Harmony AI
MANUFACTURING · PACKAGING & PRINTING · GLASS CONTAINER MANUFACTURING

AI Automation for Glass Container Manufacturing.

Harmony digitizes furnace logbooks and gob weight sheets, streams the hot end live, ties defects to molds and sections, and generates COAs from live data. One AI-native operating system over everything: your floor, your machines, and the software you already run. No rip-and-replace.

THE PROCESS

First a digital transformation. Then an AI transformation.

FIG. 01 · THE TWO ARCSARC 1 · THE DIGITAL TRANSFORMATIONARC 2 · THE AI TRANSFORMATIONCLIPBOARDS & PAPEREXCEL & TRACKERSERP & SOFTWAREPLCS & MACHINESDIGITIZECONNECTUNIFYFORWARD-DEPLOYED ENGINEERSON YOUR FLOOR · IN PERSON · IN WEEKSHARMONY AI PLATFORMONE LIVE DATA LAYERERP · PLCS · SENSORS · SCALESQUALITY · INBOXES · DOCS · SPREADSHEETSNO MES? THIS BECOMES THE MESAI SEARCH · SOURCES CITEDAI AGENTS · HUMANS APPROVEAI PRODUCTION SCHEDULINGBACK-OFFICE AUTOMATIONS
ARC 1 · DIGITALTHE DIGITAL TRANSFORMATIONCLIPBOARDSEXCELERP + SOFTWAREMACHINES (PLCS)DIGITIZE · CONNECT · UNIFYONE LIVE DATA LAYEREVERY RECORD, SYSTEM, AND MACHINEBUILT ON-SITE, IN PERSON
ARC 2 · AITHE AI TRANSFORMATIONONE LIVE DATA LAYERTHE FOUNDATIONAI SEARCHANSWERS WITH SOURCES CITEDAI AGENTSPROPOSE · A PERSON APPROVESAI PRODUCTION SCHEDULINGTHE PLAN PROPOSES, PLANNER APPROVESBACK-OFFICE AI AUTOMATIONSFINANCE · SALES · PROCUREMENT · LOGISTICSAI ON TOP OF THE FOUNDATION
Data engineering is the work.AI is what it unlocks.
THE PHASED ROLLOUT

Three phases of a digital and AI transformation.

PHASE 01PAPEREXCELCAPTUREDONE LIVEDATALAYEREVERY PAPER POINT DIGITIZED · MACHINES CONNECTED
PHASE 1

Lay the Data Foundation · Digitization

Every pen-and-paper record digitized at the station, every software system connected, and all of the data unified into one live layer. The digital transformation starts here.

PHASE 02SENSORS LIVEAI SCHEDULE BOARDPM BEFORE FAILUREPROACTIVE OPERATIONS AT SCALE
PHASE 2

Production & Operations Scale

Factory operations turn proactive: live sensors and machine data, the AI scheduling board, predictive maintenance before failure.

PHASE 03LIVE DATALAYERAGENTAGENTAGENTHUMANS APPROVEAGENTS ACT ON THE LIVE LAYER
PHASE 3

AI-Native Operations

Agents across the floor and the back office act on the live layer: quality signals, reports, copilots. Humans approve.

AI USE CASES

Twelve ways the AI layer earns its keep.

PAPERSTATION TABLETLIVECAPTURED ONCE · SAME SHIFT
U1

Paperwork Digitization

Use case 01 · PEN, PAPER & EXCEL

Problem. Furnace logbooks, gob weight sheets, cold-end defect tallies, and mold change cards, filled in by hand and keyed into Excel a day later.

Harmony AI Solution. Captured once, where they happen, on tablets at the furnace deck, the machine floor, and the cold end. AI flags an out-of-range entry the moment it is written, not a shift later.

ANY PLCLIVEONE SCREENSTREAMED IN REAL TIME
U2

Live Machine Data

Use case 02 · LIVE MACHINE DATA

Problem. Zone temps live on the furnace HMI, section timing lives in the IS machine, the lehr curve lives in a logbook. Nobody sees them next to the defect counts from the same hour.

Harmony AI Solution. Direct PLC and sensor connections, Allen-Bradley/Rockwell, Siemens, or any other brand, via OPC UA or whatever the machine speaks, streamed in real time.

