What does Private-Message Patience, Handoff and Complaint Blocking conclude, in numbers, about “Queues that Give Up: Erlang C, Erlang-A and Square-Root Staffing”?

Answer: Chapter 2 uses Erlang-A to show that for the same six-person team Erlang C reports a mean wait of 103.6 seconds while, with abandonment counted, it is only 13.1 seconds, at the price of 0.109 of high-intent buyers leaving while they wait.

Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; it is not personalised legal, financial or business advice. Marked boxes separate published evidence, runnable Python calculations (18 programs), classical sources and the author's hypotheses.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 0.109, 103.6, 13.1

load 4.8 erlangs, mean patience 120 s
agentsP(wait)P(abandon)mean wait (s)P(ab)/E[W] per min
30.7730.44152.90.500
40.6210.29935.90.500
50.4570.18822.60.500
60.3070.10913.10.500
70.1860.0587.00.500
80.1030.0293.40.500
90.0510.0131.50.500
6 agents: 0.109 of high-intent buyers leave unanswered;
Erlang C (p21) shows a 103.6 s wait for the same desk, Erlang-A 13.1 s
Show the full program
# Chapter 2. p22: Erlang-A (M/M/n+M, the Palm model) on the same desk.
# Model (illustrative parameters): as in p21, plus exponential patience
# with mean PAT minutes: a waiting buyer leaves at rate THETA = 1/PAT.
# The number in the system is a birth-death chain: births at LAM,
# deaths at min(k, n) MU + max(k - n, 0) THETA.  The chain is solved
# exactly on 0..K (K large enough that the tail is below 1e-15).
# Little's law gives the mean wait; abandonment = THETA E[queue] / LAM,
# so P(abandon) / E[wait] equals THETA exactly in this model.
import math

LAM, HANDLE, PAT, K = 1.2, 4.0, 2.0, 400
MU, THETA = 1.0 / HANDLE, 1.0 / PAT


def solve(n):
    w = [1.0]
    for k in range(1, K + 1):
        death = min(k, n) * MU + max(k - n, 0) * THETA
        w.append(w[-1] * LAM / death)
    z = math.fsum(w)
    p = [x / z for x in w]
    p_wait = math.fsum(p[n:])
    eq = math.fsum((k - n) * p[k] for k in range(n + 1, K + 1))
    return p_wait, THETA * eq / LAM, eq / LAM


print("load {:.1f} erlangs, mean patience {:.0f} s".format(
    LAM * HANDLE, 60.0 * PAT))
print(" agents  P(wait)  P(abandon)  mean wait (s)  P(ab)/E[W] per min")
for n in range(3, 10):
    pw, pab, ew = solve(n)
    print("{:7d}  {:7.3f}  {:10.3f}  {:13.1f}  {:18.3f}".format(
        n, pw, pab, 60.0 * ew, pab / ew))
pw, pab, ew = solve(6)
print("6 agents: {:.3f} of high-intent buyers leave unanswered;".format(
    pab))
print("Erlang C (p21) shows a 103.6 s wait for the same desk, Erlang-A"
      " {:.1f} s".format(60.0 * ew))

Single source and citation

Fan, Y. (2026). Patience-Decay Distributions in Private-Message Queues, Red-Amber-Green Human-AI Handoff Thresholds and Markov Blocking of Complaint Escalation (Version 1.0.0), Chapter 2: Queues that Give Up: Erlang C, Erlang-A and Square-Root Staffing. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-175 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026175,
  author    = {Fan, Yafu},
  title     = {Patience-Decay Distributions in Private-Message Queues, Red-Amber-Green Human-AI Handoff Thresholds and Markov Blocking of Complaint Escalation},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book