FA-67521 / Elevator dispatch scheduling / Open access

Hall call waiting time statistics: clock skew clamp · case 01

Negative waits pull the average down.

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

ROOT CAUSE

A skewed answer time before registration yields a negative wait.

VERIFIED REPAIR

Clamp negative waits to zero.

Unsuccessful approach: Taking the absolute value turns skew into a positive wait.

Case contract

Each hall call has a registration time and an answer time (None if still unanswered at end_s). Waits are answer - registration clamped at zero (clock skew); unanswered calls count with the censored wait end_s - registration. Report the integer mean (floor), per-mille of waits over 60 s (floor), the maximum and the count; all zeros when there are no calls.

Why this case matters

Lift group controllers make these decisions many times per minute; a wrong answer strands passengers, wastes trips or overrides a safety rule.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(x):
    waits = []
    for reg, ans in x['calls']:
        if ans is None:
            waits.append(x['end_s'] - reg)
        else:
            waits.append(ans - reg)
    if not waits:
        return {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 0}
    longs = sum(1 for w in waits if w > 60)
    return {'avg': sum(waits) // len(waits), 'long_permille': longs * 1000 // len(waits), 'max': max(waits), 'count': len(waits)}
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 17', {'end_s': 600, 'calls': [[26, None], [377, 374], [447, 492], [321, 411], [154, 151], [571, None], [283, 344]]}, {'avg': 114, 'long_permille': 428, 'max': 574, 'count': 7}), ('sampled regression 16', {'end_s': 600, 'calls': [[293, None], [565, None], [526, 586], [205, 202]]}, {'avg': 100, 'long_permille': 250, 'max': 307, 'count': 4}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('control 1', {'end_s': 600, 'calls': [[93, 138], [499, 519], [546, None], [485, None], [376, 526], [329, 419], [42, None], [303, None]]}, {'avg': 166, 'long_permille': 625, 'max': 558, 'count': 8}), ('sampled regression 4', {'end_s': 600, 'calls': [[15, None], [80, 100], [304, 324], [560, 557], [73, 78], [574, None]]}, {'avg': 109, 'long_permille': 166, 'max': 585, 'count': 6}), ('control 7', {'end_s': 600, 'calls': [[196, 216], [228, 248], [100, None], [381, 471], [370, 415], [56, 76]]}, {'avg': 115, 'long_permille': 333, 'max': 500, 'count': 6})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 57', {'end_s': 600, 'calls': [[218, 216], [34, 184]]}, {'avg': 75, 'long_permille': 500, 'max': 150, 'count': 2}), ('control 48', {'end_s': 600, 'calls': [[235, 234], [112, 202], [453, 498], [10, 30], [166, 186], [359, 358], [466, 471]]}, {'avg': 25, 'long_permille': 142, 'max': 90, 'count': 7}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('boundary: long unanswered call', {'end_s': 600, 'calls': [[400, None], [0, 10], [5, 15]]}, {'avg': 73, 'long_permille': 333, 'max': 200, 'count': 3}), ('control 12', {'end_s': 600, 'calls': [[549, None], [34, 94], [153, 243]]}, {'avg': 67, 'long_permille': 333, 'max': 90, 'count': 3}), ('control 15', {'end_s': 600, 'calls': [[427, 472], [290, 440], [125, None], [18, 79], [88, 178], [156, 306], [7, None], [120, None]]}, {'avg': 255, 'long_permille': 875, 'max': 593, 'count': 8}), ('control 18', {'end_s': 600, 'calls': [[508, 528]]}, {'avg': 20, 'long_permille': 0, 'max': 20, 'count': 1})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 4', {'end_s': 600, 'calls': [[15, None], [80, 100], [304, 324], [560, 557], [73, 78], [574, None]]}, {'avg': 109, 'long_permille': 166, 'max': 585, 'count': 6}), ('boundary: mean with a remainder', {'end_s': 600, 'calls': [[0, 10], [0, 11], [0, 11]]}, {'avg': 10, 'long_permille': 0, 'max': 11, 'count': 3}), ('boundary: no calls', {'end_s': 600, 