Machine learning model to provide predicted print material usage correction factor
AssigneeHewlett-Packard Development Company, L.P.
InventorsJeffrey H Luke, Gabriel S McDaniel, Scott K Hymas
Life of this patent
Estimating the term…
Abstract
A machine learning model provides a correction factor to predicted print material usage on a per-print job basis. The machine learning model is trained based on print job information of print jobs printed by printing devices using print material from cartridges received on a per-print job basis, and based on predicted print material usage by the printing devices in printing the print jobs received on a per-print job basis. The machine learning model is further trained based on measured actual remaining print material within the cartridges after the cartridges have been replaced within the printing devices by new cartridges.
Claims9
A method comprising:
- receiving, by a processor of a computing device, print job information of print jobs printed by printing devices using print material from cartridges, and predicted print material usage by the printing devices in printing the print jobs, on a per-print job basis;
- receiving, by the processor, actual remaining print material within the cartridges measured after the cartridges have been replaced within the printing devices by new cartridges;
- training, by the processor and based on the received print job information, the received predicted print material usage, and the received actual remaining print material, a machine learning model that provides a correction factor to the predicted print material usage on the per-print job basis;
- receiving, by the processor and from a printing device, print job information and predicted print material usage of a print job printed by the printing device using print material from a cartridge already used to print one or more other print jobs, wherein the predicted print material usage is determined by the printing device;
- applying, by the processor, the machine learning model to the received print job information and the received predicted print material usage to determine the correction factor for the print job;
- and
- sending, by the processor, the correction factor to the printing device, wherein the printing device applies the predicted print material usage to determine more accurate predicted print material usage, such that the printing device determines the more accurate predicted print material usage.
A non-transitory computer-readable data storage medium storing program code executable by a processor to perform processing comprising:
- receiving, from a printing device, print job information and predicted print material usage of a print job printed by the printing device using print material from a cartridge, wherein the predicted print material usage is determined by the printing device;
- applying a machine learning model to the received print job information and the predicted print material usage by the printing device in printing the print job to determine a correction factor to apply to the predicted print material usage to determine more accurate predicted print material usage of the print job;
- and
- sending, to the printing device, the determined correction factor for the print job, wherein the printing device applies the determined correction factor to the predicted print material usage, such that the printing device determines the more accurate predicted print material usage.
A printing device comprising:
- network hardware to communicatively connect to a computing device over a network;
- printing hardware to print a print job on print media using print material from a cartridge;
- and
- logic hardware to:
- calculate predicted print material usage based on print job information of the print job printed by the printing hardware;
- send the predicted print material usage and the print job information to the computing device;
- responsively receive a correction factor from the computing device, wherein the computing device determines the correction factor by applying a machine learning model to the predicted print material usage and the print job information;
- and
- apply the correction factor to the predicted print material usage to determine more accurate predicted print material usage of the print job.
Dependencies are inferred from the claim text.