Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
Door dwell time calculation: dwell cap · case 01
Crowded stops hold the car far beyond the maximum dwell.
Door dwell time calculation: dwell cap · case 02
Crowded stops hold the car far beyond the maximum dwell.
Door dwell time calculation: dwell cap · case 03
Crowded stops hold the car far beyond the maximum dwell.
Door dwell time calculation: dwell cap · case 04
Crowded stops hold the car far beyond the maximum dwell.
Door dwell time calculation: dwell cap · case 05
Crowded stops hold the car far beyond the maximum dwell.
Door dwell time calculation: nudging onset · case 01
The door keeps reopening on the third obstruction instead of nudging.
Door dwell time calculation: nudging onset · case 02
The door keeps reopening on the third obstruction instead of nudging.
Door dwell time calculation: nudging onset · case 03
The door keeps reopening on the third obstruction instead of nudging.
Door dwell time calculation: nudging onset · case 04
The door keeps reopening on the third obstruction instead of nudging.
Door dwell time calculation: nudging onset · case 05
The door keeps reopening on the third obstruction instead of nudging.
Load weighing dispatch modes: overload boundary · case 01
A car loaded exactly to the overload limit refuses to depart, or a heavy overload departs.
Load weighing dispatch modes: overload boundary · case 02
A car loaded exactly to the overload limit refuses to depart, or a heavy overload departs.
Load weighing dispatch modes: overload boundary · case 03
A car loaded exactly to the overload limit refuses to depart, or a heavy overload departs.
Load weighing dispatch modes: overload boundary · case 04
A car loaded exactly to the overload limit refuses to depart, or a heavy overload departs.
Load weighing dispatch modes: overload boundary · case 05
A car loaded exactly to the overload limit refuses to depart, or a heavy overload departs.
Load weighing dispatch modes: bypass threshold · case 01
A car at exactly 80% keeps stopping for hall calls, or a lightly loaded car bypasses them.
Load weighing dispatch modes: bypass threshold · case 02
A car at exactly 80% keeps stopping for hall calls, or a lightly loaded car bypasses them.
Load weighing dispatch modes: bypass threshold · case 03
A car at exactly 80% keeps stopping for hall calls, or a lightly loaded car bypasses them.
Load weighing dispatch modes: bypass threshold · case 04
A car at exactly 80% keeps stopping for hall calls, or a lightly loaded car bypasses them.
Load weighing dispatch modes: bypass threshold · case 05
A car at exactly 80% keeps stopping for hall calls, or a lightly loaded car bypasses them.
Load weighing dispatch modes: nuisance call count · case 01
Genuine car calls of a light passenger are cancelled, or prank calls survive.
Load weighing dispatch modes: nuisance call count · case 02
Genuine car calls of a light passenger are cancelled, or prank calls survive.
Load weighing dispatch modes: nuisance call count · case 03
Genuine car calls of a light passenger are cancelled, or prank calls survive.
Load weighing dispatch modes: nuisance call count · case 04
Genuine car calls of a light passenger are cancelled, or prank calls survive.
Load weighing dispatch modes: nuisance call count · case 05
Genuine car calls of a light passenger are cancelled, or prank calls survive.
Load weighing dispatch modes: nuisance load threshold · case 01
The anti-nuisance check fires at exactly ten percent load or misses a load just under a fractional threshold.
Load weighing dispatch modes: nuisance load threshold · case 02
The anti-nuisance check fires at exactly ten percent load or misses a load just under a fractional threshold.
Load weighing dispatch modes: nuisance load threshold · case 03
The anti-nuisance check fires at exactly ten percent load or misses a load just under a fractional threshold.
Load weighing dispatch modes: nuisance load threshold · case 04
The anti-nuisance check fires at exactly ten percent load or misses a load just under a fractional threshold.
Load weighing dispatch modes: nuisance load threshold · case 05
The anti-nuisance check fires at exactly ten percent load or misses a load just under a fractional threshold.
