AI Team Formation System for RFP Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack automation in identifying and forming teams of experts to respond to funding agency requests for proposals (RFPs), leading to inefficiencies and high rates of false alerts, which require manual tracking and are not suited for multidisciplinary opportunities.
Innovation Solution
A data-driven system using natural language processing and analytical techniques to recommend teams of experts from various faculties and departments based on RFP requirements, prioritizing those with complementary skills and past collaboration success, and providing budget and success probability estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword-based alert systems are used to notify researchers about RFPs, then researchers receive notifications about opportunities, but the system generates high rates of false positives and false negatives making alerts unusable
Solution Approach 1:
The patent replaces simple keyword-matching mechanics with natural language processing and machine learning models that understand semantic meaning, context, and research alignment. This substitution enables more accurate interpretation of RFP requirements and researcher expertise, significantly reducing false positives and false negatives in alert generation.
Solution Approach 2:
The patent introduces an intermediary layer of AI-based analysis between the RFP database and researcher notifications. This intermediary processes RFP text, extracts requirements, matches them against researcher profiles using semantic understanding, and filters results before generating alerts. This intermediary layer is the key to reducing false alerts while maintaining high alert accuracy.
2Measurement precision
If manual tracking of RFPs is performed by researchers, then researchers can identify suitable opportunities, but this process is time-consuming and inefficient
Solution Approach 1:
The patent enables the system to automatically perform the RFP identification and matching task that previously required manual researcher effort. The AI system continuously monitors RFP databases, extracts requirements, matches them against researcher profiles, and generates alerts without researcher intervention. This self-service automation eliminates time-consuming manual tracking while maintaining or improving identification accuracy through sophisticated matching algorithms.
Solution Approach 2:
The patent performs preliminary analysis of RFPs by pre-processing and storing extracted requirements, themes, and key criteria before researchers need to review them. This preliminary action prepares matching data in advance, enabling rapid and accurate comparison with researcher profiles when opportunities arise, thereby eliminating the need for researchers to manually analyze RFP texts at the moment of need.
3Adaptability or versatility
If existing matching systems are used for RFPs, then some opportunities can be identified, but they lack capability to suggest team formations and collaborative groups
Solution Approach 1:
The patent extends the matching system to perform multiple functions: identifying RFP opportunities, analyzing requirements, matching individual researchers, and forming collaborative teams. The same NLP and machine learning infrastructure that matches individual researchers to RFPs is extended to analyze collaboration patterns, identify complementary skill sets, and suggest optimal team compositions. This multi-functionality adds team formation capability without proportionally increasing complexity.
Solution Approach 2:
The patent merges individual researcher matching with team formation by integrating collaboration network analysis and skill complementarity assessment into the existing matching framework. The system combines data about individual expertise with data about past collaborations and potential synergies to generate comprehensive team recommendations. This merging creates a unified system that handles both individual and collaborative matching seamlessly.
Data Source
AI summary
The disclosure deals with a system and method for building teams in response to a teaming opportunity. In one exemplary embodiment disclosed herewith, a system and method for building teams for Request for Proposals (RFPs) is described where potential team participants are researchers at one or more institutions. A computer-based method and computer system, given RFPs from funding agencies like NSF, DOE and NASA, recommends a team of experts from various faculties and departments of the organization, like a university, that would best fit the needs of the RFP and have a high chance of putting a successful proposal together. The system generates teams that may match the requirements of an RFP. In addition, the system optimizes the list of teams to maximize winning success and to reduce redundancy. The system input includes RFPs and the researchers' public information. The system output is a list of proposed teams, each team with two or more members. Optionally, each team will have an estimation of the team's budget and proposal success chances. The disclosed methodology is more broadly applicable to team-building opportunities in general.