traktor manufacturing

Tractor Inspection

Component Detection and OCR

Tractor Inspection

Maddox AI ensures that all relevant parts of a customer-specific order have been correctly mounted on the customer’s tractor.

Product

Tractors

Industry

Off-Highway

Total savings / year​

65-440k €

Customer Request

Our customer produces several 1,000 tractors of different sizes and configurations each year. The tractors are finally inspected in a large light tunnel. Two specialists per shift check whether the correct configurations such as lights, rear view mirrors, hazard warnings, etc. have been installed for the respective customer, as incorrectly installed components can quickly become expensive in after-sales.

The Maddox AI inspection system supplements this inspection work and thus supports the technical experts during the inspection. All tractor types, large and small, can be photographed with the same static-mounted camera and are evaluated by the AI model. By also reading the order number for each tractor, Maddox AI enables full traceability of individual tractor orders.

Before

Previous Inspection System

After

With Maddox AI

The tractors were checked manually in two shifts by two employees. The detection accuracy varied depending on the inspector.

Relevant parts are automatically checked according to a precise defect book definition.

Because tractors are made order-specific, the plant received a larger number of customer complaints that required rework.

The number of customer complaints was significantly reduced with the help of the Maddox AI inspection solution.

It was not possible to perform a structured root cause analysis to reduce mounting errors as the current inspection results were not consistently saved digitally.

Thanks to the consistent and digitized inspection process, the expensive re-work rate can be reduced in the future.

Before

Previous Inspection System

The tractors were checked manually in two shifts by two employees. The detection accuracy varied depending on the inspector.

Because tractors are made order-specific, the plant received a larger number of customer complaints that required rework.

It was not possible to perform a structured root cause analysis to reduce mounting errors as the current inspection results were not consistently saved digitally.

After

With Maddox AI

Relevant parts are automatically checked according to a precise defect book definition.

The number of customer complaints was significantly reduced with the help of the Maddox AI inspection solution.

Thanks to the consistent and digitized inspection process, the expensive re-work rate can be reduced in the future.

100%

Tracability by reading out the production order via OCR

MES Integration

Enables use of data for process improvements

1 Tractor / Min.

Inspection speed

Would you like to learn more about our solution? – Test for free now!

Label Recognition on Cardboard Boxes​

Thanks to Maddox AI, the codes and labels on different cartons are reliably recognized, read, matched as well as rejected and re-evaluated in case of discrepancy.

Inspection of Ball Bearings

Maddox AI successfully automates the inspection of impact points, incorrectly mounted components and scratches on ball bearings.

Tractor Inspection

Maddox AI ensures that all relevant parts of a customer-specific order have been correctly mounted on the customer’s tractor.

Inspection of Printed Circuit Boards

Surface Inspection, Component Recognition and Data Matrix Code Recognition

Inspection of Printed Circuit Boards

Thanks to Maddox AI’s device and software, incorrectly assembled components and other damages can be detected and tracked to a specific PCB.

Product

Printed Circuit Boards

Industry

Electronics

Total savings / year​

110-160k €

Customer Request

Our customer manufactures printed circuit boards, which were manually inspected by two people in a three-shift operation due to the variety and complexity of the production. A classic inspection system was already used to check individual components – but a hundred percent automated end-of-line inspection did not happen, yet.

Maddox AI has developed a semi-automatic inspection system in which printed circuit boards are inspected with a high-resolution camera. Due to the high product variability, a fully automated inspection system was not possible. The system is used to detect, among other things, defectively mounted components and other damages such as scratches, or solder bumps. By reading a data matrix code, the Maddox AI system tracks defects to a specific component. In the first step, the system is loaded manually. New defect classes have already been retrained several times free of charge.

Before

Previous Inspection System

After

With Maddox AI

The printed circuit boards were inspected manually in three shifts with a total of six employees. The detection accuracy varied depending on the inspector.

All products are inspected 100% automatically according to a precise defect definition. The inspection system is operated by one employee per shift.

Due to the faulty assembly of printed circuit boards, the production was carried out with an excessively high reject rate.

Since installing Maddox AI, the scrap rate has been reduced by 25%.

Since the inspection results were not digitized, no regular structured root cause analysis was performed.