Trip flight time from a trapezoidal profile: floor rise slice · case 01
Trip distances include the rise above the destination floor.
Trip flight time from a trapezoidal profile: floor rise slice · case 02
Trip distances include the rise above the destination floor.
Trip flight time from a trapezoidal profile: floor rise slice · case 03
Trip distances include the rise above the destination floor.
Trip flight time from a trapezoidal profile: floor rise slice · case 04
Trip distances include the rise above the destination floor.
Trip flight time from a trapezoidal profile: floor rise slice · case 05
Trip distances include the rise above the destination floor.
Trip flight time from a trapezoidal profile: profile regime threshold · case 01
Medium trips are timed as if the car reached contract speed.
Trip flight time from a trapezoidal profile: profile regime threshold · case 02
Medium trips are timed as if the car reached contract speed.
Trip flight time from a trapezoidal profile: profile regime threshold · case 03
Medium trips are timed as if the car reached contract speed.
Trip flight time from a trapezoidal profile: profile regime threshold · case 04
Medium trips are timed as if the car reached contract speed.
Trip flight time from a trapezoidal profile: profile regime threshold · case 05
Medium trips are timed as if the car reached contract speed.
Trip flight time from a trapezoidal profile: cruise formula · case 01
Long trips are predicted too short.
Trip flight time from a trapezoidal profile: cruise formula · case 02
Long trips are predicted too short.
Trip flight time from a trapezoidal profile: cruise formula · case 03
Long trips are predicted too short.
Trip flight time from a trapezoidal profile: cruise formula · case 04
Long trips are predicted too short.
Trip flight time from a trapezoidal profile: cruise formula · case 05
Long trips are predicted too short.
Trip flight time from a trapezoidal profile: triangular formula · case 01
Short hops are predicted faster or slower than the car can achieve.
Trip flight time from a trapezoidal profile: triangular formula · case 02
Short hops are predicted faster or slower than the car can achieve.
Trip flight time from a trapezoidal profile: triangular formula · case 03
Short hops are predicted faster or slower than the car can achieve.
Trip flight time from a trapezoidal profile: triangular formula · case 04
Short hops are predicted faster or slower than the car can achieve.
Trip flight time from a trapezoidal profile: triangular formula · case 05
Short hops are predicted faster or slower than the car can achieve.
Trip flight time from a trapezoidal profile: millisecond rounding · case 01
Predicted trip times are a millisecond short.
Trip flight time from a trapezoidal profile: millisecond rounding · case 02
Predicted trip times are a millisecond short.
Trip flight time from a trapezoidal profile: millisecond rounding · case 03
Predicted trip times are a millisecond short.
Trip flight time from a trapezoidal profile: millisecond rounding · case 04
Predicted trip times are a millisecond short.
Trip flight time from a trapezoidal profile: millisecond rounding · case 05
Predicted trip times are a millisecond short.
Floor label and index mapping: basement order · case 01
The deepest basement is labelled B1.
Floor label and index mapping: basement order · case 02
The deepest basement is labelled B1.
Floor label and index mapping: basement order · case 03
The deepest basement is labelled B1.
Floor label and index mapping: basement order · case 04
The deepest basement is labelled B1.
Floor label and index mapping: basement order · case 05
The deepest basement is labelled B1.
Floor label and index mapping: unlucky floor skip · case 01
Floor 12 disappears instead of 13 in buildings with a mezzanine.
Floor label and index mapping: unlucky floor skip · case 02
Floor 12 disappears instead of 13 in buildings with a mezzanine.
Floor label and index mapping: unlucky floor skip · case 03
Floor 12 disappears instead of 13 in buildings with a mezzanine.
Floor label and index mapping: unlucky floor skip · case 04
Floor 12 disappears instead of 13 in buildings with a mezzanine.
