cell contacting system

Inspection of E-Modules

Surface Inspection

Inspection of E-Modules

Maddox AI enables the reliable and automated identification of defects such as sheet metal damage, bubbles in the potting, damaged conductor paths and many more.

Product

E-Modules

Industry

Automotive

Total savings / year​

240 - 380k €

Customer Request

Our customer produces several thousand e-modules every day. Due to the complex manufacturing processes, the error rate fluctuates greatly and can be as high as 20%. The quality control of the components was previously carried out manually in a 3-shift operation by a team of five employees per shift. In addition to special (UV) light, magnifying glasses were used to detect delicate defects (such as solder joint defects). As each component is checked for over 50 different defect classes, the inspection accuracy of the different inspection employees varied considerably, which led to pseudo rejects and customer complaints with significant financial risks.


With the introduction of Maddox AI, the actual inspection process could be fully automated and thus also rationalized. Defective components are now detected precisely and reliably, meaning that it is no longer necessary to rely on the subjective assessments of employees.

Before

Previous Inspection System

After

With Maddox AI

Each day, 15 employees checked the components, which led to a high level of manual effort and associated costs.

The actual quality inspection is now fully automated, which has considerably reduced the manual effort and significantly increased the efficiency of the inspections. One person per shift is still scheduled to load the inspection device.

Due to the inaccuracy of manual checks, there were several customer complaints.

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

No structured root cause analysis was carried out for the production of NOK parts.

Thanks to the digitalized inspection process, including the development of a defect heat map, the main causes of increased defect production could be identified.

Before

Previous Inspection System

Each day, 15 employees checked the components, which led to a high level of manual effort and associated costs.

Due to the inaccuracy of manual checks, there were several customer complaints.

No structured root cause analysis was carried out for the production of NOK parts.

After

With Maddox AI

The actual quality inspection is now fully automated, which has considerably reduced the manual effort and significantly increased the efficiency of the inspections. One person per shift is still scheduled to load the inspection device.

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

Thanks to the digitalized inspection process, including the development of a defect heat map, the main causes of increased defect production could be identified.

> 80%

reduced Personnel Costs

100%

Traceability of each component

7 Months

Investment Payback Period

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

Optimization of Brake Hose inspection

Medical Hose Inspection

Maddox AI’s fully automated quality inspection reliably detects various defects and impurities such as foreign bodies, bubbles, water droplets, hair and more.

Cosmetic Container Inspection

Surface Inspection, Positioning Errors and OCR

Cosmetic Container Inspection

Maddox AI’s software, as well as the implementation of three individual camera systems, can reliably inspect surfaces, labels and imprints.

Product

Cosmetic Container

Industry

Consumer Goods

Total savings / year​

90-120k €

Customer Request

Our customer produces up to 8,000 cosmetics containers per hour. Initially, the containers were randomly inspected manually by the quality department every one to two hours. In addition, gross defects were already detected by a classic inspection system.

Maddox AI has integrated three individual camera systems on the existing production line so that the front, back and bottom of the containers can be inspected. In addition to classic damages, the label as well as other imprints such as batch stamps are inspected.

Before

Previous Inspection System

After

With Maddox AI

Random manual inspections were carried out in three shifts by one employee. The detection accuracy varied depending on the inspector.

All products are 100% automatically inspected according to a precise defect book definition.

The plant received several customer complaints per year, 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 digitized inspection process, the production scrap rate has been reduced by 20%.

 

Before

Previous Inspection System

Random manual inspections were carried out in three shifts by one employee. The detection accuracy varied depending on the inspector.

The plant received several customer complaints per year, 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 100% automatically inspected according to a precise defect book definition.

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

Thanks to the digitized inspection process, the production scrap rate has been reduced by 20%.

100%

Digitized Analyses

0 € Costs

For expansion to further product country variants

80 - 130 Teile / Min.

Inspection Speed

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

Windshield Wiper Inspection

Maddox AI’s fully automated 100% quality inspection reliably detects even the smallest defects on rubber parts with complex geometries.

Inspection Cell for Plastic Seals

By integrating a rotating inspection cell with three cameras in combination with Maddox AI’s software, the plastic seals are reliably inspected 100% of the time.

