Research — Illustrative (draft, unvalidated)
Failover race — auto
Three failover sequences are held fixed while the time budget sweeps from 1 s upward. Each row flips to qualifying the moment the budget line passes the end of its bar, unless it was too small in capacity to qualify at any budget. The function scores the sequences; it never generates one.
Time budget (s)
1 sfirst qualifying budget16 s
Reading —
Caller-supplied sequences against the time budget (illustrative durations)
Failover scoring
Fastest passing
Gates (yours)
Transfer-switch margin
How the check read it
Client-side port:the logic and table in this page are a TypeScript port of the harness function, checked against values exported from the Python tests. Not a live API call. Domain maturity, quoted from the research page: “Draft; no reliability or controls engineer has reviewed it yet. The availability math is cross-checked against closed-form series, parallel and 2-of-3 cases, and an AI-assisted pre-review found and fixed three issues. IEEE 493 and ISA-75.19.01 were not read directly.” Step names follow the function’s docstring example; durations, capacities and the 90 required capacity are illustrative values, not equipment data.Real function this maps to:
evaluate_failover_sequence_candidates.Not modeled: sequence generation, interlock logic, sensor debounce, valve or pump dynamics, overlapping steps (summed end to end), and any standard-set time budget. The ride-through function is on the interactive page, not here. The tick marks the smallest budget at which any capacity-qualified sequence passes; it is where this example flips, not a code threshold.See also:Failover race, interactive · K-of-N availability, auto walk