Floor label and index mapping: unlucky floor skip · case 05
Floor 12 disappears instead of 13 in buildings with a mezzanine.
Floor label and index mapping: index range guard · case 01
A negative index returns a label from the top of the building.
Floor label and index mapping: index range guard · case 02
A negative index returns a label from the top of the building.
Floor label and index mapping: index range guard · case 03
A negative index returns a label from the top of the building.
Floor label and index mapping: index range guard · case 04
A negative index returns a label from the top of the building.
Floor label and index mapping: index range guard · case 05
A negative index returns a label from the top of the building.
Floor label and index mapping: label case folding · case 01
Lower-case labels such as g or b1 are not found.
Floor label and index mapping: label case folding · case 02
Lower-case labels such as g or b1 are not found.
Floor label and index mapping: label case folding · case 03
Lower-case labels such as g or b1 are not found.
Floor label and index mapping: label case folding · case 04
Lower-case labels such as g or b1 are not found.
Floor label and index mapping: label case folding · case 05
Lower-case labels such as g or b1 are not found.
Idle car parking by zone: zone size rounding · case 01
An extra tiny zone appears at the top, or zones are too large when floors divide evenly.
Idle car parking by zone: zone size rounding · case 02
An extra tiny zone appears at the top, or zones are too large when floors divide evenly.
Idle car parking by zone: zone size rounding · case 03
An extra tiny zone appears at the top, or zones are too large when floors divide evenly.
Idle car parking by zone: zone size rounding · case 04
An extra tiny zone appears at the top, or zones are too large when floors divide evenly.
Idle car parking by zone: zone size rounding · case 05
An extra tiny zone appears at the top, or zones are too large when floors divide evenly.
Idle car parking by zone: up-peak lobby pair · case 01
During up-peak only one car waits at the lobby.
Idle car parking by zone: up-peak lobby pair · case 02
During up-peak only one car waits at the lobby.
Idle car parking by zone: up-peak lobby pair · case 03
During up-peak only one car waits at the lobby.
Idle car parking by zone: up-peak lobby pair · case 04
During up-peak only one car waits at the lobby.
Idle car parking by zone: up-peak lobby pair · case 05
During up-peak only one car waits at the lobby.
Idle car parking by zone: down-peak ordering · case 01
In down-peak the upper zones are left without cars.
Idle car parking by zone: down-peak ordering · case 02
In down-peak the upper zones are left without cars.
Idle car parking by zone: down-peak ordering · case 03
In down-peak the upper zones are left without cars.
Idle car parking by zone: down-peak ordering · case 04
In down-peak the upper zones are left without cars.
Idle car parking by zone: down-peak ordering · case 05
In down-peak the upper zones are left without cars.
Idle car parking by zone: nearest car tie-break · case 01
Equally near cars are chosen by listing order, or a far car is chosen.
Idle car parking by zone: nearest car tie-break · case 02
Equally near cars are chosen by listing order, or a far car is chosen.
Idle car parking by zone: nearest car tie-break · case 03
Equally near cars are chosen by listing order, or a far car is chosen.
Idle car parking by zone: nearest car tie-break · case 04
Equally near cars are chosen by listing order, or a far car is chosen.
Idle car parking by zone: nearest car tie-break · case 05
Equally near cars are chosen by listing order, or a far car is chosen.
Car call registration and cancellation: door reopen condition · case 01
Pressing the current floor with doors closed is swallowed instead of registered.
Car call registration and cancellation: door reopen condition · case 02
Pressing the current floor with doors closed is swallowed instead of registered.
Car call registration and cancellation: door reopen condition · case 03
Pressing the current floor with doors closed is swallowed instead of registered.
Car call registration and cancellation: door reopen condition · case 04
Pressing the current floor with doors closed is swallowed instead of registered.
Car call registration and cancellation: door reopen condition · case 05
Pressing the current floor with doors closed is swallowed instead of registered.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