|
|
QN, a broadband provider, runs a call center to handle telephone calls. When customers
call, the call is immediately picked up by a recorded message which says:
‘Welcome to the QN sales and help line. After the tone, press (1) for sales enquiries,
press (2) for account services, press (3) for technical enquiries, press (4)
for all other enquires or press (5) to repeat this menu’.
If the customer presses numbers 1 to 4 on their handset they are then directed to the
correct line where they await their call to be answered. If they press 5, the recorded
message immediately repeats itself. Data shows that 20% of callers press ‘1’ , 15%
press ‘2’, 20% press ‘3’, 40% press ‘4’ and 5% press ‘5’. It has been found that once a
customer has chosen the correct enquiry line, s/he will only wait for a maximum of 10
minutes before hanging up if the call is not dealt with.
It has been found that 50% of ‘all other enquiries’ are in fact casual enquiries from prospective
‘new’ customers and such calls are redirected to sales. The other 50% of ‘all
other enquiries’ are redirected to ‘technical enquiries’. As such, staff working on the ‘all
other’ line do their job in two parts: first they must determine what type of call it is.
Secondly, they must redirect the call correctly.
Historical ‘sales’ records show that 1 in 10 of ‘new’ customers take up a subscription
immediately. In such circumstances, on finishing the ‘sales’ call, the new customer is
redirected by the sales staff to the technical enquiries where the ‘tech’ staff can discuss
the details of setting up a new broadband account with the new customer.
Re-directing a call on any line takes no time at all.
The call center opens at 2 pm and closes at 10 pm, Monday to Saturday; and it is operated
by eight staff in total. Two members of staff are ‘techs’ dedicated to dealing with
‘technical’ enquiries only. The other six are ‘general’ and have the training and equipment
to deal with all other types of enquiries, i.e. the ‘sales’, ‘accounts’ and ‘all other’,
although each member of staff can only deal with one enquiry at a time. During a shift
the ‘staff’ are divided into teams of two with each team being dedicated to a particular
enquiry line for the entire shift.
On average the call center receives a call every three minutes and the calls are random,
i.e. they are exponentially distributed.
Data collected over time shows the time it takes to deal with a ‘sales’ enquiry is distributed
as follows.
Time to complete sales call (minutes) Number observed
Under 20 163
20 to under 30 17
30 to under 60 20
Dealing with other enquiries takes staff the time indicated by the fitted probability distributions
in the table below:
Activity type Fitted probability distribution
Parameters (time in minutes
)
Recorded message Fixed 0.5
Technical Lognormal Mean = 15; standard deviation
= 8
Accounts Uniform Lower = 1; upper = 5
Determining type of ‘all
other’ enquiries
Average Mean = 3
(a) Create a simul8 model of this system. Call this your baseline model. Simulate the
call centre and print out the results of performance variables that you think
might be of interest to the management of QN. (50 marks).
(b) The management of QN is concerned by the number of dropped calls, i.e. customers
that hang up without being dealt with. They suppose that the number of
‘dropped’ calls can be reduced if the current members of staff were used more
creatively. Can this be done? Experiment with your model and write a report detailing
the what-if-scenarios you have tried, the results you have obtained and
your recommendations, if any, to the management of QN. |
|
[复制链接]
发表于 2010-4-26 07:20:25
