Erik Lundberg: Applied queueing theory in customer support

There are two aims of this thesis. The first aim is to study how queueing theory can be applied to customer support to obtain information which gives an insight in how to improve the ratio of answered calls to incoming calls. The second aim is to create a model for predicting the ratio of answered calls to incoming calls, given the average time that a customer spends in the queue.

The data has been collected from 26 of Mars to 1 of July 2012. To simulate the actual need for staffing, I have used the data to approximate the amount of incoming calls during a typical week and estimated how many employees that need to be answering incoming calls to obtain a chosen average time that a customer spends in the queue. The model that is used for predicting the ratio of answered calls to incoming calls is a logistic regression.

I found that the logistic regression needs more accurate data to give an accurate result, but that queueing theory nevertheless has many advantages and applications in call centres.