Ford Global Technolgoies, LLC

Power Information Network, LLC

Jie Cheng

George Carver Davis

Anya Lynn Tascillo

Stephanie Mary LaCrosse

Rose Peng

Wei Fan

Jie Du

Yi Lu

Charles Edward Newman, Jr

A method for estimating a used vehicle's market value is disclosed. The method searches for a set of comparable vehicles in an existing historical database of used vehicles. Estimations of the used vehicles are established based on each comparable vehicle (neighbor) in the set of comparable vehicles by adjusting their resale value in accordance with the individual differences found between the neighbor(s) and the used vehicle whose market value is to be estimated/predicted (the target vehicle). Finally, a distance-weighted average of all such estimations is calculated to arrive at a market value estimation for the target vehicle.

Method for estimating a used vehicle's market value

A commodity product distribution plan is used to instruct source sites as to how commodity products are to be distributed among target sites. Where the commodity products are to be sold at auction, a wide range of auction prices can be expected due to mixed models, model years, commodity attributes such as color or optional features, economic conditions, and the auction site location itself. Additional factors that contribute to realized auction prices include depreciation and interest rate costs as well as constraints on shipments and auction site capacities. The present invention provides forecast auction prices for the commodity products, taking these various factors into consideration. In this way, an optimized distribution planned aimed at maximizing the potential profit for the commodity products to be sold at auction is generated.

Optimized auction commodity distribution system, method, and computer program product

A virtual vehicle sensor includes a neural network which produces a sensor output based on a linear combination of non-linear physical signals generated by conventional physical sensors. Instead of determining an output directly, the neural network determines the polynomial coefficients as functions of the physical signals indicative of other engine operating parameters. The sensor is manufactured using relatively limited data collection to calibrate a simulation model. The output of the simulation model is used for model-based mapping to generate more comprehensive maps used for training the neural network. The trained neural network is embedded in a controller and acts as the virtual sensor to monitor engine parameters which are difficult to measure or for which conventional physical sensors do not currently exist. The virtual sensor may be used to sense parameters such as in-cylinder residual mass fraction, emission levels, in-cylinder pressure rise during combustion, and exhaust gas temperature.

Virtual vehicle sensors based on neural networks trained using data generated by simulation models

Growing community of inventors

Plymouth, MI, United States of America

Jie Cheng