penguineer live demo
Every week, penguineer runs
investigate on the off-spec lots of
Homeostat Live isidis-01, a simulated methane
plant that publishes its sensor data, lab results and data dictionary. Nobody edits these
results: this page is what the open-source engine says, right or wrong.
c7f61af.
Reports: S20260926T2000, S20261005T1700, S20261014T0500. All lab samples with their grade: lots.csv.
How it works: Homeostat Live
guide.investigate on Homeostat Live isidis-01
Window 2026-09-26 12:00 to 2026-10-31 12:00 UTC. 217 lab samples, 2026-09-26 14:00 to 2026-10-23 14:00.
| Grade | Lots | Off spec |
|---|---|---|
| None | 4 | 0 |
| feedstock | 45 | 0 |
| propellant | 73 | 14 |
| surface fuel | 95 | 4 |
Grade changes (H2.rate ramps): propellant to surface fuel 09-27 19:40-22:20; surface fuel to propellant 10-05 13:50-16:30; propellant to surface fuel 10-13 13:40-16:20; surface fuel to feedstock 10-17 21:10-23:40; feedstock to surface fuel 10-25 17:50-19:10; surface fuel to propellant 10-25 19:20-20:40; propellant to surface fuel 10-29 23:50-01:10; surface fuel to feedstock 10-30 01:20-02:40
Tags (role): JT-001 (measurement), FT-101 (measurement), H2.rate (setting), FIC-101.rsp (setpoint), FIC-101.out (output), FIC-101.mode (mode), FFC-102.ratio (setting), FFC-102.out (setpoint), FT-102 (measurement), FIC-102.out (output), FIC-102.mode (mode), TT-201 (measurement), TIC-201.sp (setpoint), TIC-201.out (output), TIC-201.mode (mode), TT-200 (measurement), TT-301 (measurement), DRY.bypass (setting)
7 off-spec episodes (off-spec lots, broken by more than 3 in-spec lots). Severity is the worst lot's distance outside the range, in range widths; marginal means every lot was within lab accuracy of the limit; h after change is hours from the last grade change to the first lot.
| First lot | Last lot | Lots | Grade | Off spec on | Worst lot | Purity | Dew | Severity | Marginal | h after change |
|---|---|---|---|---|---|---|---|---|---|---|
| S20260926T2000 | S20260927T0200 | 3 | propellant | AT-202 | S20260926T2300 | 99.7 | -70.5 | 0.19 | None | |
| S20261005T1700 | S20261006T0500 | 5 | propellant | AT-201, AT-202 | S20261005T1700 | 98.0 | -73.5 | 3.75 | 0.5 | |
| S20261006T2300 | S20261006T2300 | 1 | propellant | AT-202 | S20261006T2300 | 99.7 | -71.0 | 0.12 | yes | 30.5 |
| S20261008T0800 | S20261008T0800 | 1 | propellant | AT-202 | S20261008T0800 | 99.7 | -71.5 | 0.06 | yes | 63.5 |
| S20261009T0800 | S20261009T0800 | 1 | propellant | AT-202 | S20261009T0800 | 99.6 | -71.5 | 0.06 | yes | 87.5 |
| S20261011T0500 | S20261011T1100 | 3 | propellant | AT-202 | S20261011T0800 | 99.7 | -71.0 | 0.12 | yes | 132.5 |
| S20261014T0500 | S20261014T1400 | 4 | surface fuel | AT-201 | S20261014T1400 | 95.4 | -53.5 | 0.30 | 12.7 |
Episode S20261005T1700 (propellant)
Lot S20261005T1700, sampled 0.5 h after a grade change:
investigate: lot S20261005T1700, AT-201 98 % (spec >= 99.5), sampled 2026-10-05 17:00 at SP-product
#1 DRY.bypass high score 1.00 regime_change
The regime changed from 2026-10-05 13:40 to 16:40 (DRY.bypass 0.03 to 0, H2.rate 9 to 4, TIC-201.sp 395 to 340), 0.3 h before the sample; the lot was made during it or right after it.
