Direct Answer

After the monitoring platform goes live, there is a pile of alarms but nobody closes them out. How should the disposition process be designed? The answer: four elements — graded thresholds (normal / attention / warning / alarm / linkage), the responsible person and response time limit for each grade, automatic work-order dispatch with closure records, and automatic escalation on timeout. The platform keeps "alarm–dispatch–disposition–review" fully recorded so every alarm has a result instead of sinking into the sea.

After an campus distribution monitoring system went live, it generated over 400 alarms in the first month. Operators were overwhelmed; repeated alarms made up 60%; nothing was closed out; hidden risks kept recurring. Redone with the four-element scheme: alarms graded into five levels (normal/attention/warning/alarm/linkage), each with a responsible person and time limit (e.g., alarm level = duty electrician, respond in 30 min, dispose in 8 h); the platform auto-dispatched, disposition was filled back, and timeouts auto-escalated to the supervisor. One month later, repeated alarms dropped 80%; every alarm had a closure record; the closure rate exceeded 95%. The core of an alarm-disposition scheme: grading separates priorities, responsible persons give everything an owner, closure records give every item a result, and escalation keeps the loop unbroken.

Because:

  • No grading = no priority: all alarms at one level; the operator does not know what to handle first;
  • No responsible person = nobody owns it: an alarm with no owner recurs with nobody managing it;
  • No closure record = no loop: disposition not filled back makes alarms untraceable and problems "invisible";
  • No escalation = no safety net: no mechanism intervenes on timeout; critical alarms can be drowned out.

Key insight: an alarm-disposition scheme solves management closure, not alarm technology. Grading, responsibility, closure, escalation — with all four, alarms have results.

At one plant, a night alarm (busbar temperature, warning level) was not responded to by the duty electrician within 40 minutes; the system auto-escalated to the supervisor with a phone call; the supervisor coordinated and completed disposition with closure within 2 hours. Without the escalation mechanism, that alarm might have waited until the next day — 8 hours of sustained high temperature, risks compounding. Automatic timeout escalation is the loop's safety net, keeping alarms from being drowned out at night.

1. Graded Thresholds and Response-Time Table

Step one: how alarms are graded and time limits set.

Five-level alarm definition:

  • Normal: data within baseline; no attention needed;
  • Attention: over baseline but below threshold, stable trend — record and observe;
  • Warning: over threshold with rising trend; near-term inspection needed (e.g., 24–72 h);
  • Alarm: clear anomaly; dispose promptly (e.g., respond in 30 min, dispose in 8 h);
  • Linkage: severe anomaly with high risk; automatic linked disposition (e.g., trip, fire-fighting activation, shedding non-critical load).

Where graded thresholds come from:

  • OEM recommended values (temperature, PD, leakage thresholds);
  • On-site baseline measurement (background noise, normal fluctuation range);
  • Historical incident review (precursor values before incidents);
  • National standards / procedures (follow the regulation where one exists).

Response-time table:

  • Attention: review within 48 h (record and observe);
  • Warning: on-site check within 24 h;
  • Alarm: respond in 30 min, dispose in 8 h (duty electrician);
  • Linkage: dispose immediately (auto linkage + manual confirmation).

Meaning of grading:

  • Operators see priority at a glance (which to handle first);
  • Response time limits are clear (no delay, no misjudgment);
  • Alarm volume matches manpower (no longer one pot of everything after grading).

Key insight: graded thresholds are the priority ruler. Five-level definition + threshold sources + response limits — priority and rhythm are clear the moment an alarm arrives.

2. Responsible-Person Matrix

Step two: who gets the alarm. Responsibility lands on a person:

Responsible-person matrix design:

  • Each alarm type defines a responsible role (job-level):

- Temperature/PD/leakage alarm → duty electrician / inspector; - Electrical-fire alarm → duty electrician + fire linkage (double confirmation); - System/communication alarm → platform administrator; - Linkage-disposition alarm → shift leader / supervisor (command decision);

  • Backup responsible person (takes over when the primary is absent);
  • Contact methods (phone + platform message).

Matrix elements:

  • Alarm type (temperature/PD/leakage/fire/communication);
  • Responsible role (who disposes);
  • Response time limit (how fast to respond);
  • Disposition authority (what they can do: inspect/isolate/de-energize);
  • Reviewer (who checks disposition quality).