'calls': []}, {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 0}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('control 23', {'end_s': 600, 'calls': [[121, 141], [388, None], [577, None]]}, {'avg': 85, 'long_permille': 333, 'max': 212, 'count': 3}), ('control 26', {'end_s': 600, 'calls': [[205, 225], [336, 381]]}, {'avg': 32, 'long_permille': 0, 'max': 45, 'count': 2}), ('control 29', {'end_s': 600, 'calls': [[26, 31], [282, 287], [487, 577], [127, 187], [568, None], [130, 280], [335, 395], [184, 274]]}, {'avg': 61, 'long_permille': 375, 'max': 150, 'count': 8})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 44', {'end_s': 600, 'calls': [[294, 299], [351, 396], [215, 276], [154, 244], [532, 577], [340, 337], [420, 481], [449, 510]]}, {'avg': 46, 'long_permille': 500, 'max': 90, 'count': 8}), ('sampled regression 38', {'end_s': 600, 'calls': [[292, 382], [521, 518]]}, {'avg': 45, 'long_permille': 500, 'max': 90, 'count': 2}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('control 34', {'end_s': 600, 'calls': [[526, 531], [100, 160], [338, None], [152, 242], [413, 433], [56, 117]]}, {'avg': 83, 'long_permille': 500, 'max': 262, 'count': 6}), ('sampled regression 37', {'end_s': 600, 'calls': [[542, 547], [483, None], [421, 418], [216, None]]}, {'avg': 126, 'long_permille': 500, 'max': 384, 'count': 4}), ('control 40', {'end_s': 600, 'calls': [[286, 346], [342, None], [309, 308], [91, 96], [519, 524], [518, None], [12, 73]]}, {'avg': 67, 'long_permille': 428, 'max': 258, 'count': 7})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 77', {'end_s': 600, 'calls': [[391, 388]]}, {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 1}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('boundary: long unanswered call', {'end_s': 600, 'calls': [[400, None], [0, 10], [5, 15]]}, {'avg': 73, 'long_permille': 333, 'max': 200, 'count': 3}), ('boundary: mean with a remainder', {'end_s': 600, 'calls': [[0, 10], [0, 11], [0, 11]]}, {'avg': 10, 'long_permille': 0, 'max': 11, 'count': 3}), ('control 45', {'end_s': 600, 'calls': [[54, 204], [360, None], [102, None], [487, None], [396, None]]}, {'avg': 241, 'long_permille': 1000, 'max': 498, 'count': 5}), ('control 48', {'end_s': 600, 'calls': [[235, 234], [112, 202], [453, 498], [10, 30], [166, 186], [359, 358], [466, 471]]}, {'avg': 25, 'long_permille': 142, 'max': 90, 'count': 7}), ('control 51', {'end_s': 600, 'calls': [[553, None], [377, 422], [370, None], [180, 241]]}, {'avg': 95, 'long_permille': 500, 'max': 230, 'count': 4})]]
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 fixture	Actual	Expected	Outcome
regression: answer logged before registration	`{'avg': 14, 'count': 2, 'long_permille': 0, 'max': 30}`	`{'avg': 15, 'count': 2, 'long_permille': 0, 'max': 30}`	Failed
sampled regression 17	`{'avg': 113, 'count': 7, 'long_permille': 428, 'max': 574}`	`{'avg': 114, 'count': 7, 'long_permille': 428, 'max': 574}`	Failed
sampled regression 16	`{'avg': 99, 'count': 4, 'long_permille': 250, 'max': 307}`	`{'avg': 100, 'count': 4, 'long_permille': 250, 'max': 307}`	Failed
boundary: unanswered call at the end	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 100}`	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 100}`	Passed
boundary: wait of exactly sixty seconds	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 61}`	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 61}`	Passed
control 1	`{'avg': 166, 'count': 8, 'long_permille': 625, 'max': 558}`	`{'avg': 166, 'count': 8, 'long_permille': 625, 'max': 558}`	Passed
sampled regression 4	`{'avg': 108, 'count': 6, 'long_permille': 166, 'max': 585}`	`{'avg': 109, 'count': 6, 'long_permille': 166, 'max': 585}`	Failed
control 7	`{'avg': 115, 'count': 6, 'long_permille': 333, 'max': 500}`	`{'avg': 115, 'count': 6, 'long_permille': 333, 'max': 500}`	Passed

