Solutions · Simulation
Understand the range of outcomes before you commit.
Explore demand, queues, service levels and uncertain conditions through distributions and scenarios — not a single number pretending to be certain.
60 calls an hour: 7 people, running at 85.7% — 6 would be at 100%.
60 calls an hour: 7 people on the desk
The hour
7people on the desk
- Calls
- 60 an hour
- Work
- 60 × 6 = 360 talk-minutes
- Supplied
- 7 × 60 = 420 minutes
- Load
- 85.7%
Deliberately under full. A desk supplying exactly the minutes the calls demand has nothing left for a busy ten minutes, and the queue that starts there does not clear on its own.
Every desk size tried
| People | Load | Spare |
|---|---|---|
| 5 | 120% | none |
| 6 | 100% | none |
| 7 | 85.7% | yes — recommended smallest |
| 8 | 75% | yes |
| 9 | 66.7% | yes |
6 people would run at exactly 100% — every minute of the hour already spoken for. 7 is the first size with anything left over, which is why it is the answer rather than 8.
Checks
- Passed: Workload from the quoted call volume
- Passed: Every desk size from 1 up tried
- Passed: Smallest size with spare capacity kept
What this brief supplied
- Call volume
- 60 an hour
- Call length
- 6 minutes on average
- Staffed time
- 60 minutes an hour per person
- How long callers wait
- Not stated — not stated in this brief. This is workload arithmetic, not a queueing model. It answers whether the desk keeps up over an hour; it cannot tell you a wait time or a service level.
- How call length varies
- Not stated — not stated in this brief. Six minutes is an average, and the brief gives no spread. Variation is exactly what the spare capacity is for, but sizing it properly would need the real distribution.
- Arrival pattern within the hour
- Not stated — not stated in this brief. Calls are treated as spread across the hour. A lunchtime peak would need staffing by the interval rather than by the hour.
- Breaks, training and admin
- Not stated — not stated in this brief. Each person is credited with a full hour on calls. Any time off the phones would mean a larger desk than this.
A fan of futures, one median
Trajectories fan out from today; the teal line holds the middle and the histogram shows the spread. That spread — not one number — is the honest answer.
What you can decide
Questions with ranges for answers
Probability-weighted forecasts
Demand, revenue, or load projected as a range with likelihoods, not a single guess.
Queues and wait times
Service desks, docks, and lines: how long the wait gets and what staffing changes it.
What-if scenarios
Change an assumption and re-run; compare the resulting ranges side by side.
Decisions in stages
Choices that unfold over time under uncertainty, played forward before you make the first one.
How it runs
Rehearse, then decide
Frame the uncertainty
What varies, how much, and what you control — written out, not modeled by hand.
Run the futures
The runs condense into a range, a median, and the chance of the outcome you care about.
Decide on the range
The recommendation names the odds it rests on, so the risk you accept is explicit.
More solutions