Dispatch logic:

  • Platform auto-matches the responsible person by alarm type (auto dispatch);
  • Responsible person accepts the order (confirm receipt);
  • Timeout without acceptance → backup/escalation (auto flow);
  • Disposition complete → reviewer confirms (quality gate).

Meaning of the matrix:

  • Everything has an owner (every alarm has a home);
  • Independent of personal experience (job-based; the post remains when the person leaves);
  • Dispatch is automated (operators miss nothing).

Key insight: the responsible-person matrix is the ownership ruler. Type → role → time limit → authority → review, five elements define responsibility — every alarm has an owner.

3. Work-Order Flow Process

Step three: how alarms flow. The work order is the closure carrier:

Work-order flow:

  • Platform alarm → auto-generate work order (with alarm data);
  • Work order dispatched → responsible person accepts (confirm receipt);
  • On-site disposition → record actions (inspect/handle/replace);
  • Result filled back → disposition result, photos, data uploaded;
  • Review and closure → reviewer confirms; work order closed.

Work-order content design:

  • Alarm info: type, location, value, time;
  • Disposition guidance: suggested actions (inspect/isolate/de-energize);
  • Responsibility info: responsible person, time limit, contact;
  • Result records: process, result, residual risk;
  • Timestamps: alarm–accept–dispose–closure full timeline.

Flow principles:

  • Every step has a timestamp (fully traceable);
  • Every step has an owner (no break in the flow);
  • Disposition results are retained (review sees them);
  • Not closed until closed out (problems do not go "invisible").

Meaning of the flow:

  • Alarms are visible from "issued" to "closed";
  • Recurring problems have evidence to check;
  • Management closure (no more sinking into the sea).

Key insight: the work-order flow is the closure carrier. Alarm → dispatch → disposition → review → closure, five steps — each with a stamp, an owner, and a trace.

4. Timeout Escalation Rules

Step four: what if nobody handles it. Escalation is the safety net:

Timeout escalation rules:

  • Response timeout (e.g., alarm level not accepted in 30 min) → escalate to the responsible person's superior;
  • Disposition timeout (e.g., not complete in 8 h) → escalate to the supervisor;
  • Consecutive timeouts → escalate to higher levels (shift leader → supervisor → department head);
  • Escalation channels: platform message + SMS + phone (multi-channel reach).

Escalation trigger conditions:

  • Acceptance timeout (auto-detected);
  • Disposition timeout (deadline passed without closure);
  • Re-measurement failed (data still abnormal after closure);
  • Linkage alarm (immediately escalate to the shift leader).

Effect of escalation:

  • Critical alarms are not drowned out (safety net engages);
  • Responsibility is enforced (timeout = accountability issue, forcing response);
  • Management visibility (escalation records = management evidence).

Escalation notes:

  • Reasonable escalation thresholds (not hypersensitive, not sluggish);
  • Escalation records retained (time, level);
  • The original responsible person still follows up after escalation (no abdication).

Typical linkage-disposition scenarios:

  • Busbar-temperature linkage level: auto-shed non-critical load (protect critical load) + notify the shift leader;
  • Electrical-fire linkage level: auto-activate fire alarm + isolate the faulty circuit + double confirmation (on site + platform);
  • Cable-trench flooding linkage level: auto-start the drain pump + water-level alarm + warning linkage;
  • Manual review is mandatory after any linked action (an automatic action ≠ a solved problem; confirm on site);
  • All linkage actions are recorded (when, which device, which action) for later review.

Key insight: timeout escalation is the loop's safety net. One level up on response timeout, one level up on disposition timeout — alarms are never drowned out at night.

5. Using Closed-Loop Data for Review

Step five: what comes after closure. The data must be reviewed:

Review dimensions:

  • Alarm distribution: type/location/period (where the hotspots are);
  • Repeated alarms: repeat count at the same point (problems not cured);
  • Disposition duration: average alarm-to-closure time (response efficiency);
  • False-alarm rate: share confirmed as non-defect (threshold reasonableness).

Review methods:

  • Weekly/monthly reports: closure rate, average disposition time, repeated-alarm TOP 10;
  • Trend analysis: alarm-volume trend (rising = aging equipment / unreasonable thresholds);
  • Root-cause analysis: repeated-alarm causes (thresholds too strict? real equipment problems?);
  • Optimization iteration: threshold adjustment, responsible-person adjustment, process optimization.