SHA-256 / 17cbae8ab1bb1ebe1e5878283ef97a1cdd299f6081e6d8367d310e079d6b296d

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(x):
    waits = []
    for reg, ans in x['calls']:
        if ans is None:
            waits.append(x['end_s'] - reg)
        else:
            waits.append(abs(ans - reg))
    if not waits:
        return {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 0}
    longs = sum(1 for w in waits if w > 60)
    return {'avg': sum(waits) // len(waits), 'long_permille': longs * 1000 // len(waits), 'max': max(waits), 'count': len(waits)}
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 17', {'end_s': 600, 'calls': [[26, None], [377, 374], [447, 492], [321, 411], [154, 151], [571, None], [283, 344]]}, {'avg': 114, 'long_permille': 428, 'max': 574, 'count': 7}), ('sampled regression 16', {'end_s': 600, 'calls': [[293, None], [565, None], [526, 586], [205, 202]]}, {'avg': 100, 'long_permille': 250, 'max': 307, 'count': 4}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('control 1', {'end_s': 600, 'calls': [[93, 138], [499, 519], [546, None], [485, None], [376, 526], [329, 419], [42, None], [303, None]]}, {'avg': 166, 'long_permille': 625, 'max': 558, 'count': 8}), ('sampled regression 4', {'end_s': 600, 'calls': [[15, None], [80, 100], [304, 324], [560, 557], [73, 78], [574, None]]}, {'avg': 109, 'long_permille': 166, 'max': 585, 'count': 6}), ('control 7', {'end_s': 600, 'calls': [[196, 216], [228, 248], [100, None], [381, 471], [370, 415], [56, 76]]}, {'avg': 115, 'long_permille': 333, 'max': 500, 'count': 6})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 57', {'end_s': 600, 'calls': [[218, 216], [34, 184]]}, {'avg': 75, 'long_permille': 500, 'max': 150, 'count': 2}), ('control 48', {'end_s': 600, 'calls': [[235, 234], [112, 202], [453, 498], [10, 30], [166, 186], [359, 358], [466, 471]]}, {'avg': 25, 'long_permille': 142, 'max': 90, 'count': 7}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('boundary: long unanswered call', {'end_s': 600, 'calls': [[400, None], [0, 10], [5, 15]]}, {'avg': 73, 'long_permille': 333, 'max': 200, 'count': 3}), ('control 12', {'end_s': 600, 'calls': [[549, None], [34, 94], [153, 243]]}, {'avg': 67, 'long_permille': 333, 'max': 90, 'count': 3}), ('control 15', {'end_s': 600, 'calls': [[427, 472], [290, 440], [125, None], [18, 79], [88, 178], [156, 306], [7, None], [120, None]]}, {'avg': 255, 'long_permille': 875, 'max': 593, 'count': 8}), ('control 18', {'end_s': 600, 'calls': [[508, 528]]}, {'avg': 20, 'long_permille': 0, 'max': 20, 'count': 1})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 4', {'end_s': 600, 'calls': [[15, None], [80, 100], [304, 324], [560, 557], [73, 78], [574, None]]}, {'avg': 109, 'long_permille': 166, 'max': 585, 'count': 6}), ('boundary: mean with a remainder', {'end_s': 600, 'calls': [[0, 10], [0, 11], [0, 11]]}, {'avg': 10, 'long_permille': 0, 'max': 11, 'count': 3}), ('boundary: no calls', {'end_s': 600, 'calls': []}, {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 0}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('control 23', {'end_s': 600, 'calls': [[121, 141], [388, None], [577, None]]}, {'avg': 85, 'long_permille': 333, 'max': 212, 'count': 3}), ('control 26', {'end_s': 600, 'calls': [[205, 225], [336, 381]]}, {'avg': 32, 'long_permille': 0, 'max': 45, 'count': 2}), ('control 29', {'end_s': 600, 'calls': [[26, 31], [282, 