A-FMDS: System Overview
A 10,000-foot view of ProAlert's closed-loop daily management system
A 10,000-foot view of ProAlert's closed-loop daily management system
A-FMDS (the Automated Floor Management Daily System) is ProAlert Core's daily management spine: the connective tissue that turns a collection of strong but separate modules (telemetry, CMMS (Computerized Maintenance Management System), scrap tracking, RCA (Root Cause Analysis), reporting) into one closed-loop execution system. It answers a single question, continuously and automatically:
"Where are we missing the mark right now, and who is doing what, by when, to fix it?"
The name reads exactly as it works. FMDS (Floor Management Daily System) is the established lean-manufacturing discipline: the daily rhythm a plant uses to review targets, surface the gaps against them, and drive each gap to closure with clear ownership. The leading "A" is Automated. That is what ProAlert adds. ProAlert already sees the floor in real time, so it can detect the gap, raise the problem, route it, escalate it, and capture the learning without anyone re-keying data into a whiteboard.
Gap → Problem → Action flow that
every detection source feeds and every screen consumesA scrap-rate breach, a sensor alarm, a missed schedule, a failed inspection, a manually-raised concern: all of them become the same kind of record in A-FMDS. That is the whole point. One place to look, one lifecycle to learn, one audit trail to trust.
FMDS works. The discipline is proven across lean manufacturers worldwide. The gap is not in the methodology; it is in the execution. FMDS without automation depends entirely on the right people doing the right things, at the right time, every shift. When a strong lean leader is present and engaged, the board reflects reality. When that person rotates out, goes on vacation, or moves to another plant, the discipline erodes and the board becomes theatre.
The honest reason most FMDS programs fade is not laziness or bad intent. It is that they ask people to do extra work on top of everything else they are responsible for. Anything extra gets squeezed out when the floor gets busy. Three structural gaps make this predictable:
| The gap | What breaks without automation | How A-FMDS closes it |
|---|---|---|
| No single source of targets | Targets live in spreadsheets, whiteboards, and tribal knowledge. "What good looks like" varies by shift and by supervisor. Nobody agrees on the baseline. | One Target entity covers every scope and metric,
with revisions, effective dates, and approval. Every screen in
the system uses the same number. |
| Gaps surface only when someone raises them | The morning board reflects what the outgoing supervisor remembered to write down. Events that happened overnight when no one was watching go unreported. | Every detector in ProAlert automatically produces a
Gap record the moment an actual misses its target.
Nothing surfaces only when someone chooses to surface it. |
| Ownership and follow-through are voluntary | Problems scatter across call logs, NCRs (Nonconformance Records), work orders, and verbal hand-offs. Most issues age quietly until the next shift meeting when someone asks again. | One Problem with an enforced lifecycle, one
configurable escalation engine, and a permanent audit trail.
The system chases it so the supervisor does not have to. |
FMDS initiatives fail not because the discipline is wrong but because they rely on people to do it consistently, without error, on top of everything else they are responsible for. Any process that depends on constant human discipline eventually loses to a busy shift. A-FMDS makes the discipline automatic. The system holds regardless of who is on the floor that day.
A-FMDS is a loop, not a list. Every operational issue travels the same path from "we missed" to "we learned," and the learning feeds the next cycle.
Every Gap, Problem, Action, and Escalation carries an append-only event stream: a permanent, ordered record of everything that happened to it. That stream is the audit trail, the basis of reporting, and the fuel for future analytics. Records are written once and never quietly edited.
A-FMDS introduces a small set of new entities. Each is auditable, each carries an event stream, and each is deliberately generic so any source can feed it and any screen can consume it.
| Entity | Purpose | Key fields |
|---|---|---|
| Target | What good looks like for a metric at a scope. Versioned and approved. | Metric, direction, target / warning / critical thresholds, scope, revision |
| Gap | A recorded instance of an actual missing a target. | Gap number, metric, expected, actual, severity, source, status |
| Problem | An operational issue someone is accountable for resolving. | Problem number (PRB-…), category, severity,
priority, owner, status, due date |
| Action | A concrete step taken to resolve a Problem (or other parent). | Action number (ACT-…), type, owner, status,
due date, verification flag |
| Escalation Rule | A timed ladder of "who to notify when this sits too long." | Applies-to type, filters, levels (delay, notify target, channel) |
| Coaching Note F4 | Structured supervisor coaching tied to a problem or person. | Coachee, anchor, structured prompts |
| LSW (Leader Standard Work) Routine F4 | A recurring checklist a leader executes. | Items, schedule, execution records |
| Tiered Meeting F4 | Tier 1 / 2 / 3 huddle with an auto-built agenda from open gaps. | Tier, agenda items, attendees, action links |
A Gap is a measurement: "scrap was 5%, target 2%." A Problem is accountability: "this is mine to fix." Not every gap becomes a problem; a gap is promoted when it needs an owner. One problem can also roll up many recurring gaps.
