{{product.tagline}}
{{product.description}}
-
FactorySight learns from the knowledge you already have.
Manuals, job cards, photos, voice notes and repair records become one knowledge base, organised by machine.
-
Every answer shows where it came from.
The manual page, the closest reference images and the machine's own repair history sit beside each answer, one tap away.
-
It runs inside the plant, on the phones your teams carry.
FactorySight is served from infrastructure you control, so manuals, photos and maintenance records stay on your network.
- Identify the machineScan its QR code, or enter its ID.
- Ask your wayType, speak or photograph the fault.
- Get the stepsNumbered guidance, read aloud if you want.
- See the sourceThe manual page and the matching reference images.
- Log the fixA drafted record of issue, cause and resolution.
- Build the historyEvery log informs the next answer on that machine.
Recording of FactorySight on a phone, with sound. The technician scans a demo unit, sends a photo and a question, receives cited steps, then saves the repair as a log that appears in the machine's history and dashboard. The demo unit stands in for a machine.
FactorySight is a memory for each machine, not a search box over manuals.
A manual explains how a machine should work. The people who maintain it know how it actually fails. FactorySight keeps both, and connects them to the machine in front of the technician.
- Job 1
Capture
It turns manuals, job cards, photos and voice notes into a knowledge base for each machine.
- Job 2
Guide
It answers a question asked by text, photo or voice with steps and the sources behind them.
- Job 3
Remember
It records each intervention as a structured log, so the next answer starts from it.
A fault becomes shared knowledge in six steps.
The same loop runs on every machine, and each pass through it leaves the knowledge base a little better.
Pick the machine.
Scan the QR code on the equipment or enter its ID, so every question and log is tied to that machine.
Describe the problem your way.
Type it, say it or photograph it. A photo and a question can go together.
Follow cited steps.
The answer gives numbered steps and points to the manual page, the closest reference images and earlier repairs on this machine.
Save what was done.
A record is drafted from the conversation: reported issue, root cause and resolution. The technician edits and saves it, or uploads a job card.
Start from there next time.
The log joins the machine's history and informs the next answer on the same fault.
See the pattern.
A dashboard for each machine shows events by type and period, with a searchable maintenance history.
There are three ways for know-how to reach the shop floor, and each has a different weakness.
This comparison is qualitative. The right route depends on how much of your know-how is written down and where your data is allowed to go.
| Route | What it knows | How an answer is checked | What happens after the fix |
|---|---|---|---|
| Manuals and experienced colleagues | What is printed, and what the people on shift remember. | By asking someone who has seen the fault before. | A paper job card, or nothing. The knowledge leaves with the person. |
| General AI assistant | Public information about similar machines, not your machines or their history. | It usually cannot be checked against your own documents. | Nothing is kept for the next shift, and data may leave your network. |
| FactorySight in your plant | Your manuals, your reference images and the repair history of each machine. | Against the cited manual page, the matching images and the earlier logs. | A structured log that feeds the next answer and the dashboard. |
What matters most depends on how much of your know-how is written down today.
Every answer comes with four kinds of evidence.
The technician sees why the assistant said what it said, and can check it before picking up a tool.
| Evidence | What it shows | Why it matters |
|---|---|---|
| Sources | The document and page the guidance was taken from. | The step can be read in the original manual. |
| Image analysis | The reference images that most closely match the photo. | The technician can confirm it is the same part and the same fault. |
| Maintenance history | Earlier interventions on this machine: issue, cause and resolution. | A repeat fault is recognised as one. |
| Conversation memory | What has already been asked and answered. | Follow-up questions do not start from zero. |
A cited answer can still be wrong.
The citation makes an answer checkable, not correct. For safety-critical steps, read the cited page.
The technician decides.
{P} recommends the next best action. A person carries it out and confirms the result.
FactorySight is a mobile assistant, a knowledge layer and a server inside your plant.
FactorySight has three layers. Each one can be inspected.
Chat for questions, Log for maintenance records, Dashboard for each machine's history, and Tools for recent equipment and connection status.
A question is matched against the machine's documents, reference images and earlier logs, and the answer is written from what is found.
The assistant, the knowledge base and the maintenance records sit on infrastructure you control. Technicians reach it from the devices they already use.
- Phone
- Tablet
- QR code
- Local server
Technicians fix the machine, and FactorySight keeps what they learn.
FactorySight finds, cites and records. People diagnose, repair and decide. The chart is an example of how maintenance knowledge moves today.
The maintenance lead decides which machines and documents are loaded first, and who may change them.
Technicians scan the machine, ask, follow the cited steps and save the log.
Senior experts add what the manuals leave out. Reliability engineers use the history to spot repeat faults.
Knowledge owners keep manuals current and review what has been logged.
