FAILURE MAP
← Case archive

FA-63031 / Insurance premium rating / Open access

Parallelogram on-level factor: Each segment is weighted with the following rate level · case 01

Average earned level is overstated so the on-level factor is too small.

Verified by executionVariant 1 · 8 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

Area before a change is multiplied by the level after it.

THE FAILURE

Area before a change is multiplied by the level after it.

Unsuccessful approach: Weighting by the area after the change measures the wrong region.

Case contract

Input changes [[month offset from calendar-year start, change per mille]] in order. Annual policies written uniformly; the share of calendar-year earned exposure written after offset d=m/12 is (1-d)^2/2 for 0<=d<1, 1-(1+d)^2/2 for -1<d<0, 0 for d>=1 and 1 for d<=-1. Levels are cumulative products. OLF = current level / earned-weighted average level, returned *1000 rounded half-up.

Why this case matters

On-level factors restate historical earned premium at current rates before comparing to losses.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json
import math
from fractions import Fraction
N = 1
observations = []
def solve(x):
    def after(m):
        d=Fraction(m,12)
        if d>=1: return Fraction(0)
        if d<=-1: return Fraction(1)
        if d>=0: return (1-d)**2/2
        return 1-(1+d)**2/2
    level=Fraction(1); levels=[level]
    for m,c in x['changes']:
        level*=1+Fraction(c,1000); levels.append(level)
    avg=Fraction(0)
    prev=Fraction(1)
    for i,(m,c) in enumerate(x['changes']):
        a=after(m)
        avg+=levels[i+1]*(prev-a)
        prev=a
    avg+=levels[-1]*prev
    return math.floor(levels[-1]/avg*1000+Fraction(1,2))
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['single mid-year change', {'changes': [[3, 100]]}, 1070], ['change in prior year', {'changes': [[-7, 45]]}, 1004], ['two changes straddling', {'changes': [[-8, 45], [2, -50]]}, 969], ['three changes', {'changes': [[-9, 30], [3, 100], [8, -50]]}, 1020], ['change at year start', {'changes': [[0, 45]]}, 1022], ['change long before period', {'changes': [[-14, -50], [3, 100]]}, 1070], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[10, -80]]}, 921]], [['single mid-year change', {'changes': [[5, 60]]}, 1049], ['change in prior year', {'changes': [[-8, 30]]}, 1002], ['two changes straddling', {'changes': [[-6, 80], [6, 60]]}, 1062], ['three changes', {'changes': [[-6, 30], [4, 30], [7, 100]]}, 1120], ['change at year start', {'changes': [[0, 80]]}, 1038], ['change long before period', {'changes': [[-14, 100], [4, -50]]}, 961], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[6, -80]]}, 929]], [['single mid-year change', {'changes': [[8, 60]]}, 1056], ['change in prior year', {'changes': [[-3, -50]]}, 985], ['two changes straddling', {'changes': [[-7, 100], [2, 100]]}, 1071], ['three changes', {'changes': [[-11, -50], [0, -50], [10, 100]]}, 1070], ['change at year start', {'changes': [[0, 100]]}, 1048], ['change long before period', {'changes': [[-14, -50], [4, 30]]}, 1023], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[4, -80]]}, 937]], [['single mid-year change', {'changes': [[4, 80]]}, 1061], ['change in prior year', {'changes': [[-9, 60]]}, 1002], ['two changes straddling', {'changes': [[-4, 80], [8, 100]]}, 1112], ['three changes', {'changes': [[-2, 30], [2, 60], [8, 60]]}, 1108], ['change at year start', {'changes': [[0, 45]]}, 1022], ['change long before period', {'changes': [[-14, 30], [4, 60]]}, 1046], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[2, -80]]}, 946]], [['single mid-year change', {'changes': [[9, -50]]}, 951], ['change in prior year', {'changes': [[-5, -50]]}, 991], ['two changes straddling', {'changes': [[-8, 80], [5, -50]]}, 962], ['three changes', {'changes': [[-3, 80], [5, 30], [7, 60]]}, 1103], ['change at year start', {'changes': [[0, 100]]}, 1048], ['change long before period', {'changes': [[-14, 45], [4, 60]]}, 1046], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[4, -80]]}, 937]]]
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
single mid-year change10001070Failed
change in prior year10001004Failed
two changes straddling997969Failed
three changes9661020Failed
change at year start10001022Failed
change long before period10001070Failed
no changes10001000Passed
rate decrease1000921Failed