287], [487, 577], [127, 187], [568, None], [130, 280], [335, 395], [184, 274]]}, {'avg': 61, 'long_permille': 375, 'max': 150, 'count': 8})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 44', {'end_s': 600, 'calls': [[294, 299], [351, 396], [215, 276], [154, 244], [532, 577], [340, 337], [420, 481], [449, 510]]}, {'avg': 46, 'long_permille': 500, 'max': 90, 'count': 8}), ('sampled regression 38', {'end_s': 600, 'calls': [[292, 382], [521, 518]]}, {'avg': 45, 'long_permille': 500, 'max': 90, 'count': 2}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('control 34', {'end_s': 600, 'calls': [[526, 531], [100, 160], [338, None], [152, 242], [413, 433], [56, 117]]}, {'avg': 83, 'long_permille': 500, 'max': 262, 'count': 6}), ('sampled regression 37', {'end_s': 600, 'calls': [[542, 547], [483, None], [421, 418], [216, None]]}, {'avg': 126, 'long_permille': 500, 'max': 384, 'count': 4}), ('control 40', {'end_s': 600, 'calls': [[286, 346], [342, None], [309, 308], [91, 96], [519, 524], [518, None], [12, 73]]}, {'avg': 67, 'long_permille': 428, 'max': 258, 'count': 7})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 77', {'end_s': 600, 'calls': [[391, 388]]}, {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 1}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('boundary: long unanswered call', {'end_s': 600, 'calls': [[400, None], [0, 10], [5, 15]]}, {'avg': 73, 'long_permille': 333, 'max': 200, 'count': 3}), ('boundary: mean with a remainder', {'end_s': 600, 'calls': [[0, 10], [0, 11], [0, 11]]}, {'avg': 10, 'long_permille': 0, 'max': 11, 'count': 3}), ('control 45', {'end_s': 600, 'calls': [[54, 204], [360, None], [102, None], [487, None], [396, None]]}, {'avg': 241, 'long_permille': 1000, 'max': 498, 'count': 5}), ('control 48', {'end_s': 600, 'calls': [[235, 234], [112, 202], [453, 498], [10, 30], [166, 186], [359, 358], [466, 471]]}, {'avg': 25, 'long_permille': 142, 'max': 90, 'count': 7}), ('control 51', {'end_s': 600, 'calls': [[553, None], [377, 422], [370, None], [180, 241]]}, {'avg': 95, 'long_permille': 500, 'max': 230, 'count': 4})]]
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 fixture	Actual	Expected	Outcome
regression: answer logged before registration	`{'avg': 16, 'count': 2, 'long_permille': 0, 'max': 30}`	`{'avg': 15, 'count': 2, 'long_permille': 0, 'max': 30}`	Failed
sampled regression 17	`{'avg': 115, 'count': 7, 'long_permille': 428, 'max': 574}`	`{'avg': 114, 'count': 7, 'long_permille': 428, 'max': 574}`	Failed
sampled regression 16	`{'avg': 101, 'count': 4, 'long_permille': 250, 'max': 307}`	`{'avg': 100, 'count': 4, 'long_permille': 250, 'max': 307}`	Failed
boundary: unanswered call at the end	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 100}`	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 100}`	Passed
boundary: wait of exactly sixty seconds	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 61}`	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 61}`	Passed
control 1	`{'avg': 166, 'count': 8, 'long_permille': 625, 'max': 558}`	`{'avg': 166, 'count': 8, 'long_permille': 625, 'max': 558}`	Passed
sampled regression 4	`{'avg': 109, 'count': 6, 'long_permille': 166, 'max': 585}`	`{'avg': 109, 'count': 6, 'long_permille': 166, 'max': 585}`	Passed
control 7	`{'avg': 115, 'count': 6, 'long_permille': 333, 'max': 500}`	`{'avg': 115, 'count': 6, 'long_permille': 333, 'max': 500}`	Passed