Fully automated Wire Inspection

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

Inspection of Vacuum Interrupters

Surface Inspection

Inspection of Vacuum Interrupters

Maddox AI enables the reliable and automated identification of surface defects such as soiling, metallic abrasion, impact marks, scratches, chipping on ceramics and many other defects.

Product

Vacuum interrupters

Industry

Energy Technology

Total savings / year​

60 - 80k €

Customer Request

Our customer produces over 1,000 switch tubes every day, which are inspected 100% manually by at least one specialist per shift. Due to the intricate defects (flaking, scratches, contamination) that have to be found on a complex component geometry with copper and ceramic surfaces, magnifying glasses and special lighting were used in the manual inspection. Despite these tools, the inspection accuracy varied considerably between employees per shift.


With the introduction of Maddox AI, inspection has been fully automated so that defective components are detected accurately and reliably. In particular, components are now inspected consistently across all shifts and according to clearly defined defect criteria.

Before

Previous Inspection System

After

With Maddox AI

Quality control was carried out completely manually, which was very time-consuming and costly.

The quality check is now fully automated, which has significantly reduced the manual effort and considerably increased the efficiency of the checks.

As the test results were not recorded digitally, there was no structured root cause analysis for faulty (NOK) parts.

Thanks to the digitalized inspection process, the reject rate in production has been reduced.

On average, only 20% of the rejected parts were really faulty. 80% of the parts could still be sold without any problems / with minor reworking.

Thanks to the increased precision of the Maddox AI inspection system, the pseudo reject rate has been reduced by around 75%, virtually eliminating re-inspections and avoiding unnecessary costs.

Before

Previous Inspection System

Quality control was carried out completely manually, which was very time-consuming and costly.

As the test results were not recorded digitally, there was no structured root cause analysis for faulty (NOK) parts.

On average, only 20% of the rejected parts were really faulty. 80% of the parts could still be sold without any problems / with minor reworking.

After

With Maddox AI

The quality check is now fully automated, which has significantly reduced the manual effort and considerably increased the efficiency of the checks.

Thanks to the digitalized inspection process, the reject rate in production has been reduced.

Thanks to the increased precision of the Maddox AI inspection system, the pseudo reject rate has been reduced by around 75%, virtually eliminating re-inspections and avoiding unnecessary costs.

100%

Digitized Analyses

75%

Less Pseudo Scrap

15 Month

Investment Payback Period

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

Inspection of Ball Bearings

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

Inspection of E-Modules

Maddox AI enables the reliable and automated identification of defects such as sheet metal damage, bubbles in the potting, damaged conductor paths and many more.

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.

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!

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.
Medical Hose

Medical Hose Inspection

Surface Inspection

Medical Hose Inspection

Maddox AI’s fully automated quality inspection reliably detects various defects and impurities such as foreign bodies, bubbles, water droplets, hair and more.

Product

Medical Hose

Industry

Pharmaceutical and Medical

Total savings / year​

150 - 170k €

Customer Request

Our customer produces many kilometers of medical tubing every day in a 3-shift operation, whereby quality control was previously carried out manually by two employees per shift, who mainly check the tubing for bubbles by touch.
With the introduction of Maddox AI, the inspection was fully automated. Defective components are now reliably detected and ejected automatically without the need for employees to carry out the inspection.

Before

Previous Inspection System

After

With Maddox AI

The medical tubes (5 different variants) were tested manually in 3 shifts by a total of 6 employees.

All product variants are now inspected 100% automatically. Employees are no longer required for inspection.

As the test results were not recorded digitally, there was no structured root cause analysis for faulty (NOK) parts.

Thanks to the digitalized inspection process, the reject rate in production has been reduced.

On average, the plant received 5-10 customer complaints per year, which led to re-sorting actions and quality measures.

There have been no more customer complaints since the installation of Maddox AI.

Before

Previous Inspection System

The medical tubes (5 different variants) were tested manually in 3 shifts by a total of 6 employees.

As the test results were not recorded digitally, there was no structured root cause analysis for faulty (NOK) parts.

On average, the plant received 5-10 customer complaints per year, which led to re-sorting actions and quality measures.

After

With Maddox AI

All product variants are now inspected 100% automatically. Employees are no longer required for inspection.

Thanks to the digitalized inspection process, the reject rate in production has been reduced.

There have been no more customer complaints since the installation of Maddox AI.

