What does Pocket Command of Container Fleets conclude, in numbers, about “Heartbeats and Probes: Late Is Not Dead”?

Answer: Chapter 4 shows that the phi accrual detector falls far short of its nominal false-alarm rate on a long-tailed link, 0.0001 a day nominal against 51.00 observed at a threshold of 8, so that reliable verdicts need confirmation from several vantage points.

Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; the book offers non-personalised operating guidance, promises no availability figure, and was written without connecting to any machine. Marked boxes separate published evidence, runnable Python calculations (22 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: 0.0001, 51.00

rulefalse suspicions/daynominal/daydetect after crash s
phi 1515.71864.000012.7
phi 2114.0086.400014.8
phi 369.868.640016.3
phi 558.000.086418.6
phi 851.000.000121.3
phi 8 +20s14.710.000141.3
fixed 30 s26.14-30.0
Show the full program
# Chapter 4. p41: phi accrual failure detection on a jittery link
# Heartbeats every 10 s with N(0, 1 s) jitter; 1 in 100 intervals gets
# an extra exponential delay with mean 15 s (a mobile or tunnel hiccup).
# The detector fits a normal to the last 100 intervals and suspects the
# peer once phi = -log10(1 - F(t)) passes the threshold. Seven days of
# synthetic heartbeats, fixed seed. A fixed 30 s timeout is the control;
# the last row adds an acceptable pause of 20 s to the phi 8 limit.
# Window statistics are refreshed every 20 heartbeats.
import random
import statistics
from collections import deque

rng = random.Random(41641)
N = 7 * 8640
gaps = []
for _ in range(N):
    g = max(0.5, rng.gauss(10.0, 1.0))
    if rng.random() < 0.01:
        g += rng.expovariate(1.0 / 15.0)
    gaps.append(g)

print("rule        false suspicions/day  nominal/day  detect after crash s")
for phi, pause in ((1.0, 0), (2.0, 0), (3.0, 0), (5.0, 0), (8.0, 0),
                   (8.0, 20)):
    z = statistics.NormalDist().inv_cdf(1.0 - 10.0 ** -phi)
    win, fs, det, limit = deque(maxlen=100), 0, [], None
    for i, g in enumerate(gaps):
        if len(win) == 100:
            if limit is None or i % 20 == 0:
                mu, sd = statistics.fmean(win), statistics.pstdev(win)
                limit = mu + max(sd, 0.1) * z + pause
            det.append(limit)
            if g > limit:
                fs += 1
        win.append(g)
    nominal = 8640 * 10.0 ** -phi
    tag = "phi {:g}{}".format(phi, " +20s" if pause else "")
    print("{:<10}  {:>20.2f}  {:>11.4f}  {:>20.1f}".format(
        tag, fs / 7.0, nominal, statistics.fmean(det)))
late = 0
for g in gaps:
    if g > 30.0:
        late += 1
print("fixed 30 s  {:>20.2f}  {:>11}  {:>20.1f}".format(
    late / 7.0, "-", 30.0))

Single source and citation

Fan, Y. (2026). Pocket Command of Multi-Container Fleets: Mobile Remote Control, Exception-Only Push Filtering and Away-From-Keyboard Continuity (Version 1.0.0), Chapter 4: Heartbeats and Probes: Late Is Not Dead. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-416 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026416,
  author    = {Fan, Yafu},
  title     = {Pocket Command of Multi-Container Fleets: Mobile Remote Control, Exception-Only Push Filtering and Away-From-Keyboard Continuity},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book