ECLSS Decision-Support Console

Mission Controls

Simulation rate 2.0 h/s

Degradation intensity 1.0×

Decision weighting balanced

Console Readouts

System health

100%

Lowest RUL

— h

Recommended action

All nominal

Priority Ranking

How it works

Each subsystem publishes a composite stress index S (0 = nominal, 100 = out-of-spec) aggregated from its own raw telemetry — CO2 ppm for the scrubber, cell current for the O2 generator, conductivity for water recovery, coolant ΔT for thermal control. The console never looks at raw sensors directly; it runs three real decision-support algorithms on the index stream:

EWMA:      E(t) = λ·S(t) + (1-λ)·E(t-1),  λ = 0.15
Anomaly z: z = (E(t) - μ) / σ            (control-chart deviation)
Trend:     slope = smoothed dE/dt         (mission-hours)
RUL:       R = max(0, 100 - E(t)) / slope  (hours to the 100 limit)
Priority:  P = C · (wZ·severity + wU·urgency)

severity = anomaly z-score normalized to [0,1] — how far off the statistical baseline the subsystem currently reads.
urgency = 1 − R/250, normalized RUL pressure — how soon the degradation trend reaches the failure threshold if nothing is done.
C is the subsystem's fixed mission-criticality weight (the CO2 scrubber and O2 generator outrank water recovery and thermal control because crew survival time without them is shorter).
Decision weighting slider moves wZ ↔ wU — pure-left ranks by "is it acting weird right now", pure-right ranks by "is it about to run out of margin".
Inject Random Fault shoves one subsystem's index sharply upward, showing how a sudden anomaly changes the ranked recommendation in real time.

This is the same class of system real ECLSS mission-ops teams use: statistical process control plus trend-based Remaining-Useful-Life estimation, combined into a single weighted decision matrix so operators know which of several simultaneously-drifting subsystems to service first.

Nominal Elevated Critical

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.