85 %

Reduced pseudo scrap

> 90 %

Time savings on recalibrations

50 m / min

Inspection speed

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

Tractor Inspection

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

Optimization of Brake Hose inspection

Exhaust Filter Inspection

Surface Inspection

Exhaust Filter Inspection

Maddox AI’s consistent, fully automated 100% quality inspection reliably detects defects in over 50 different product types.

Product

Exhaust Filter

Industry

Automotive

Total savings / year​

50 - 60k €

Customer Request

Our customer produces filters in various designs in a 3-shift operation. Quality control was previously carried out using a rule-based camera system. However, due to the diversity of production, the existing camera system had to be recalibrated daily in order to limit the pseudo reject rate. Despite the regular recalibration of the camera system, the old camera control led to a pseudo reject rate of up to 80%, which in turn led to high scrap / re-sorting costs.


With the introduction of Maddox AI, the inspection system only needs to be recalibrated / retrained in a few cases and the average pseudo reject rate has been reduced by over 90%. Defective filters are now reliably rejected, while OK filters are reliably classified as OK even with different production batches.

Before

Previous Inspection System

After

With Maddox AI

The existing inspection system had to be recalibrated several times a day.

Das Inspektionssystem muss dank der Maddox AI Software nicht mehr nachkalibriert werden und kann auch mit unterschiedlichen Produkttypen zuverlässig umgehen.

Depending on the product type / batch, up to 80% of the filters were classified as NOK, although the majority of these were OK parts.

Thanks to the increased precision of the Maddox AI inspection system, the pseudo reject rate has been reduced by around 90%.

New product types had to be laboriously stored individually in the inspection software.

New filter types are automatically and reliably inspected by the Maddox AI software.

Before

Previous Inspection System

The existing inspection system had to be recalibrated several times a day.

Depending on the product type / batch, up to 80% of the filters were classified as NOK, although the majority of these were OK parts.

New product types had to be laboriously stored individually in the inspection software.

After

With Maddox AI

The inspection system no longer needs to be recalibrated thanks to the Maddox AI software and can reliably handle different product types.

Thanks to the increased precision of the Maddox AI inspection system, the pseudo reject rate has been reduced by around 90%.

New filter types are automatically and reliably inspected by the Maddox AI software.

> 95%

Reduced recalibration time

> 95%

Reduced pseudo rejects

>50

Inspected product variants on a line

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

Inspection of Ball Bearings

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

Inspection of Medical Syringes

Surface defects such as impurities, burrs and cracks on tips of syringes can now be reliably checked automatically with the help of Maddox AI.

Inspection of E-Modules

Maddox AI enables the reliable and automated identification of defects such as sheet metal damage, bubbles in the potting, damaged conductor paths and many more.

Inspection of Medical Syringes

Surface Inspection

Inspection of
Medical Syringes

Surface defects such as impurities, burrs and cracks on tips of syringes can now be reliably checked automatically with the help of Maddox AI.

Product

Medical Syringes

Industry

Pharma

Total savings / year

70-120k €

Customer Request

Our customer produces up to 3,000 cartridges per day. Here, especially the cartridge tips need to be checked. Before using the Maddox AI system, this was done manually by two employees per shift. Depending on the shift and personnel, the sorting accuracy varied considerably.

Maddox AI has developed a semi-automatic inspection device that uses two cameras to inspect the cartridges from the top and focus on the tip. The system detects contaminations, burrs and cracks, among other things.

Before

Previous Inspection System

After

With Maddox AI

The cartridges were checked manually in three shifts by a total of three to 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.

The plant received an average of two customer complaints per year, 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 digitized inspection process, the production scrap rate has been reduced by 20%.

Before

Previous Inspection System

The cartridges were checked manually in three shifts by a total of three to six employees. The detection accuracy varied depending on the inspector.

The plant received an average of two customer complaints per year, 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. The inspection system is operated by one employee per shift.

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

Thanks to the digitized inspection process, the production scrap rate has been reduced by 20%.

100%

Digitized Analyses

60%

Less Pseudo Scrap

MES Integration

Enables use of data for process improvements

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

Inspection of Vacuum Interrupters

Maddox AI enables the reliable and automated identification of surface defects such as soiling, metallic abrasion, impact marks, scratches, chipping on ceramics and many other defects.

Gear Inspection

Maddox AI enables the reliable automation of surface inspections for metal objects, accurately detecting defects such as scratches and dents.

