Working out headcount

Without a sample sheet, your request doesn't make sense to me!! Please read the yellow banner (top of page) and post a sample sheet with some manually calculated results.

Most places use erlang to work out headcount for a call centre, but is there a way to do calls and tickets together to work out headcount?

There's no straight forward way to do it.

Here's general guide line.
1. Using Erlang C traffic model, calculate number of staff required to maintain your SL target.
2. Using same model, calculate average occupancy rate per agent, making sure to account for breaks etc.
3. Then using average handle time of ticket(s). Calculate how many agents are required to process queue(s), meeting SL agreement.

EDIT:
Ex: Lets say to meet 80/20 SL, you need 15 agents, but this puts them at 85% occupancy. This leave little to no room for them to work ticket queue. So you will need to add more agents.
Let's say that you add more agents to bring occupancy rate down to 75%. Then you have about 10% of agent's time to work on ticket queue. Calculate if this is enough to meet your need to process tickets, using avg ticket handle time.

This will give you base value to start your model. Then you'll need to adjust according to your specific business and mix of agent skill level.

It can be bit more complicated, if you mix agent skill sets which are not identical across agents, as well as mixing dedicated agent (working queue) etc.

As well, most call centre does not have flat call volume every hour, there is hourly peak and valleys, which should be accounted for. So your staffing should have varied start time etc.

FYI - Never staff agents for over 88% occupancy or more (for entire shift). This will burn out agents and will result in high-turnover rate, which will increase your staffing cost in the end.
Target is usually between 80% to 87% occupancy to strike balance between keeping agents busy and giving enough breathing time.

EDIT2: I used to be SL manager, WF manager at mid-size call centre (150 to 200 seats depending on season). That had mix of calls & ticket queue. Let me see if I have any sample from my old job. It's been at least few years since I last used Erlang C model for staffing... so don't hold your breath.

Tickets can be measured, not in open-close duration. But on actual time spent handling it. Typically speaking, no tickets are worked on constantly for more than few hours (typically less than 15 min, and don't count time spent by IT or other staff outside of front-line).

Get enough sample to calculate average handle time for tickets. This will give you basis for staffing calculation.

If you have some support staff (QA, HelpDesk, etc), that can support queue, this will help as well to buffer peak volume.

Start off by measuring your current staffing vs current volume stats. Check if all the targets are met. Adjust your model accordingly (if meeting all targets, then try reducing # of staff, else increase staff).

Once you build your model, you'll need to schedule periodic review to validate it. Making adjustments as necessary.

For interval, I'd recommend using 30min or Hourly level break-down for each day of the week. Then create schedule to fill in staffing requirements for each hour.

Sorry, I looked through my archives, but I can't locate the one I used to use.