How to calculate amount of call center's operators? Erlang Calculator on python

Question

I have stuck with a problem of predicting how many operators are needed for call center to answer the calls (80%) in less than 20 seconds (in queue). Just something like this Erlang Calculator

But I need to implement that in a python script. I did found basic Erlang-c formula but it only returns probability that an arriving customer will need to queue.

Appreciate any help!

Answer 1

def ErlangC(A, N):
    if (N-A<=0): 
        return 1 
    L = (A**N / factorial(N)) * (N / (N - A))
    sum_ = 0
    for i in range(N):
        sum_ += (A**i) / factorial(i)
    return (L / (sum_ + L))

def calc_sl(row):
    if row['duration'] == 0:
        return 0
    A = row['duration']
    T = 20 # targeted waiting time
    aht = row['duration'] * 3600 / row['calls']
    for staff in range(1000):
        P = ErlangC(A, staff)
        sl = (1 - (P * math.exp(-((staff - A) * (T / aht))))) * 100
        if sl >= 80: # percent of calls to be in tagreted waiting time
            return staff

Answer 2

You can use pyworkforce for this problem, you just need to define your parameters and goals (all the time parameters are in minutes), for example:

from pyworkforce.queuing import ErlangC

erlang = ErlangC(transactions=100, asa=20/60, aht=3, interval=30, shrinkage=0.3)

positions_requirements = erlang.required_positions(service_level=0.8, max_occupancy=0.85)
print("positions_requirements: ", positions_requirements)

This will return the needed resources considering shrinkage = 0 (raw_resources), the resources under the provided shrinkage, the probability that a transaction waits in the queue, the expected service level and the expected resources occupancy

Output:

{'raw_positions': 14,
 'positions': 20,
 'service_level': 0.8883500191794669,
 'occupancy': 0.7142857142857143, 
 'waiting_probability': 0.1741319335950498}

Answer 3

You didn't even state your criteria of filtering. If I have to guess I guess you want to filter out the third element, which is not a valid JSON. You can wrap JSON.parse() in a try/catch block to silent the JSON parse error, return a null then filter it out.

const arr = [
  '{"Topic":"TRACK","SID":"1","GPSlat":"634312740","GPSlong":"104014684","GPSat":"43788","GPSspd":"20","GPShead":"29499222","GPSepoch":"1638451303"}',
  '{"Topic":"TRACK","SID":"1","GPSlat":"634312740","GPSlong":"104014684","GPSalt":"43788","GPSspd":"20","GPShead":"29499222","GPSepoch":"1638451303"}',
   "{'143567890'}",
  '{"Topic":"TRACK","SID":"1","GPSlat":"634312740","GPSlong":"104014684","GPSalt":"43788","GPSspd":"20","GPShead":"29499222","GPSepoch":"1638451303"}',
  '{"Topic":"Impact","SID":"1","GPSlat":"634312740","GPSlong":"104014684","GPSalt":"43788","GPSspd":"20","GPShead":"29499222","GPSepoch":"1638451303"}'
 
];

const output = arr.map( it => {
  try{
    return JSON.parse(it)
  }
  catch(e){
    return null
  }
}).filter(it => !!it);
console.log(output);