Optimization of Brake Hose inspection

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!

Medical Hose Inspection

Maddox AI’s fully automated quality inspection reliably detects various defects and impurities such as foreign bodies, bubbles, water droplets, hair and more.

Inspection of Ball Bearings

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

Windshield Wiper Inspection

Surface Inspection

Windshield Wiper Inspection

Maddox AI’s fully automated 100% quality inspection reliably detects even the smallest defects on rubber parts with complex geometries.

Product

Windshield Wiper

Industry

Automotive

Total savings / year​

80 - 132k €

Customer Request

Our customer produces rubber hoses in a 3-shift operation. Quality control was carried out manually & randomly by 1 employee per shift. With the introduction of Maddox AI, the inspection was fully automated. Defective components are now reliably detected and ejected automatically without the need for employees to carry out the inspection.

 

Before

Previous Inspection System

After

With Maddox AI

Quality control was carried out manually by 1 employee per shift, which was time-consuming and costly.

The quality check is now fully automated, which has significantly reduced the manual effort and considerably increased the efficiency of the checks.

As the test results were not recorded digitally, there was no structured root cause analysis for faulty (NOK) parts.

Thanks to the digitalized inspection process, the reject rate in production has been reduced.

On average, the plant received 3-4 customer complaints per year, which led to re-sorting actions and quality measures.

There have been no more customer complaints since the installation of Maddox AI.

Before

Previous Inspection System

Quality control was carried out manually by 1 employee per shift, which was time-consuming and costly.

As the test results were not recorded digitally, there was no structured root cause analysis for faulty (NOK) parts.

On average, the plant received 3-4 customer complaints per year, which led to re-sorting actions and quality measures.

After

With Maddox AI

The quality check is now fully automated, which has significantly reduced the manual effort and considerably increased the efficiency of the checks.

Thanks to the digitalized inspection process, the reject rate in production has been reduced.

There have been no more customer complaints since the installation of Maddox AI.

16 Months

Investment Payback Period

No

customer complaints

20 m / min.

Inspection speed

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

Inspection of Ball Bearings

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

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.

Cosmetic Container Inspection

Maddox AI’s software, as well as the implementation of three individual camera systems, can reliably inspect surfaces, labels and imprints.

Fully automated Wire Inspection

Surface Inspection

Fully automated Wire Inspection

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

Product

Wire

Industry

Energy Technology

Total savings / year​

360-650k €

Customer Request

Our customer manufactures transformers where various wire elements are wound at high speeds. The winding systems sometimes operate at over 10 meters per second, making a 100% manual inspection impossible. If a transformer is wound incorrectly, it can lead to complete failure, resulting in significant costs.

Maddox AI has developed a smart camera device that uses mirrors to monitor 100% of the wire’s surface area. This device detects issues such as discoloration, burn marks, and insulation faults. 

Before

Previous Inspection System

After

With Maddox AI

It was manually checked in three shifts by an employee on a sample basis. The detection accuracy varied depending on the inspector.

All products are 100% automatically inspected according to a precise defect book definition, resulting in reduced scrap.

The plant received an average of 1-3 customer complaints per year, which had to be manually rechecked.

Since the installation of Maddox AI, no further customer complaints have been received.

Since the inspection results were not digitized, no structured root cause analysis was conducted for the production of non-conforming (NOK) parts.

Thanks to the digitized inspection process, several process optimizations, such as supplier management, have been implemented.

Before

Previous Inspection System

It was manually checked in three shifts by an employee on a sample basis. The detection accuracy varied depending on the inspector.

The plant received an average of 1-3 customer complaints per year, which had to be manually rechecked.

Since the inspection results were not digitized, no structured root cause analysis was conducted for the production of non-conforming (NOK) parts.

After

With Maddox AI

All products are 100% automatically inspected according to a precise defect book definition, resulting in reduced scrap.

Since the installation of Maddox AI, no further customer complaints have been received.

Thanks to the digitized inspection process, several process optimizations, such as supplier management, have been implemented.

100%

Digitized analyses

MES Integration

Enables the use of data for process improvements

180 Images / Second

Inspection speed

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

Inspection of E-Modules

Maddox AI enables the reliable and automated identification of defects such as sheet metal damage, bubbles in the potting, damaged conductor paths and many more.

Optimization of Brake Hose inspection

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.
Maddox AI logo