Inspection Of Steering Races Using AI-Based Machine Vision System To Double The Productivity | Qualitas Technologies

CLIENT/INDUSTRY BACKGROUND

The client is a leading innovator, manufacturer, and supplier of bearing. The headquarters of the company is in Mumbai, India. It has been successfully catering to the needs of its clients (big market players) for more than 50 years. The revenue of the company, for the year 2020 was more than 13.4 million U.S. dollars.

CLIENT’S PROBLEMS

No automation is placed at the end of the production line to inspect the steering races. This makes the inspection process the bottleneck point in the manufacturing line. Some Common defects are shown below –

Inspection Of Steering Races Using AI-Based Machine Vision System To Double The Productivity | Qualitas Technologies

Inspection Of Steering Races Using AI-Based Machine Vision System To Double The Productivity | Qualitas Technologies

PROBLEM IMPLICATIONS

  1. Low productivity is one of the outcomes as there is no automation at the inspection station. As a result, there is a shortage in supply.
  2. The client is losing customers as there is no automation in the inspection process.

CLIENT REQUIREMENTS

  1. To automate the inspection of steering races using an AI-based machine vision system. 
  2. To double the productivity from 2000 to 4000 per day by incorporating automated visual inspection systems.

CURRENT PROCESS

Inspection Of Steering Races Using AI-Based Machine Vision System To Double The Productivity | Qualitas Technologies

After assembling, the steering races are routed to the inspection station. The inspection is being done prior to packaging. Multiple lines are in operation with 10 personnel deployed at the end of each line for manual inspection. The finished races are collected in a bin, which is further picked by the operators for visual inspection. The defective bearings are placed in the rejection bins. 25 percent of the total workforce is being utilized in the inspection station. Due to limited human efficiency in terms of inspection, the number of shipped steering races is 2500 to 3000 per day and the inspection cycle time is 3-4 seconds 

BUSINESS IMPACT

  1. Penalties are being imposed by the customers due to delays in deliveries. 
  2. Recall rates are high as the defective races are delivered to the customers. 
  3. The cost of full-time employment of the operators and training them is high

SOLUTION USING MACHINE VISION

A 5 megapixels EagleEye camera with 16mm lens configuration will be installed to acquire clear images of the smallest feature i.e. 1 mm in the races. The frame rates would be greater than the speed of the production line and therefore, 15 frame rates would be a perfect fit in this case as the races will be stationary. The resolution of the camera sensor will be 2592×1944 to capture the complete area of interest as FOV (Field Of View) will be maintained to 60mmx45mm with the working distance to be 300mm. 

Dome lights (red in color) will be used with the fiber enclosure to focus the projection of lights onto the races. The reason for using the red lights is to minimize the reflection on the metal surface. 

SETUP

Inspection Of Steering Races Using AI-Based Machine Vision System To Double The Productivity | Qualitas Technologies

IMAGES 

DETECTING DEFECTS IN QEP(QUALITAS EAGLE-EYE® PLATFORM) 

Inspection Of Steering Races Using AI-Based Machine Vision System To Double The Productivity | Qualitas Technologies

CONCLUSION

Followings are the potential benefits of a machine vision system deployed on a production line – 

  1. Achieved accuracy of defects detection on steering races could be more than 97 percent.
  2. The cycle time of the inspection can be significantly reduced to 0.5 seconds.
  3. Manpower can be reduced by 90 percent and utilized in other complex operations.
Get In Touch With Us


Leave a Reply

SCHEDULE A CALL