Contradiction:

ImproveRFP response volumeVSAvoidresource efficiency

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent applies local quality by differentiating RFP responses based on their individual characteristics and award likelihood scores, rather than treating all RFPs uniformly. The system identifies specific high-value RFPs worth pursuing and low-value ones to avoid, allocating resources locally to where they generate maximum return

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system segments RFPs into distinct categories based on clustering analysis of attributes and historical performance, separating high award likelihood RFPs from low likelihood ones. This segmentation enables targeted resource allocation where energy is concentrated on promising opportunities rather than dispersed across all RFPs

Inventive Principle:

Principle #1Segmentation

3Loss of energy

If hoteliers focus on high award likelihood RFPs identified through machine learning clustering, then resource efficiency improves, but the system complexity increases

Engineering Contradiction:

Improveresource efficiencyVSAvoidsystem complexity

Core Design Contradiction:

Loss of energyVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary layer between raw RFP data and decision-making, consisting of the machine learning model and clustering algorithm. This intermediary automatically processes complex analyses, attribute extraction, and pattern recognition, shielding users from system complexity while delivering resource efficiency benefits

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12561604B2System and method for iterative data clustering using machine learning

Publication Date: 2026.02.24 CVENT INC

AI summary

Systems, methods, and non-transitory computer-readable storage media which train a machine learning algorithm using a training set of Requests for Proposals (RFPs), then cluster a second set of RFPs using the trained machine learning algorithm. Distinct clusters are then compared to historical data, and an outlier is identified. An alert regarding that outlier is then transmitted across a network to an entity associated with the outlier.