Intelligent Lender Selection via Machine Learning Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The vehicle purchasing process is complex and time-consuming, particularly in the financing stage, where manual selection of lenders can lead to inefficiencies and human bias, resulting in potential revenue loss and high customer dropout rates.
Innovation Solution
An intelligent lender selection system that utilizes machine learning modules to predict the probability of loan approval and determine optimal lending entities based on customer and vehicle data, reducing the time spent on lender selection and mitigating human bias.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual selection of lending entities is used, then human bias can be introduced, but the process requires significant time (15-40 minutes) and expertise
Solution Approach 1:
The patent replaces the manual mechanical process of lender selection with an automated machine learning system. The ML model analyzes customer data, vehicle data, and lender characteristics to automatically predict loan approval probabilities and select optimal lenders, eliminating the need for manual expert evaluation while reducing process time from 15-40 minutes to instantaneous automated decision-making
Solution Approach 2:
The system enables self-service by allowing the automated ML model to independently perform lender selection without requiring human F&I manager intervention. The model autonomously processes customer information, evaluates multiple lenders, and generates loan application recommendations, freeing human workers from this time-consuming task
2Productivity
If manual lender selection is performed, then human bias results in lost potential revenue, but the process is complex and requires multiple steps
Solution Approach 1:
The patent transforms the complex qualitative judgment process into quantitative parameter analysis. The ML model evaluates multiple parameters including customer credit score, income, vehicle value, lender approval rates, and profitability metrics simultaneously, converting subjective human bias into objective data-driven decisions that maximize revenue while simplifying the process
3Reliability
If the financing process takes 15-40 minutes, then customers may be disconnected and walk away (15-20% dropout rate), but thorough lender evaluation is needed
Solution Approach 1:
The system performs preliminary actions by pre-evaluating multiple lenders and pre-calculating loan approval probabilities before the customer completes the purchase. The ML model has lenders pre-ranked based on historical data and customer profiles, so when a customer is ready to finance, the optimal lender is already identified, eliminating waiting time and preventing customer dropout
Data Source
AI summary
Methods for intelligent lender selection is provided. A customer identifier associated with a customer is received and customer variables are determined based on the customer identifier. Vehicle variables associated with a vehicle to be transferred to the customer are received. Loan variables associated with a loan application by the customer for transfer of the vehicle to the customer are determined based on the vehicle variables and the customer variables. A probability of acceptance of the loan application by the customer for transfer of the vehicle by each of a plurality of lending entities is predicted based on the loan variables, the vehicle variables, and the customer variables. One or more lending entities are filtered from the plurality of lending entities based on the probability of acceptance and a minimum probability of acceptance defined by an administrator. The loan application is submitted to each of the one or more lending entities.