Tool 01

Flowstate Atlas

Map a process. Get the engineering. Shaped to your factory.

How the walkthrough works

DEMANDCONSTRAINTTAKTP1InputP2ProcessP3OutputWIPWIPPCE37%14 SECTIONS9/14

An engineer walks through 14 observation sections. The inference engine calculates efficiency, constraint signals, cost of loss, and generates the improvement plan automatically.

What the engine produces

Process Cycle Efficiency

VA time as % of total lead time

Capacity vs Demand Gap

Units of constraint headroom

Cost of Loss

Rand-quantified across 6 cost lanes

Improvement Plan

Prioritised matrix with ROI per action

Tracking

Log key metrics per shift. Dashboards show trends, targets vs actuals, and whether improvements are holding.

Analysis layer

Every figure shows its formula, inputs, and confidence rating. An AI layer generates diagnostic narratives, root cause analysis, and an executive briefing. A chat interface answers questions about the specific process.

NO AI ESTIMATES. ENGINEERING FORMULAS ONLY.

Atlas Services

No two factories run the same Atlas.

Every account is unique. Three services shape Atlas to your operation — from a single process diagnosis through to a fully commissioned factory floor.

SYS.DIAGNOSEFrom R1,950

01

Process Health Check

Know exactly what your factory costs you.

A Flowstate engineer walks your floor through a 14-section observation. Every process timed, every loss calculated in Rand. You receive a full engineering report with cycle efficiency, constraint, cost of loss, and a prioritised improvement plan within five working days.

  • 14-section engineering walkthrough

  • Cost of loss in Rand per process

  • Constraint and takt analysis

  • Prioritised improvement plan with ROI

SYS.COMMISSIONQuoted

02

Atlas Commissioning

Your whole factory in Atlas. Every process. Running.

A full end-to-end commissioning engagement. Every process on your floor mapped, costed and live in your Atlas account. Your team trained until they can find your biggest loss in three clicks and log a shift unaided. Quoted per factory.

  • Full factory mapped with measured cycle times

  • Cost of loss in Rand across every process

  • Product lines, targets and shift tracking live

  • Team trained to owner and supervisor level

SYS.BUILDFrom R3,000

03

Custom Build

New functionality engineered into your Atlas account.

Need a job tracker, a planning board, a quality capture module, or a customer report that runs itself? A Flowstate engineer scopes and builds exactly what your operation needs. Fixed price, written acceptance criteria, delivered into your own Atlas account.

  • Job trackers, planning boards, quality capture

  • Custom reports and integrations

  • Fixed price, written brief, defects free

  • You own it, maintained under Custom Care

How a build works

01

Consult

02

Specify

03

Build

04

Review & Refine

05

Run

Fixed price. Written brief. Defects fixed free.

Tool 02

Automated Data Collection & Live Monitoring

See your factory in real time.

Data pipeline architecture

MACHINESGATEWAYCLOUDDASHBOARDSPLCSCADASENSORERPEDGEGATEWAYSECURESTORAGELIVEOPERATORINPUT

Sensors, PLCs, SCADA and ERP feed a single normalised data store. Operators add downtime reasons via tablet. All streams converge in live dashboards.

Dashboard views

OEE

Availability × Performance × Quality

Downtime Pareto

Ranked by total loss time

Output vs Target

Shift-by-shift actuals vs plan

Maintenance Signals

Performance trend alerts

Why it matters

The data collected here does two things: it validates what the Process Mapper identified during the walkthrough, and it tracks whether improvement projects are holding across shifts over time. You own the data. The system is built to remain usable after handover.

Tool 03

On-Site Engineering

The work the tools can't do remotely.

Improvement project structure

OUTPUTINTERVENTION+72% OUTPUTBEFOREAFTERBASELINE

Every project is baselined before any change. Improvement is implemented on the floor and measured in the same data system. The Process Mapper is updated to reflect the new operating state.

What a project looks like

01

Baseline captured

Before any change is made

02

Root cause confirmed

Machine data, not assumption

03

Improvement implemented

On the floor by a Flowstate engineer

04

Before/after measured

In the same data system

05

Process map updated

New operating state becomes the baseline