Back to results
US 11,537,966 B2Granted
Systems and methods for use of a global registry with automated demand profiling via machine learning to optimize inventory management
AssigneeDemand Driven Technologies, Inc.
InventorsScott E. Hrastar, Hugh Dylan Broome
Life of this patent
Estimating the term…
Abstract
In various embodiments, the disclosed system may be utilized in managing supply and demand chains to optimize creation and management of inventory buffer levels via machine learning. In several embodiments, optimizing inventory buffer levels may lead to minimizing costs while maximizing the availability and timeliness of product deliveries. In multiple embodiments, the disclosed system may utilize supply chain data and performance monitoring with automated machine learning and derivation of demand signatures and profiles to provide optimized creation and management of inventory buffer levels.
Claims17
A method, comprising:
- receiving, by at least one processor, supply chain data corresponding to a plurality of parts over a computer network from at least one customer database;
- training, by the at least one processor, an inventory machine learning process using the supply chain data;
- performing the inventory machine learning process by:
- determining, by the at least one processor, a demand profile for each of the plurality of parts based on the supply chain data;
- defining, by the at least one processor, a number of similarity groups for grouping the plurality of parts according to their respective demand profiles;
- selecting, by the at least one processor, a similarity group center for each of the number of similarity groups, wherein the similarity group center comprises a demand profile for a representative part;
- applying, by the at least one processor, a calculation subprocess of the inventory machine learning process on each respective similarity group by iteratively:
- comparing, by the processor executing code to perform a similarity function, the demand profiles for each of the plurality of parts to the demand profiles of the similarity group centers to determine similarity metrics between the demand profiles for each of the plurality of parts and the demand profiles of the similarity group centers;
- assigning, by the at least one processor, each of the plurality of parts into a respective similarity group based on the similarity metrics between the demand profiles for each of the plurality of parts and the demand profiles of the similarity group centers;
- calculating, by the at least one processor, a representative demand profile for each respective similarity group;
- and
- determining, by the at least one processor, whether the calculated representative demand profile for each respective similarity group meets a predetermined group threshold;
- in response to the predetermined group threshold being met, determining, by the at least one processor, calculated inventory parameters for each respective similarity group based on the optimization calculation subprocess;
- and
- associating, by the at least one processor, the calculated inventory parameters for each respective similarity group and the calculated representative demand profile for each respective similarity group with each of the plurality of parts assigned to its respective similarity group.
12 more elements
An inventory management system, comprising:
- a memory configured to store supply chain data;
- and
- at least one processor operatively connected to the memory, the at least one processor configured for:
- receiving supply chain data corresponding to a plurality of parts over a computer network from at least one customer database;
- storing the supply chain data in the memory;
- training an inventory machine learning process using the supply chain data;
- performing the inventory machine learning process by:
- determining a demand profile for each of the plurality of parts based on the supply chain data;
- defining a number of similarity groups for grouping the plurality of parts according to their respective demand profiles;
- selecting a similarity group center for each of the number of similarity groups, wherein the similarity group center comprises a demand profile for a representative part;
- applying a calculation subprocess of the inventory machine learning process on each respective similarity group by iteratively:
- comparing, via a similarity function, the demand profiles for each of the plurality of parts to the demand profiles of the similarity group centers to determine similarity metrics between the demand profiles for each of the plurality of parts and the demand profiles of the similarity group centers;
- assigning each of the plurality of parts into a respective similarity group based on the similarity metrics between the demand profiles for each of the plurality of parts and the demand profiles of the similarity group centers;
- calculating a representative demand profile for each respective similarity group;
- and
- determining whether the calculated representative demand profile for each respective similarity group meets a predetermined group threshold;
- in response to the predetermined group threshold being met, determining calculated inventory parameters for each respective similarity group based on the calculation subprocess;
- and
- associating the calculated inventory parameters for each respective similarity group and the calculated representative demand profile for each respective similarity group with each of the plurality of parts assigned to its respective similarity group.
16 more elements
Dependencies are inferred from the claim text.