Assigning support tickets to support agents

A computer-implemented method for assigning support tickets to support agents, the computer-implemented method comprising:

• providing a computer server including a machine-learning model, an agents and topics data structure and an agents and complication levels data structure;
• training, by the computer server, using a history of closed support tickets, the machine-learning model stored on the computer server to estimate a plurality of complexities resolving closed support tickets, and identify a plurality of support agents who had a plurality of skills handling a plurality of closed support ticket topics and a plurality of experiences resolving closed support tickets of the plurality of complexities, in response to receiving the plurality of closed support tickets;
• identifying a topic of a support ticket, in response to receiving the support ticket;
• estimating, by the trained machine-learning model, one of a plurality of levels of complication of resolving the support ticket based on a body of the support ticket:
• identifying, by the trained machine-learning model, a first set of support agents who have skills handling the identified topic of the support ticket, in the agents and topics data structure which retrieves any topics of support tickets and identifiers of support agents who handle the topics;
• identifying, by the trained machine-learning model, a second set of support agents who have experiences resolving support tickets of the estimated level of complication, in the agents and complication levels data structure which retrieves any levels of complication for resolving support tickets and identifiers of support agents who handle the levels of complication;
• projecting workload availabilities, based on at least one of a time zone compatibility, a time off schedule, and active work hours associated with each support agent of at least some of the first set of support agents and the second set of support agents, for the support ticket;
• generating, by the computer server, a plurality of support agent scores based on the skills handling the identified topic of the support ticket, the experiences resolving the support tickets of the estimated level of complication, and the projected workload availabilities for the support ticket;
• and
• assigning, by the computer server, based on the plurality of support agent scores, the support ticket to a support agent who is identified in both the first set of support agents, identified in the agents and topics data structure, who have skills handling the identified topic of the support ticket, and the second set of support agents, identified in the agents and complication levels data structure, who have experiences resolving support tickets of the estimated level of complication, wherein the agents and topics data structure is mutually exclusive of the agents and complication levels data structure.

A computer program product, comprising a non-transitory computer-readable medium having a computer-readable program code embodied therein to be executed by one or more processors of a server computer, the server computer including a machine-learning model, an agents and topics data structure and an agents and complication levels data structure, the program code including instructions to:

• train, by the server computer, the machine-learning model using a history of closed support tickets to estimate a plurality of complexities resolving closed support tickets, and identify a plurality of support agents who had a plurality of skills handling a plurality of closed support ticket topics and a plurality of experiences resolving closed support tickets of the plurality of complexities, in response to receiving the plurality of closed support tickets;
• identify a topic of a support ticket, in response to receiving the support ticket;
• estimate, by the trained machine-learning model, one of a plurality of levels of complication of resolving the support ticket based on a body of the support ticket:
• identify, by the trained machine-learning model, a first set of support agents who have skills handling the identified topic of the support ticket, in the agents and topics data structure which retrieves any topics of support tickets and identifiers of support agents who handle the topics;
• identify, by the trained machine-learning model, a second set of support agents who have experiences resolving support tickets of the estimated level of complication, in the agents and complication levels data structure which retrieves any levels of complication for resolving support tickets and identifiers of support agents who handle the levels of complication;
• project workload availabilities, based on at least one of a time zone compatibility, a time off schedule, and active work hours associated with each support agent of at least some of the first set of support agents and the second set of support agents, for the support ticket;
• generate, by the server computer, a plurality of support agent scores based on the skills handling the identified topic of the support ticket, the experiences resolving the support tickets of the estimated level of complication, and the projected workload availabilities for the support ticket;
• and
• assign, by the server computer, based on the plurality of support agent scores, the support ticket to a support agent who is identified in both the first set of support agents, identified in the agents and topics data structure, who have skills handling the identified topic of the support ticket, and the second set of support agents, identified in the agents and complication levels data structure, who have experiences resolving support tickets of the estimated level of complication, wherein the agents and topics data structure is mutually exclusive of the agents and complication levels data structure.