US 12,475,018 B2
Utilization of a resource to perform a task based on a predicted capacity of the resource
Eric Zavesky, Austin, TX (US); Ni An, Fremont, CA (US); Prateek Baranwal, Dallas, TX (US); Kelly Dowd, Dallas, TX (US); James Pratt, Round Rock, TX (US); and Emily Williams, Dallas, TX (US)
Assigned to AT&T Intellectual Property I, L.P., Atlanta, GA (US)
Filed by AT&T Intellectual Property I, L.P., Atlanta, GA (US)
Filed on Jul. 29, 2022, as Appl. No. 17/816,187.
Prior Publication US 2024/0037007 A1, Feb. 1, 2024
Int. Cl. G06F 11/34 (2006.01); G06F 9/50 (2006.01)
CPC G06F 11/3442 (2013.01) [G06F 9/505 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
identifying, by a system comprising a processor, a first capacity associated with completion of first tasks of a first task type;
identifying, by the system, collected production data representing a result of a workflow executed by a first worker resource of a plurality of worker resources having a plurality of tasks of a plurality of task types, wherein each of the plurality of tasks are associated with different capability and capacity;
generating a first training data set from the collected production data, wherein the first training data set comprises a first pseudorandom selection of the collected production data;
training, by the system, using the first capacity associated with completion of first tasks of a first task type and the first training data set, an artificial neural network (ANN) to predict resource capacities of the plurality of worker resources;
receiving, by the system, from the ANN, a prediction that the first worker resource has the first capacity;
generating a second training data set from the collected production data, wherein the second training data set comprises a second pseudorandom selection of the collected production data; and
providing subsequent training of the ANN using the second training data set.