SHA-256 / 16f15d57a912a78d4a017eba81682f94818559eddd3a76c7deaf1a2946f9a516

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json
import math
from fractions import Fraction
N = 1
observations = []
def solve(x):
    def after(m):
        d=Fraction(m,12)
        if d>=1: return Fraction(0)
        if d<=-1: return Fraction(1)
        if d>=0: return (1-d)**2/2
        return 1-(1+d)**2/2
    level=Fraction(1); levels=[level]
    for m,c in x['changes']:
        level*=1+Fraction(c,1000); levels.append(level)
    avg=Fraction(0)
    prev=Fraction(1)
    for i,(m,c) in enumerate(x['changes']):
        a=after(m)
        avg+=levels[i]*a
        prev=a
    avg+=levels[-1]*prev
    return math.floor(levels[-1]/avg*1000+Fraction(1,2))
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['single mid-year change', {'changes': [[3, 100]]}, 1070], ['change in prior year', {'changes': [[-7, 45]]}, 1004], ['two changes straddling', {'changes': [[-8, 45], [2, -50]]}, 969], ['three changes', {'changes': [[-9, 30], [3, 100], [8, -50]]}, 1020], ['change at year start', {'changes': [[0, 45]]}, 1022], ['change long before period', {'changes': [[-14, -50], [3, 100]]}, 1070], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[10, -80]]}, 921]], [['single mid-year change', {'changes': [[5, 60]]}, 1049], ['change in prior year', {'changes': [[-8, 30]]}, 1002], ['two changes straddling', {'changes': [[-6, 80], [6, 60]]}, 1062], ['three changes', {'changes': [[-6, 30], [4, 30], [7, 100]]}, 1120], ['change at year start', {'changes': [[0, 80]]}, 1038], ['change long before period', {'changes': [[-14, 100], [4, -50]]}, 961], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[6, -80]]}, 929]], [['single mid-year change', {'changes': [[8, 60]]}, 1056], ['change in prior year', {'changes': [[-3, -50]]}, 985], ['two changes straddling', {'changes': [[-7, 100], [2, 100]]}, 1071], ['three changes', {'changes': [[-11, -50], [0, -50], [10, 100]]}, 1070], ['change at year start', {'changes': [[0, 100]]}, 1048], ['change long before period', {'changes': [[-14, -50], [4, 30]]}, 1023], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[4, -80]]}, 937]], [['single mid-year change', {'changes': [[4, 80]]}, 1061], ['change in prior year', {'changes': [[-9, 60]]}, 1002], ['two changes straddling', {'changes': [[-4, 80], [8, 100]]}, 1112], ['three changes', {'changes': [[-2, 30], [2, 60], [8, 60]]}, 1108], ['change at year start', {'changes': [[0, 45]]}, 1022], ['change long before period', {'changes': [[-14, 30], [4, 60]]}, 1046], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[2, -80]]}, 946]], [['single mid-year change', {'changes': [[9, -50]]}, 951], ['change in prior year', {'changes': [[-5, -50]]}, 991], ['two changes straddling', {'changes': [[-8, 80], [5, -50]]}, 962], ['three changes', {'changes': [[-3, 80], [5, 30], [7, 60]]}, 1103], ['change at year start', {'changes': [[0, 100]]}, 1048], ['change long before period', {'changes': [[-14, 45], [4, 60]]}, 1046], ['no changes', {'changes': []}, 1000], ['rate decrease', {'changes': [[4, -80]]}, 937]]]
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
single mid-year change18621070Failed
change in prior year5601004Failed
two changes straddling601969Failed
three changes7791020Failed
change at year start10221022Passed
change long before period6691070Failed
no changes10001000Passed
rate decrease34500921Failed

SHA-256 / 32a3389156b40b776ceb3c3e0382e8384802b25c00330f55cc08275f660dd331

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 8 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Sign in to the archive ↗

Verification & scope

A deterministic bounded teaching model of a stipulated toy rating contract. Amounts are integer cents and factors exact rationals; it makes no claim of conformance with any filed rating plan, bureau manual or jurisdiction. 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 2026-09-29T14:47:10.388181+00:00.

Case digest / 220614b53dbc2a7d9fcecb34093d309dd9c8db465567bfc41a79af429beb3069