01020304DEFECT FOUNDSIGNAL SENT UPSTREAM, SAME SHIFTFIXED AT THE START OF THE LINE
U3

In-Line Quality Intelligence

Use case 03 · IN-LINE QUALITY

Problem. A defect spike shows up at the cold end hours after forming. The ware in the lehr already carries the same fault, and the mold and section behind it were never written down next to the count.

Harmony AI Solution. Checks and defect counts captured at the cold end, tied to mold, section, and gob weight.

ASK ANYTHINGAI SEARCHALL SYSTEMSANSWER + SOURCES CITEDSECONDS, NOT PHONE CALLS
U4

AI Search Across All Data

Use case 04 · AI SEARCH

Problem. "Which molds ran that batch?" or "What did the furnace record before the defect spike?" means the logbook shelf, a historian query, and three phone calls.

Harmony AI Solution. Answered in seconds with the source shown, across the ERP, the historian, the digitized logbooks, and dispatch records. Idle floor TVs become live AI line boards.

THE PLAN PROPOSESPLANNER APPROVESCONSTRAINTS MODELED, NEVER GUESSED
U5

AI Production Scheduling

Use case 05 · AI SCHEDULING

Problem. Job changes sequenced by hand against furnace pull, mold availability, lehr capacity, and order due dates, rebuilt every time an order moves.

Harmony AI Solution. AI production scheduling models the planner's own constraints and proposes the sequence. The planner stays in charge and approves.

MATERIAL INLOSSLIVE, BY LOTA BAD LOT SHOWS IN AN HOUR
U6

Yield Optimization

Use case 06 · YIELD OPTIMIZATION

Problem. Gob weights sit on paper sheets and pack-to-melt gets computed at month end, long after the glass that missed the ratio went to cullet.

Harmony AI Solution. Gob weights stream live. Pack-to-melt ratio by shift and by mold, so a drifting section shows in an hour, not in the monthly report.

STATIONREPAIRRFID TAGLAST KNOWN LOCATION · GENEALOGY
U7

RFID + Serial Genealogy

Use case 07 · MOLD GENEALOGY

Problem. Mold life and repairs live on cards in the mold shop, and when a customer calls about a bad lot, the genealogy gets rebuilt by hand from logbooks.

Harmony AI Solution. Mold life, repairs, and defect history tracked in one place. Batch to forming to pallet, one-up and one-back in seconds instead of days.

SCRAPTRIMEVERY LOSS, SEEN WHERE IT HAPPENS
U8

Energy

Use case 08 · ENERGY

Problem. Furnace and compressor energy shows up on the utility bill weeks later, as one number nobody can act on.

Harmony AI Solution. Furnace and compressed-air energy metered live, per ton of glass melted. ML finds the setpoint patterns that melt the same glass for less.

CAPTUREDUNIFIEDACTED ONONE LIVE LAYER, HUMANS DECIDE
U9

AI Workflow Automation

Use case 09 · JOB CHANGE ANALYTICS

Problem. Changeover length lives in memory. Nobody can say what a job change really costs, or why the same change runs an hour longer on nights.

Harmony AI Solution. Every mold and job change timed live, by job, section, and crew. The plant learns which changes run long and why, and the schedule starts planning with real numbers.

SYSTEM40FLOOR12TRUESTOCKAI AGENTPOSUPPLIERPO DRAFTED AUTOMATICALLY · A PERSON APPROVES
U10

Inventory + Shortage Intelligence

Use case 10 · INVENTORY

Problem. Batch material counts drift, molds sit unlogged in the repair shop, and a packaging stockout gets discovered at the palletizer.

Harmony AI Solution. An AI-maintained true-stock layer for batch materials, molds, and packaging. Material-readiness before a job change releases; count drift flagged before it stops a line.

STATION 01STATION 02STATION 03DEFECT FOUNDSIGNALUPSTREAM
U11

Quality Records

Use case 11 · QUALITY RECORDS

Problem. COAs and customer quality records assembled by hand from logbooks, Excel, and the cold-end tallies, hours per lot, every lot.

Harmony AI Solution. COAs and quality records generated from the live data the floor already captured. An audit request is answered in minutes, with sources shown.

INVOICESPOSREPORTSAI AGENT, WORKINGAPPROVEDFINANCE · SALES · PROCUREMENT · LOGISTICS
U12

Back-Office AI Automations

Use case 12 · FULL-SCOPE AI AGENTS · FINANCE, SALES, PROCUREMENT, LOGISTICS

Problem. Dispatch paperwork, invoicing, AR follow-ups, and the daily production report eat office hours every single day.

