3D view of the measurement system

The AI measurement system

Your machines.
Now AI-ready.

Retrofit without touching the process. On-premise AI finds the use cases that pay off for you.

Industrial controller.

A compact industrial computer sits on the DIN rail in the control cabinet.

Fieldbus coupler.

Connects the measurement modules to the controller via EtherCAT.

Measurement modules as needed.

Plug-in modules capture your machine’s signals, combined to suit the application.

Right inside your machine.

The measurement system sits in the control cabinet. Attached non-invasively, without touching the controller.

Analysed by your local AI.

An AI server on site analyses the data and automatically derives use cases. No cloud.

Get ready for intelligent manufacturing.

One local AI server for all the machines in your production.

The challenge

Your machines have been running reliably for years. Yet most of them are not connected. The data that decides wear, quality and downtime is lost every day.

Avg. 9.9

years is the age of a machine in a machining company.1

1/3

of machines are connected.1 The rest deliver no data.

72%

of SMEs considering AI lack the know-how.2

  1. Boos et al. (2020): Tooling in Germany. WZL of RWTH Aachen University & Fraunhofer IPT.
  2. Destatis (2025): Reasons for not using AI, ICT survey, reporting year 2025.

Intelligent machines

What if your machines
became intelligent?

They predict their own maintenance. Companies that maintain predictively instead of reactively have far fewer unplanned stoppages.

Schematic view of one year: today, breakdowns happen unplanned. With AI, part of them become planned maintenance that takes place before the breakdown.
−52.7%
unplanned stoppages1
≈ €40,000
lower downtime costs per year2
  1. Thomas, D. S. & Weiss, B. A. (2021): Maintenance Costs and Advanced Maintenance Techniques in Manufacturing Machinery. IJPHM. doi:10.36001/ijphm.2021.v12i1.2883
  2. Worked example: 10 machines × 50 h unplanned downtime/year × €150/h, minus 52.7% (exactly €39,525).

Solution

Our solution
in three steps.

Our measurement system is attached to the machine without interfering and collects fine-grained process data. We develop the best use cases into a finished product.

  1. Capture

    Measurement system

    Attached non-invasively. First dataset in 6 to 8 weeks.

  2. Analyse

    Agentic AI

    Analyses the data on-premise and automatically derives use cases, such as wear prediction.

  3. Prove

    Study and prototype development

    Feasibility proven, then developed into a product.

Data sovereignty

Your data never
leaves the shop floor.

The entire AI analysis runs on your premises. No cloud, no upload, no loss of know-how.

Your premises Machine Measurement system Local AI Cloud Your premises Machine Measurement system Local AI Cloud

100 % on-premise

Measurement data, models and the language model run on hardware in your network.

GDPR by design

No transfer to third parties. 60% of SMEs considering AI cite data protection concerns as an obstacle.1

Your process know-how stays yours

Process parameters and insights never leave the plant and remain your competitive edge.

  1. Destatis (2025): Reasons for not using AI, ICT survey, reporting year 2025.

Comparison

Feasibility studies
SMEs can afford.

We automate what costs weeks of manual work in conventional projects.

Why we do it differently

  • 100% on-premise
  • Our own AI models
  • Fits any machine
  • Everything from one source
ComparisonConventional data science consultingOpti Solutions
Data accessIntegration with the controller, often invasiveNon-invasive retrofit on the existing cable
AnalysisManual work by data scientistsAutomated by an agentic AI pipeline
Data storageOften in the cloudFully on-premise
OutcomeReport with recommendationsFeasibility study, then development into a product

Use cases

What your data
can tell you.

Wear prediction

Detect tool and grinding wheel wear before it costs quality. Plan changes instead of reacting.

Process monitoring via acoustic emission

High-frequency acoustic emission signals reveal contact, engagement and irregularities in real time.

Anomaly detection

The AI learns your machine's normal behaviour and flags deviations long before an operator notices.

Quality prediction

Infer part quality from process signals and reduce inspection effort where it matters.

In the field

Acoustic emission on a deburring machine.

At an industrial partner, we placed the box between the existing sensor cable of an acoustic emission monitoring system. It reads the machine's start and stop signals and records every measurement automatically. A one-year endurance test confirmed gap-free recording and synchronisation.

Schematic illustration, not measured data

Sample rate in pilot
10 kS/s
Changes to the controller
0
Endurance test
1 year

Working together

How we work together.

  1. Feasibility study

    With our measurement system, we show where AI pays off.

  2. Implementation

    We turn the best use cases into products.

  3. Operation

    You rent our AI box including software, operation and replacement. We take care of everything.

Specs

Technical specifications

Status: prototype. The industrial version of the measurement system is planned for 2027.

Data capture

Measurement modules
Plug-in, combined to suit the application
Fieldbus
EtherCAT
Installation
Non-invasive, without touching the controller

Processing

Edge computer
Industrial controller on the DIN rail
AI analysis
Local AI server on site
AI runtime
Local language model, container architecture

Integration

Protocols
OPC UA, MQTT
Operation
Web dashboard via local Wi-Fi
Mounting
35 mm DIN rail (EN 60715)

About

Born in research.
Built for production.

Opti Solutions builds on ARTHUR, a distributed measurement system developed at Furtwangen University in the research project DQ-Meister and refined in PräziLoop. We are turning that experience into a product for mid-sized manufacturers.

M.Sc. Niels Schneider

Co-founder, AI and technology

Doctoral researcher at Furtwangen University (HFU). Researches explainable artificial intelligence.

M.Sc. Sinan Harkci

Co-founder, software and sales

Senior software developer for industrial software, experienced in process automation. pilot projects and sales.

Why us

Network
Lecture series and workshops, plus ongoing exchange with companies in the industry.
Research
Research and publications on applying industrial AI.
Practice
Tested on real machines and production data. First prototypes and pilot projects.

Let’s make your existing
machines AI-ready.

Local. Fast. Affordable.

We are looking for pilot customers from the region. Tell us about your machine and we will get back to you promptly with an initial assessment.