SHA-256 / 1021464fd5ae468708c9888c98422e027a41159154042ef8c29524b72a680556

3 / The verified repair

Exit 0

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(x):
    waits = []
    for reg, ans in x['calls']:
        if ans is None:
            waits.append(x['end_s'] - reg)
        else:
            waits.append(max(0, ans - reg))
    if not waits:
        return {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 0}
    longs = sum(1 for w in waits if w > 60)
    return {'avg': sum(waits) // len(waits), 'long_permille': longs * 1000 // len(waits), 'max': max(waits), 'count': len(waits)}
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 17', {'end_s': 600, 'calls': [[26, None], [377, 374], [447, 492], [321, 411], [154, 151], [571, None], [283, 344]]}, {'avg': 114, 'long_permille': 428, 'max': 574, 'count': 7}), ('sampled regression 16', {'end_s': 600, 'calls': [[293, None], [565, None], [526, 586], [205, 202]]}, {'avg': 100, 'long_permille': 250, 'max': 307, 'count': 4}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('control 1', {'end_s': 600, 'calls': [[93, 138], [499, 519], [546, None], [485, None], [376, 526], [329, 419], [42, None], [303, None]]}, {'avg': 166, 'long_permille': 625, 'max': 558, 'count': 8}), ('sampled regression 4', {'end_s': 600, 'calls': [[15, None], [80, 100], [304, 324], [560, 557], [73, 78], [574, None]]}, {'avg': 109, 'long_permille': 166, 'max': 585, 'count': 6}), ('control 7', {'end_s': 600, 'calls': [[196, 216], [228, 248], [100, None], [381, 471], [370, 415], [56, 76]]}, {'avg': 115, 'long_permille': 333, 'max': 500, 'count': 6})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 57', {'end_s': 600, 'calls': [[218, 216], [34, 184]]}, {'avg': 75, 'long_permille': 500, 'max': 150, 'count': 2}), ('control 48', {'end_s': 600, 'calls': [[235, 234], [112, 202], [453, 498], [10, 30], [166, 186], [359, 358], [466, 471]]}, {'avg': 25, 'long_permille': 142, 'max': 90, 'count': 7}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('boundary: long unanswered call', {'end_s': 600, 'calls': [[400, None], [0, 10], [5, 15]]}, {'avg': 73, 'long_permille': 333, 'max': 200, 'count': 3}), ('control 12', {'end_s': 600, 'calls': [[549, None], [34, 94], [153, 243]]}, {'avg': 67, 'long_permille': 333, 'max': 90, 'count': 3}), ('control 15', {'end_s': 600, 'calls': [[427, 472], [290, 440], [125, None], [18, 79], [88, 178], [156, 306], [7, None], [120, None]]}, {'avg': 255, 'long_permille': 875, 'max': 593, 'count': 8}), ('control 18', {'end_s': 600, 'calls': [[508, 528]]}, {'avg': 20, 'long_permille': 0, 'max': 20, 'count': 1})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 4', {'end_s': 600, 'calls': [[15, None], [80, 100], [304, 324], [560, 557], [73, 78], [574, None]]}, {'avg': 109, 'long_permille': 166, 'max': 585, 'count': 6}), ('boundary: mean with a remainder', {'end_s': 600, 'calls': [[0, 10], [0, 11], [0, 11]]}, {'avg': 10, 'long_permille': 0, 'max': 11, 'count': 3}), ('boundary: no calls', {'end_s': 600, 'calls': []}, {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 0}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('control 23', {'end_s': 600, 'calls': [[121, 141], [388, None], [577, None]]}, {'avg': 85, 'long_permille': 333, 'max': 212, 'count': 3}), ('control 26', {'end_s': 600, 'calls': [[205, 225], [336, 381]]}, {'avg': 32, 'long_permille': 0, 'max': 45, 'count': 2}), ('control 29', {'end_s': 600, 'calls': [[26, 31], [282, 287], [487, 577], [127, 187], [568, None], [130, 280], [335, 395], [184, 274]]}, {'avg': 61, 'long_permille': 375, 'max': 150, 'count': 8})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 44', {'end_s': 600, 'calls': [[294, 299], [351, 396], [215, 276], [154, 244], [532, 577], [340, 337], [420, 481], [449, 510]]}, {'avg': 46, 'long_permille': 500, 'max': 90, 'count': 8}), ('sampled