ruled out:
TT-201: follows the loop:TIC-201 setpoint change
FT-102: follows the loop:FIC-102 setpoint change
FT-101: follows the loop:FIC-101 setpoint change
TIC-201.sp: moved with the regime change
FFC-102.out: moved with the regime change
FIC-101.rsp: moved with the regime change
TT-301: did not move in the lot
TT-200: did not move in the lot
JT-001: did not move in the lot
note: some flows_to edges have no dead time; they count as 0 min
note: the lot spans a regime change; it counts as the regime it ended in
note: 62 earlier lots skipped: different regime
note: AT-201 started to move with lot S20260927T1700, before it left spec
note: the regime changed from 2026-10-05 13:40 to 16:40 (DRY.bypass 0.03 to 0, H2.rate 9 to 4, TIC-201.sp 395 to 340), 0.3 h before the sample
note: no online analyzer on the sampled stream: onset and correlation tests skipped
report: reports/S20261005T1700/report.html
Episode S20261014T0500 (surface fuel)
Lot S20261014T0500:
investigate: lot S20261014T0500, AT-201 95.8 % (spec >= 96), sampled 2026-10-14 05:00 at SP-product
#1 JT-001 low score 0.23 process
JT-001 (EL-101) ran 24.8 % high from 2026-10-14 02:00, 0 min upstream of the sample point.
#2 FT-101 low score 0.23 process
FT-101 (S-H2) ran 2.22 Nm3/h high from 2026-10-14 02:00, 0 min upstream of the sample point.
#3 D-401 low score 0.20 unmeasured_disturbance
no measured tag upstream moved enough to explain AT-201; something unmeasured changed in D-401 or just before the sample (a catalyst, a fouled surface, a feed component nobody analyzes)
#4 TT-200 low score 0.00 process
TT-200 (S-feed) ran 6.86 degC low from 2026-10-14 02:00, 0 min upstream of the sample point.
ruled out:
TT-301: did not move in the lot
TT-201: did not move in the lot
FT-102: did not move in the lot
note: some flows_to edges have no dead time; they count as 0 min
note: 66 earlier lots skipped: different regime
note: no online analyzer on the sampled stream: onset and correlation tests skipped
report: reports/S20261014T0500/report.html
Episode S20260926T2000 (propellant)
Lot S20260926T2000:
investigate: lot S20260926T2000, AT-202 -71.5 degC (spec <= -72), sampled 2026-09-26 20:00 at SP-product
#1 JT-001 medium score 1.00 process
JT-001 (EL-101) ran 24.9 % high from 2026-09-26 17:00, 0 min upstream of the sample point.
#2 FIC-101.rsp medium score 1.00 setpoint_change
The setpoint FIC-101.rsp (loop:FIC-101) was changed 1 Nm3/h high from 2026-09-26 17:00, 0 min upstream of the sample point.
#3 FFC-102.out medium score 1.00 setpoint_change
The setpoint FFC-102.out (loop:FIC-102) was changed 0.251 Nm3/h high from 2026-09-26 17:00, 0 min upstream of the sample point.
#4 TT-200 medium score 0.28 process
TT-200 (S-feed) ran 4.87 degC low from 2026-09-26 17:00, 0 min upstream of the sample point.
ruled out:
FT-102: follows the loop:FIC-102 setpoint change
FT-101: follows the loop:FIC-101 setpoint change
TT-301: its shift of 9.27 is 87% predicted from FT-101 (regression on the reference lots)
FT-101: its shift of 1 is 121% predicted from JT-001 (regression on the reference lots)
TT-201: did not move in the lot
note: some flows_to edges have no dead time; they count as 0 min
note: no in-spec lot of AT-202 before S20260926T2000 in the same regime: compared with the in-spec lots after it (S20260927T0500, S20260927T0800, S20260927T1100)
note: no online analyzer on the sampled stream: onset and correlation tests skipped
report: reports/S20260926T2000/report.html