关于「配比:远源份额的上限由近源弹性决定」,《跨境双源与断供期权》给出的定量结论是什么?
结论:第三章说明最佳远源份额(基准 0.55)的上限由近源量弹性决定,并用逐日仿真发现:库存系统若不剔除逾期的远源货,每年缺货从 11.4 件涨到 259.5 件。
数据口径:书中的价格、交期、断供频率、毛利与产能均为示意;全部计算由书中程序生成;本书是非个人化法律、税务或理财建议;书中每一个到岸价都假定已包含如实申报后应缴的税费,不讨论任何低报、拆单规避或灰色清关手法,海关与关税规则只述大意,以上架前人工复核为准。 书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 11.4, 259.5
| case | fill | lost/yr | on_hand | near_share |
|---|---|---|---|---|
| drop overdue, no outages | 0.9992 | 11.4 | 520.5 | 0.447 |
| drop overdue, outages | 0.9985 | 21.8 | 511.9 | 0.480 |
| count overdue, no outages 0.9821 | 259.5 | 439.7 | 0.438 | |
| count overdue, outages | 0.9866 | 195.1 | 445.5 | 0.474 |
| p21 formula: on hand about 576, lost/yr 9.6 (no outages), 18.0 (outages) |
查看完整验算程序
# Chapter 3. p32: a day-by-day simulation of the base-surge split.
# Model: 50 years of days (first year discarded). Daily demand is
# normal(40, 12) rounded, lost if not on hand. The far source ships a
# constant 308 units (alpha 0.55 of 14 days) every 14 days; its lead time
# is 28 days + gamma delay (mean 8, sd 6) + 20 days with probability 0.2,
# planned at 40. The near source reviews every 7 days, lead time
# normal(5, 1), at most 36 units a day of capacity (base 18 times 2), and
# orders up to 480 + SS on: stock on hand + its own pipeline + far
# shipments planned to land within its 12-day protection interval.
# "drop overdue" leaves out far shipments already past their planned
# date; "count overdue" keeps them. Outages: rate 0.5 a year, mean 40
# days; shipments that fall due inside an outage are cancelled. The SS
# of 359 comes from the p21 formula. Fixed seed; illustrative values.
import math
import random
K_SHAPE = (8.0 / 6.0) ** 2
THETA = 8.0 / K_SHAPE
DAYS, BURN, SS = 50 * 365, 365, 359
def run(offset, drop_overdue, outages):
rng = random.Random(20261113 + offset)
on_hand, near_pipe, far_pipe = 600, [], []
down_until, next_out = -1, rng.expovariate(0.5 / 365.0)
lost = dem_tot = held = 0
far_in = near_in = 0
for t in range(DAYS):
if outages and t >= next_out:
down_until = t + rng.expovariate(1.0 / 40.0)
next_out = down_until + rng.expovariate(0.5 / 365.0)
for pipe, src in ((far_pipe, "f"), (near_pipe, "n")):
keep = []
for item in pipe:
if item[0] <= t:
on_hand += item[2] # step
if t >= BURN:
if src == "f":
far_in += item[2] # step
else:
near_in += item[2] # step
else:
keep.append(item)
pipe[:] = keep
dem = max(0, int(round(rng.gauss(40.0, 12.0))))
sold = min(dem, on_hand)
on_hand -= sold # step
if t >= BURN:
lost += dem - sold # step
dem_tot += dem # step
held += on_hand # step
if t % 14 == 0 and t >= down_until:
lt = 28.0 + rng.gammavariate(K_SHAPE, THETA)
if rng.random() < 0.2:
lt += 20.0 # step
far_pipe.append((t + int(round(lt)), t + 40, 308))
if t % 7 == 3:
due = [x[2] for x in far_pipe if x[1] <= t + 12
and (x[1] >= t or not drop_overdue)]
ip = on_hand + int(math.fsum(x[2] for x in near_pipe))
ip += int(math.fsum(due)) # step
qty = min(36 * 7, max(0, 480 + SS - ip))
lt_n = max(1, int(round(rng.gauss(5.0, 1.0))))
near_pipe.append((t + lt_n, t + 5, qty))
yrs = (DAYS - BURN) / 365.0
return (1.0 - lost / dem_tot, lost / yrs, held / (DAYS - BURN),
near_in / (near_in + far_in))
print("case fill lost/yr on_hand near_share")
for drop in (True, False):
for out in (False, True):
f, l, h, s = run(0, drop, out)
name = ("drop" if drop else "count") + " overdue, " + (
"outages" if out else "no outages")
print("%-25s %6.4f %9.1f %8.1f %10.3f" % (name, f, l, h, s))
print("p21 formula: on hand about %.0f, lost/yr %.1f (no outages), "
"%.1f (outages)" % (359.4 + 0.55 * 40 * 7 + 0.45 * 40 * 3.5,
1146.0 / 120.0, (1146.0 + 1016.0) / 120.0))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《跨境供应链双源冗余与断供实物期权、物流延误清关波动下的最优安全库存配比母典》第 3 章「配比:远源份额的上限由近源弹性决定」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-291 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026291,
author = {Fan, Yafu},
title = {Dual-Source Redundancy and Supply-Disruption Real Options in Cross-Border Chains: Optimal Safety Stock and Source Splits under Shipping Delays and Customs Variability},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}