regression 38', {'end_s': 600, 'calls': [[292, 382], [521, 518]]}, {'avg': 45, 'long_permille': 500, 'max': 90, 'count': 2}), ('boundary: unanswered call at the end', {'end_s': 600, 'calls': [[500, None], [10, 30]]}, {'avg': 60, 'long_permille': 500, 'max': 100, 'count': 2}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('control 34', {'end_s': 600, 'calls': [[526, 531], [100, 160], [338, None], [152, 242], [413, 433], [56, 117]]}, {'avg': 83, 'long_permille': 500, 'max': 262, 'count': 6}), ('sampled regression 37', {'end_s': 600, 'calls': [[542, 547], [483, None], [421, 418], [216, None]]}, {'avg': 126, 'long_permille': 500, 'max': 384, 'count': 4}), ('control 40', {'end_s': 600, 'calls': [[286, 346], [342, None], [309, 308], [91, 96], [519, 524], [518, None], [12, 73]]}, {'avg': 67, 'long_permille': 428, 'max': 258, 'count': 7})], [('regression: answer logged before registration', {'end_s': 600, 'calls': [[100, 98], [200, 230]]}, {'avg': 15, 'long_permille': 0, 'max': 30, 'count': 2}), ('sampled regression 77', {'end_s': 600, 'calls': [[391, 388]]}, {'avg': 0, 'long_permille': 0, 'max': 0, 'count': 1}), ('boundary: wait of exactly sixty seconds', {'end_s': 600, 'calls': [[0, 60], [100, 161]]}, {'avg': 60, 'long_permille': 500, 'max': 61, 'count': 2}), ('boundary: long unanswered call', {'end_s': 600, 'calls': [[400, None], [0, 10], [5, 15]]}, {'avg': 73, 'long_permille': 333, 'max': 200, 'count': 3}), ('boundary: mean with a remainder', {'end_s': 600, 'calls': [[0, 10], [0, 11], [0, 11]]}, {'avg': 10, 'long_permille': 0, 'max': 11, 'count': 3}), ('control 45', {'end_s': 600, 'calls': [[54, 204], [360, None], [102, None], [487, None], [396, None]]}, {'avg': 241, 'long_permille': 1000, 'max': 498, 'count': 5}), ('control 48', {'end_s': 600, 'calls': [[235, 234], [112, 202], [453, 498], [10, 30], [166, 186], [359, 358], [466, 471]]}, {'avg': 25, 'long_permille': 142, 'max': 90, 'count': 7}), ('control 51', {'end_s': 600, 'calls': [[553, None], [377, 422], [370, None], [180, 241]]}, {'avg': 95, 'long_permille': 500, 'max': 230, 'count': 4})]]
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 fixture	Actual	Expected	Outcome
regression: answer logged before registration	`{'avg': 15, 'count': 2, 'long_permille': 0, 'max': 30}`	`{'avg': 15, 'count': 2, 'long_permille': 0, 'max': 30}`	Passed
sampled regression 17	`{'avg': 114, 'count': 7, 'long_permille': 428, 'max': 574}`	`{'avg': 114, 'count': 7, 'long_permille': 428, 'max': 574}`	Passed
sampled regression 16	`{'avg': 100, 'count': 4, 'long_permille': 250, 'max': 307}`	`{'avg': 100, 'count': 4, 'long_permille': 250, 'max': 307}`	Passed
boundary: unanswered call at the end	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 100}`	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 100}`	Passed
boundary: wait of exactly sixty seconds	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 61}`	`{'avg': 60, 'count': 2, 'long_permille': 500, 'max': 61}`	Passed
control 1	`{'avg': 166, 'count': 8, 'long_permille': 625, 'max': 558}`	`{'avg': 166, 'count': 8, 'long_permille': 625, 'max': 558}`	Passed
sampled regression 4	`{'avg': 109, 'count': 6, 'long_permille': 166, 'max': 585}`	`{'avg': 109, 'count': 6, 'long_permille': 166, 'max': 585}`	Passed
control 7	`{'avg': 115, 'count': 6, 'long_permille': 333, 'max': 500}`	`{'avg': 115, 'count': 6, 'long_permille': 333, 'max': 500}`	Passed

SHA-256 / 18dcd5ee199ed8d54c2eb5adffec017334816f2b11c92f1ace101acd525839d2

Verification & scope

Stipulated toy lift-control contract for a bounded teaching model; it makes no claim of conformance to any lift code or vendor dispatcher and omits real safety cases. 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 2026-09-29T14:47:53.726448+00:00.

Case digest / 962cabd62e77b7142939bcabaf75b258daf3570c8c9a18faac1f8742a168cb00