A Problem is the issue and its owner. An Action is one executable step. A problem usually needs several actions: contain it now, find the root cause, fix it permanently, verify the fix held. Each is separately owned and tracked.
A-FMDS organises everything by category: the lean SQDCP (Safety, Quality, Delivery, Cost, People) framework that a daily management board is built around. Every Problem is filed under one of these, so a board can show one row per category and a plant can run its huddle top to bottom.
A-FMDS adds two more categories beyond classic SQDCP, because ProAlert tracks them too: Maintenance (equipment health, work orders) and Kaizen (continuous-improvement ideas). Together they are the complete set of problem categories.
The A-FMDS Daily Board is the screen a plant stands in front of each morning. Today it shows the live stream of recent gaps and problems; the full SQDCP grid (one cell per category per shift, click a red cell to drill straight to the problem) arrives in Phase F3.
A-FMDS was designed to wrap and link, never to rip out. Existing modules keep doing their jobs; A-FMDS connects them.
| Existing ProAlert capability | Relationship to A-FMDS |
|---|---|
| Scrap threshold & sensor violations | Wrapped: each violation now also raises a Gap. The old tables and approval flows stay. |
| Schedule discrepancies | Wrapped: become Gaps; their existing escalation level maps onto Gap severity. |
| Nonconformance register (NCR / ISO 9001) | Linked: an NCR can originate or specialise a Problem; the register stays the system of record for quality. |
| Work orders (CMMS) | Linked: a work order is an Action's execution sink; closing the WO can complete the Action. |
| Root-cause analysis & Kaizen | Linked: a Problem can open an RCA or a Kaizen idea; their countermeasures surface as Actions. |
| Andon calls & safety incidents | Linked: long downtime calls and safety incidents can promote to a Problem. |
| SignalR real-time hub & daily metrics | Reused: A-FMDS rides the existing hub for live updates; no new infrastructure. |
A-FMDS produces clean, typed event streams but does not analyse them itself. A separate, pluggable intelligence layer will eventually consume those streams to rank recurring problems and suggest fixes. A-FMDS defines the seam; the analytics plug in without changing the spine.
A-FMDS is delivered in five phases, F1 through F5. Each phase is independently useful; the loop gets more complete as the phases land.
| Phase | Scope |
|---|---|
| F1 | Event → Gap → Problem → live board update: the minimum viable loop. Targets, Gap detection, the Daily Board, and gap promotion. |
| F2 | Full Problem lifecycle, the generic Action, the configurable escalation engine, and the Problems, Actions, and Escalations management screens. |
| F3 | The full SQDCP Daily Board with TV display mode, drill-down, an Andon overlay, and accountability reporting and export. |
| F4 | Coaching notes, Leader Standard Work (LSW) routines, and Tiered Meetings with auto-built agendas. |
| F5 | Knowledge base, recurrence-driven learning, mobile push notifications, mobile pages, and security tightening. |
| Term | Meaning |
|---|---|
| A-FMDS | Automated Floor Management Daily System: ProAlert's closed-loop daily management spine, described in this document. |
| FMDS | Floor Management Daily System: the lean-manufacturing discipline A-FMDS automates. |
| Target | The defined "good" value for a metric at a scope. |
| Gap | A recorded instance of an actual value missing its target. |
| Problem | An owned operational issue, numbered
PRB-YYYYMMDD-NNNN. |
| Action | An executable step against a problem,
numbered ACT-YYYYMMDD-NNNN. |
| Promote | Converting a Gap into a Problem so it gets an owner. |
| SQDCP | Safety, Quality, Delivery, Cost, People: the daily-management category framework. |
| Escalation | The timed, automatic raising of a stalled problem up the chain of command. |
| Recurrence key | A stable fingerprint that lets A-FMDS recognise "the same kind of problem" happening again. |
| LSW | Leader Standard Work: a recurring checklist a leader is expected to perform (Phase F4). |
This document is the concept. To see how people actually use A-FMDS day to day (the Board, Problems, Actions, and Escalations screens), read the A-FMDS User Guide.