FAILURE MAP
← Case archive

FA-65581 / Ecological population dynamics / Open access

Levins metapopulation with habitat destruction: equilibrium destruction term · case 01

Equilibrium occupancy ignores habitat destruction.

Verified by executionVariant 1 · 7 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The destroyed fraction is not subtracted from the equilibrium.

VERIFIED REPAIR

Restore the equilibrium destruction term rule: `eq = max(0.0, 1 - destroyed - e / c)`.

Unsuccessful approach: Multiplying by 1-D understates the extinction threshold effect.

Case contract

Occupancy p: p += c*p*(1-D-p) - e*p each year, clamped to [0, 1-D]; equilibrium max(0, 1-D-e/c); return [equilibrium rounded 6, trajectory rounded 6]; None for c<=0, e<0 or D outside [0,1).

Why this case matters

Population projections set harvest quotas, conservation status and pest-control timing; a wrong update order, boundary or rate conversion silently changes management advice.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(c, e, destroyed, p0, years):
    if c <= 0 or e < 0 or not 0 <= destroyed < 1:
        return None
    eq = max(0.0, 1 - e / c)
    p = p0
    traj = []
    for _ in range(years):
        p = p + c * p * (1 - destroyed - p) - e * p
        p = min(max(p, 0.0), 1 - destroyed)
        traj.append(round(p, 6))
    return [round(eq, 6), traj]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])]]
for label, args, expected in fixtures[N - 1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression: butterfly network[0.8, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]][0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]Failed
control: intact habitat[0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]][0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]Passed
regression: heavy destruction extinction[0.333333, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]][0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]Failed
regression: overshoot clamp[0.98, [0.7, 0.665, 0.689937]][0.68, [0.7, 0.665, 0.689937]]Failed
regression: no extinction[1.0, [0.284, 0.388966, 0.508231, 0.627697]][0.9, [0.284, 0.388966, 0.508231, 0.627697]]Failed
control: invalid destructionNoneNonePassed
regression: empty network[0.8, [0.0, 0.0, 0.0]][0.6, [0.0, 0.0, 0.0]]Failed

SHA-256 / a320225317005294ffd22c20de1af9ba17ee620b45d54c23f3cd477a939ba427

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(c, e, destroyed, p0, years):
    if c <= 0 or e < 0 or not 0 <= destroyed < 1:
        return None
    eq = max(0.0, (1 - destroyed) * (1 - e / c))
    p = p0
    traj = []
    for _ in range(years):
        p = p + c * p * (1 - destroyed - p) - e * p
        p = min(max(p, 0.0), 1 - destroyed)
        traj.append(round(p, 6))
    return [round(eq, 6), traj]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])]]
for label, args, expected in fixtures[N - 1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression: butterfly network[0.64, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]][0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]Failed
control: intact habitat[0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]][0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]Passed
regression: heavy destruction extinction[0.166667, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]][0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]Failed
regression: overshoot clamp[0.686, [0.7, 0.665, 0.689937]][0.68, [0.7, 0.665, 0.689937]]Failed
regression: no extinction[0.9, [0.284, 0.388966, 0.508231, 0.627697]][0.9, [0.284, 0.388966, 0.508231, 0.627697]]Passed
control: invalid destructionNoneNonePassed
regression: empty network[0.64, [0.0, 0.0, 0.0]][0.6, [0.0, 0.0, 0.0]]Failed

SHA-256 / 1826fab9194487c6b635208b6bbb6371c2f9159d4753619fc0feca04a769a088

3 / The verified repair

Exit 0
"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(c, e, destroyed, p0, years):
    if c <= 0 or e < 0 or not 0 <= destroyed < 1:
        return None
    eq = max(0.0, 1 - destroyed - e / c)
    p = p0
    traj = []
    for _ in range(years):
        p = p + c * p * (1 - destroyed - p) - e * p
        p = min(max(p, 0.0), 1 - destroyed)
        traj.append(round(p, 6))
    return [round(eq, 6), traj]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: overshoot clamp', (2.5, 0.05, 0.3, 0.6, 3), [0.68, [0.7, 0.665, 0.689937]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])],
 [('regression: butterfly network',
   (0.5, 0.1, 0.2, 0.1, 6),
   [0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]),
  ('control: intact habitat',
   (0.4, 0.1, 0.0, 0.05, 5),
   [0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]),
  ('regression: heavy destruction extinction',
   (0.3, 0.2, 0.5, 0.3, 6),
   [0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]),
  ('regression: no extinction', (0.6, 0.0, 0.1, 0.2, 4), [0.9, [0.284, 0.388966, 0.508231, 0.627697]]),
  ('control: invalid destruction', (0.5, 0.1, 1.0, 0.1, 3), None),
  ('regression: empty network', (0.5, 0.1, 0.2, 0.0, 3), [0.6, [0.0, 0.0, 0.0]]),
  ('regression: high turnover',
   (1.2, 0.8, 0.25, 0.4, 5),
   [0.083333, [0.248, 0.198995, 0.171376, 0.15327, 0.140407]])]]
for label, args, expected in fixtures[N - 1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression: butterfly network[0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]][0.6, [0.125, 0.154687, 0.18913, 0.227984, 0.27039, 0.314952]]Passed
control: intact habitat[0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]][0.75, [0.064, 0.081562, 0.103369, 0.130106, 0.162367]]Passed
regression: heavy destruction extinction[0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]][0.0, [0.258, 0.225131, 0.198669, 0.176895, 0.158663, 0.143177]]Passed
regression: overshoot clamp[0.68, [0.7, 0.665, 0.689937]][0.68, [0.7, 0.665, 0.689937]]Passed
regression: no extinction[0.9, [0.284, 0.388966, 0.508231, 0.627697]][0.9, [0.284, 0.388966, 0.508231, 0.627697]]Passed
control: invalid destructionNoneNonePassed
regression: empty network[0.6, [0.0, 0.0, 0.0]][0.6, [0.0, 0.0, 0.0]]Passed

SHA-256 / 9cb7b199534e8b6d0ddeb16b8d0c045d5d184078049026de3264d9077d42b4db

Verification & scope

Deterministic bounded teaching model with a stipulated contract; not a validated scientific or public-health modelling library. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:47:35.313395+00:00.

Case digest / e7c36c5af323f24a407f5b9e74a146e5cc8b962855d20a0ef2369db5ae01f5bd