Value of review:

  • Alarms turn from "noise" into "data assets" (hotspots = retrofit priorities);
  • Thresholds keep improving (false-alarm rate falls);
  • Processes keep improving (disposition efficiency rises);
  • Management decisions have a basis (equipment retrofit, staffing).

Review metrics:

  • Closure rate (target 95%+);
  • Average disposition duration (continuously falling);
  • Repeated-alarm rate (continuously falling);
  • False-alarm rate (target < 20%).

Alarm noise-reduction measures:

  • Threshold calibration: adjust per on-site baseline measurement (too strict = more false alarms);
  • Suppression rules: merge short-interval repeated alarms at the same point (prevent bombardment);
  • Condition filtering: reasonable fluctuations during load peaks / maintenance are not reported;
  • Grading noise reduction: attention/warning levels are only recorded, not disruptive (focus on alarm level);
  • Goal: bring alarm volume down to a level operators can actually handle.

Shift and handover management:

  • Handover: unclosed work orders are handed over face to face (no break on shift change);
  • Taking-over review: confirm the status of in-progress work orders (accepted/in disposition/pending closure);
  • Shift log: alarm-disposition situations recorded (handover has evidence);
  • Night-shift assurance: alarm-level response responsibility is clear (duty electrician + backup contact);
  • Holidays: backup mechanism (primary on leave, backup covers).

Pre/post comparison reference:

  • Before: 400+ alarms/month, 60% repeated, nothing closed;
  • After: repeated alarms down 80%, closure rate 95%+;
  • Disposition duration: from ~24 h average to ~6 h (escalation mechanism drives it);
  • False-alarm rate: from 40% to 15% (threshold calibration + suppression rules);
  • Key takeaway: operators shift from "overwhelmed" to "handling by level," management has evidence.

Key insight: review makes the loop run better and better. Alarm distribution, repeated alarms, disposition duration, false-alarm rate — four dimensions of review keep the loop improving.

6. Product Landing: Multi-Level Alarm Linkage Scheme

FEXLINK products bring the multi-level alarm linkage scheme to the field:

  • ESF Electrical Fire Monitor: temperature/leakage/electrical-fire monitoring, graded alarm reporting;
  • ESI Digital Status Monitor: switch-state/linkage-signal monitoring, linkage-disposition triggering;
  • Digital Electrical Intelligent Supervision System: platform alarm grading, auto dispatch, work-order closure, timeout escalation;
  • The platform outputs five-level alarms + the responsible-person matrix + full work-order trace;
  • Review reports auto-generated (closure rate, disposition duration, repeated alarms).

Design principle in one sentence: turn "alarm bombardment" into "orderly disposition closure." Grading, dispatch, closure, escalation, review — every alarm has a result.

7. Common Scheme-Design Mistakes

  1. Alarms not graded: all alarms at one level. Consequence: no priority; operators overwhelmed. Correct: five-level definition + response limits.
  2. No responsible-person matrix: alarms unclaimed. Consequence: recurring problems nobody manages. Correct: type → role → time-limit responsibility.
  3. Disposition without closure: no fill-back after handling. Consequence: records missing; problems "invisible." Correct: work-order closure with disposition traces.
  4. No escalation mechanism: nobody intervenes on timeout. Consequence: critical alarms drowned out at night. Correct: auto-escalate on response/disposition timeout.
  5. Closed-loop data unused: stored but never reviewed. Consequence: alarms remain noise; thresholds unoptimized. Correct: weekly/monthly review, continuous iteration.

Scope and Limitations

This article provides an overall framework for a multi-level alarm and linked-disposition scheme for distribution systems (graded thresholds, responsible-person matrix, work-order flow, timeout escalation, closed-loop review) for disposition-process design on distribution monitoring platforms. Graded thresholds and response time limits should be adjusted to on-site equipment, staffing, and shift systems; linked disposition (tripping, fire-fighting linkage, etc.) involves safety operations and must be performed per procedures by qualified personnel with approval; escalation rules must match the organizational structure. Online monitoring data is managed uniformly through the platform to form trend alerts. All cases are anonymized (a certain campus / plant) and do not refer to any specific customer.