Made in logo

Documentation

Getting Started

Learn how to set up and deploy Made OS in your facility.

This guide walks you through the steps to get your first line connected and receiving actionable intelligence.

Prerequisites

Before starting, ensure the following are in place:

  • An active MadeOS account (create an account and request access at app.madeos.ai)
  • Facility network access (VLAN or dedicated connection)
  • PLCs or sensors with standard industrial protocol OPC UA
  • At least one industrial PC or server on-site (minimum 6 cores, 8 GB RAM, 200 GB storage)
  • Google Chrome or Mozilla Firefox (latest version) for the dashboard

Deployment Options

MadeOS supports two deployment paths. Choose the one that matches your infrastructure:

FeatureEdge DeviceSoftware + API
InstallationOn-site by Made OS teamSelf-install on your PC or server
AI ProcessingLocal inference on NVIDIA JetsonManaged cloud inference
LatencySub-10ms localCloud round-trip
HardwareUp to 67 TOPS NVIDIA inferenceYour existing industrial PC
Setup Time1–2 days per lineUnder 1 hour

Step 1 — Create Account and Request Access

Create an account at app.madeos.ai - sign up:

signup

After creating your account, you will need to confirm your email before accessing the app.

Once your email is confirmed, you can log in to the app and request access to the MadeOS platform. For security reasons, you will need to book a call with our team to verify your identity and get access.

request access

Step 2 — API client setup

For the API deployment path, you will need to set up an API client that can connect to the MadeOS cloud.

First, download the MadeIO API client software for your preferred OS:

For Windows:

Once downloaded, run the installer and the software will be installed on your system.

You will find the application in the Start menu or on the desktop.

app in windows

Open the application and you will see the following to connect to MadeOS API:

madeio init windows

For Ubuntu/Debian:

Once downloaded, you will need to go to the directory using the terminal where the .deb file is located and run the following command:

sudo dpkg -i filename.deb

After the installation is complete, you can find the application in the applications menu.

madeio in ubuntu

Open the application and you will see the following to connect to MadeOS API:

madeio init ubuntu

Step 3 — Connect PLCs and Machines to MadeOS

Once access is confirmed by the MadeOS team, the next step is to configure your Production Line with your PLCs and Machines in the MadeOS platform. The connection is established read-only — MadeOS does not write to your control systems.

Before you begin, you will need to:

  1. Configure the OPC UA server on your PLC/Machine, ensuring it is accessible with the following settings:
    • Host: opc.tcp://{ip}
    • Port: 4840 or your custom port
    • Security: None (or appropriate security settings)
    • Access control: Allow connections from MadeOS platform IP
  2. Connect PLCs/Machines and the computer that is running the MadeIO software to the same network via Ethernet-RJ-45.

First Production Line

Log in to the MadeOS platform after getting access.

Create a Plant and add your first Production Line.

Create plant:

madeos plant

madeos plant form

Create production line:

madeos pl form

madeos pl created

Once the Production Line is created, copy the IO Controller Key from the production line details, paste it into the MadeIO software and click “Connect”.

madeos pl details

madeio running

Once the connection is established, the MadeIO software will try to connect to every device in your plant and production line, sync devices and send real-time data to the MadeOS platform.

Finally, you will see the in the Devices Resources dialog that your plant and production line are now connected.

madeio devices

Now, you are ready to configure your Digital Twin.

Step 4 — Initial Configuration

After the data connection is established, you will need to configure your Digital Twin:

  1. Establish baseline normal behavior (nominal ranges per shift and product type)
  2. Define alert thresholds and risk prioritization rules
  3. Configure operator notification channels (web dashboard, WhatsApp, or email)
  4. Set up user accounts and role-based access

Step 5 — Go Live

On day one of live operation you will begin receiving:

  • Anomaly alerts when operational drift is detected
  • Risk-ranked event queue prioritizing intervention by impact and window
  • Prescriptive actions — what to adjust, by how much, and when

The model accuracy improves over the first 60 days as it learns your facility’s specific behavior.

Next Steps