We can also quote,
the service times of calls that are random and whose distribution may depend on
the type of call and the agent that processes the call. The waiting time of a
customer in these systems is very difficult to determine. In fact, for some
call centers, most calls to their arrival find that all agents with the skillsto handle them are busy. These calls are stored in queues invisible to clients
until an agent with the skills to handle them is available to serve them.
Due to the complexity of current models and routing, the amount of time a client has to wait may depend on the length of its queue, the current and future lengths of the other queues, from the group of agents, from the period of the day, etc. For more details on call center problems,2003, Poole and Mandela 2002 and Poole 2013.
In this thesis, we have worked on click here two problems of multi-skill call centers. The first is the prediction of customer waiting time in real time. The second is the modeling and simulation of the agents service lives. We are interested in this work to develop and study methods to estimate the wait time of a client when he arrives at the call center in order to announce this information or part of it to the customer. This information can also be used to make a possible recall proposal to the client in case the estimated waiting time is greater than a certain threshold ex 30 minutes. This threshold can be interpreted by the call center manager as the waiting period beyond which the likelihood of the client leaving the queue before being served High, or be interpreted as call center congestion.
Due to the complexity of current models and routing, the amount of time a client has to wait may depend on the length of its queue, the current and future lengths of the other queues, from the group of agents, from the period of the day, etc. For more details on call center problems,2003, Poole and Mandela 2002 and Poole 2013.
In this thesis, we have worked on click here two problems of multi-skill call centers. The first is the prediction of customer waiting time in real time. The second is the modeling and simulation of the agents service lives. We are interested in this work to develop and study methods to estimate the wait time of a client when he arrives at the call center in order to announce this information or part of it to the customer. This information can also be used to make a possible recall proposal to the client in case the estimated waiting time is greater than a certain threshold ex 30 minutes. This threshold can be interpreted by the call center manager as the waiting period beyond which the likelihood of the client leaving the queue before being served High, or be interpreted as call center congestion.