Thanks to the digitalized inspection process, process optimizations could be realized (e.g. direct feedback loop for machine setting).

Before

Previous Inspection System

The printed circuit boards were inspected manually in three shifts with a total of six employees. The detection accuracy varied depending on the inspector.

Due to the faulty assembly of printed circuit boards, the production was carried out with an excessively high reject rate.

Since the inspection results were not digitized, no regular structured root cause analysis was performed.

After

With Maddox AI

All products are inspected 100% automatically according to a precise defect definition. The inspection system is operated by one employee per shift.

Since installing Maddox AI, the scrap rate has been reduced by 25%.

Thanks to the digitalized inspection process, process optimizations could be realized (e.g. direct feedback loop for machine setting).

0€ costs

For expansion to further product variants

100%

Tracability by reading the data matrix code

40%

less pseudo scrap

Would you like to learn more about our solution? – Test for free now!

Cable Inspection

Maddox AI enables the efficient and reliable automation of high-voltage cable inspections. It accurately detects and classifies surface defects such as holes, bubbles, weld marks, and other irregularities.

Inspection of Ball Bearings

Maddox AI successfully automates the inspection of impact points, incorrectly mounted components and scratches on ball bearings.

Optimization of Brake Hose inspection

Maddox AI can efficiently and reliably automate the inspection of surface defects, such as scratches, burns, or stripping cuts.

Inspection of Ball Bearings

Surface Inspection and Component Detection

Inspection of
Ball Bearings

Maddox AI successfully automates the inspection of impact points, incorrectly mounted components and scratches on ball bearings.

Product

Ball Bearings

Industry

Metal

Total savings / year​

190-350k €

Customer Request

Our customer produces over 20,000 ball bearings daily, which were inspected both by a classic inspection system and, in some cases, by human inspectors. The classic inspection system often mistook color differences for real defects. This resulted in an excessively high pseudo scrap rate, which had to be checked manually at a high cost.

To massively reduce the pseudo scrap rate, the Maddox AI inspection system combines existing camera hardware with the Maddox AI software. Classic defect types include damages such as impact points, incorrectly mounted components or scratches.

Before

Previous Inspection System

After

With Maddox AI

A classic rule-based system was used for the initial inspection. This was followed by a manual inspection in three shifts by two employees per shift to recheck rejects and pseudo-rejects.

All products are inspected 100% automatically according to a precise defect definition. This made it possible to reduce pseudo scrap by more than 90%, eliminating the need for manual rechecking.

The plant received two relevant customer complaints in the last two years, which had to be manually rechecked.

No further customer complaints have been received since the installation of Maddox AI.

Since the inspection results were not digitized, no regular structured root cause analysis was performed.

Thanks to the digitalized inspection process, process optimizations could be realized (e.g. direct feedback loop for machine setting).

Before

Previous Inspection System

A classic rule-based system was used for the initial inspection. This was followed by a manual inspection in three shifts by two employees per shift to recheck rejects and pseudo-rejects.

The plant received two relevant customer complaints in the last two years, which had to be manually rechecked.

Since the inspection results were not digitized, no regular structured root cause analysis was performed.

After

With Maddox AI

All products are inspected 100% automatically according to a precise defect definition. This made it possible to reduce pseudo scrap by more than 90%, eliminating the need for manual rechecking.

No further customer complaints have been received since the installation of Maddox AI.

Thanks to the digitalized inspection process, process optimizations could be realized (e.g. direct feedback loop for machine setting).

98%

Time Saving for Recalibrations

> 90%

Less Pseudo Scrap

30 parts / min.

Inspection Speed

Would you like to learn more about our solution? – Test for free now!

Inspection of Injection Molded Parts

Surface defects such as improper cutting edges, cracks, scratches, and contamination can be reliably and automatically identified on injection-molded parts with the help of Maddox AI.

Label Recognition on Cardboard Boxes​

Thanks to Maddox AI, the codes and labels on different cartons are reliably recognized, read, matched as well as rejected and re-evaluated in case of discrepancy.

Fully automated Wire Inspection

Surface defects such as dents, burnt spots, or insulation faults can be reliably and automatically checked using Maddox AI.
Maddox AI logo