Harmony AI Solution. Full-scope AI agents run the workflows across financial, sales, procurement, and logistics: deductions and AR worked automatically, POs parsed from the inbox, and reports written before the morning meeting.

Scope. Financials, sales, procurement, and logistics automations, in the same platform.

01 / 12
YOUR LIKELY STACK

The systems and machines we meet on your floor.

SYSTEMS
SAPOracleHomegrown systemsAVEVA / PI historiansIgnitionExcelCustom appsLocal servers
MACHINES
Batch housesFurnacesIS forming machinesAnnealing lehrsCold-end check stationsPalletizers

You might run any of these. Or none of them. Harmony is software and hardware agnostic, and we connect live machine data from any PLC brand and any software or custom app.

Every agent proposes. A person approves. That line never moves.

AI AGENTS

Agents that watch, flag, and act. Humans decide.

Gob Weight Drift Watch
Cold-End Defect Cluster Agent
Furnace Zone Watch
Mold Life Tracker
Job Change Timer
Pack-to-Melt Agent
Dispatch Paperwork Agent
AR + Collections Copilot
Daily Report Writer
Custom Agents, built for your floor and back office
THE PATTERN WE SEE

The problems we see on every glass container manufacturing floor, and why they happen.

The ERP, the historian, and the floor never connect

SAP or Oracle holds the orders, the PI historian holds the furnace tags, Ignition runs the screens, and the logbooks hold everything else.

Furnace logs and gob weights are written by hand

The furnace logbook, gob weight sheets, cold-end defect tallies, and mold change cards are filled in on paper.

Defects are counted at the cold end, hours after forming

Ware spends hours in the lehr between forming and the cold end.

The furnace runs on a few operators' judgment

The operator who can read a zone temp curve and know the melt is going sour is retiring, and the IS mechanic who knows which section drifts after a mold change has no.

The back office runs on manual work

POs are typed in from emails, invoices keyed into the ERP, deductions and AR worked from spreadsheets, and reports built by hand before the morning meeting.

ON-SITE, IN PERSON · SOFTWARE + HARDWARE AGNOSTIC · READ-ONLY PILOTS · US-HOSTED OR ON-PREM · HUMANS APPROVE EVERY ACTION

THE PILOT

$15–20K. 4–6 weeks. One painful problem.

We pick one specific pain point, send engineers to live in your factory for 4–6 weeks, and solve it. One-time payment, working software by week 3, and no obligation to go further. If you do, you choose the modules that you like.

FIRST CALLSECOND CALLCHOOSE PILOTDATA ACCESS FROM ITON-SITEBUILDTESTROIREVIEW
QUESTIONS, ANSWERED
Our IS machines and furnace controls are decades old. Can you really get live data off them?

Yes. Direct PLC and sensor connections, Allen-Bradley/Rockwell, Siemens, Omron, Mitsubishi, or any other brand, via OPC UA or whatever the machine speaks. Where a controller predates all of that, we add sensors and meters at cost. Software and hardware agnostic: if the furnace, the IS machine, or the lehr has a signal, it streams into the AI layer in real time.

We already run a PI historian and Ignition. What does Harmony add?

Historians collect tags; they do not connect them to anything. The AI layer sits on top of PI and Ignition and joins the furnace data to the cold-end defect counts, the mold history, and the orders in the ERP, then puts AI search, ML pattern detection, and agents over all of it. Additive by default. The AI layer sits on top of what you already run, and when a system deserves to be replaced, we replace it.

Do you replace our ERP or MES?

Sometimes we're the layer on top of your MES. Sometimes we are the MES. Where there is no MES, or a fragmented one, the Harmony platform becomes the plant's MES: station capture, live line boards, quality, genealogy, and scheduling on one AI-native layer. The ERP stays.

How does the AI get our data?

Read-only connections, scheduled exports, or on-prem. Security is reviewed together in scoping. Read-only pilots. Every agent proposes. A person approves.

Do you work remotely?

No. We always start a pilot in person and on-site, with forward-deployed engineers on your floor. Very rarely do we do anything remotely.

What does it cost after the pilot?

A per-plant annual platform: unlimited seats, no user fees, hardware at cost. The pilot itself runs 4-6 weeks against one painful problem, with working software by week 3. Baseline first, then measure. Estimates are labeled as estimates.

Harmony AI puts the batch house, the furnace, and the cold end on one live layer, so the hot end hears about a defect while it can still act.