Skip to content

Sink Dependency Dashboard

Monitoring dashboard — instantiates Source–Sink Viability Management

Tracks each sink's dependency in real time — how much support it draws, how close it sits to its viability threshold, and which flows it relies on — so hidden fragility and lock-in surface before an interruption exposes them.

The Sink Dependency Dashboard is the standing monitor trained on the demand side. For every unit that persists only on imported support, it displays the shape of that dependence — the amount, reliability, and duration of support required, the pathways it arrives by, and how far the unit currently sits above the threshold below which it is nonviable. Its distinguishing question is not "is this sink worth supporting" (a decision) nor "where does the surplus come from" (the source side) but how dependent, how fragile, and how exposed to interruption each sink is right now — the intelligence a support decision needs but does not itself make.

Example

A national grid operator has three regions that import most of their electricity across long inter-ties from a generation-rich fourth region. On paper all four "have power," which hides the asymmetry. The Sink Dependency Dashboard makes it explicit. For each importing region it shows import dependence as a share of demand (Region A: ≈78%), the reliability of the supplying inter-ties, how long the region could self-supply if a line tripped (Region A: ≈40 minutes of local generation and storage before load-shed), and its distance to the viability threshold — the demand level below which it cannot island safely. Region A lights red not because it lacks power today but because its dependence is deep, its buffer thin, and its two inter-ties share a single corridor, so one fault interrupts both. The dashboard doesn't decide whether to build local generation or sign a firmer import contract; it makes the fragility legible so that decision isn't made blind.

How it works

  • Dependence profile, not just consumption. For each sink it carries amount, reliability, duration, arrival pathway, and interruption risk together — the full shape of the reliance, not a single usage number.
  • Distance to threshold. Every sink is shown against its viability threshold, so the reading is "how much headroom before nonviability," not raw demand.
  • Flow provenance. It maps which inbound flows each sink depends on and how concentrated they are, surfacing single-corridor and single-supplier exposure a steady-state view hides.

It reports dependence; it does not judge whether the dependence is wise or plan its exit.

Tuning parameters

  • Threshold definition — where "nonviable" is drawn for each sink; a conservative threshold flags fragility early, an aggressive one waits until the sink is nearly gone.
  • Reliability window — how far back interruption history is weighted; short windows react to the last outage, long ones smooth over the rare catastrophic one.
  • Interruption-risk granularity — whether shared pathways are modelled as correlated failure or treated independently; ignoring correlation badly understates single-corridor exposure.
  • Alerting sensitivity — how thin a buffer trips a warning; sensitive settings catch lock-in early but risk normalising red.
  • Dependence horizon — whether the board shows only current dependence or projects how it trends as the sink grows or decays.

When it helps, and when it misleads

Its strength is separating apparent health ("the lights are on") from structural dependence, and surfacing the interruption exposure — shared corridors, single suppliers — that a steady-state view hides until the day it fails. Its failure modes: a dashboard normalises whatever it shows, so a chronically red sink can quietly become accepted rather than acted on; and by making dependence merely visible it can encourage the soft budget constraint,[1] where a sink that knows support will always appear loses the pressure to close its own deficit. The classic misuse is treating deep-but-stable dependence as fine because nothing has broken yet — mistaking the absence of an interruption for resilience. The discipline is to read threshold distance and interruption risk together — stable-but-thin-and-concentrated is the dangerous quadrant — and to route a persistently red sink to a decision (recover, convert, taper, or firm up) rather than leave it glowing on the board forever.

How it implements the components

The Sink Dependency Dashboard fills the sink-intelligence components — the machinery that characterises dependence and fragility — and only those:

  • sink_dependency_profile — it assembles amount, reliability, duration, pathway, and interruption risk into one live profile per sink.
  • sink_viability_threshold — it shows each sink's headroom above the level below which it cannot persist.
  • support_flow_map — it maps the inbound flows each sink relies on and how concentrated they are.

It does NOT monitor the sources that fund these flows — that's the Source Depletion Dashboard — and it does NOT decide whether to sustain, taper, or convert a dependent sink, which the Support Taper Plan and Restoration Priority Matrix handle.

References

[1] The soft budget constraint (János Kornai's term for chronically-subsidised enterprises) describes a unit that survives on the expectation of rescue rather than on its own balance, and so loses the incentive to close its own deficit. A dependency dashboard can inadvertently reinforce it by making rescue look routine. ↩