All functions have to be ensured to be working well prior to shipping the products to clients. In addition to manual inspection, contactless automatic optical inspection (AOI) or automatic visual inspection (AVI) equipment can be employed during the production process. With high-precision lens and machine vision, the equipment identifies defective products. Integrating such testing equipment into the ADC System can greatly enhance the efficiency of quality management.

With rich integration analysis experience in various fields, TYNE has an in-depth understanding of industry characteristics and improvement needs, and has developed a system that integrates unique deep learning methodologies and statistical algorithms based on solid theoretical foundation. The system can not only automatically categorize defective images, but also accurately identify the location of defects and support automatic reworking or repair decisions. It is highly accurate and expandable and can greatly reduce inspection manpowers and increase efficiency and the level of intelligence of factories.

Combining TYNE ADC with TYNE AYEDAS System allows quick defect inspection while enabling the identification of the root cause of the defect. An improvement can be done from the fundamental level while improving the yield rate and production efficiency, which are all helpful for the enhancement of competitiveness of the manufacturing industry.

Offline Modeling

Offline Server can be connected through the user interface of TYNE ADC System for defect inspection model training and its effect verification. Users can easily transfer the model trained by Offline Server to Online Server.

Online Operation

Online Server loads the corresponding model in real time for defect classification and location identification, and integrates the output with the production process. The whole process is carried out automatically, replacing as-is manual work.

Users can upload the classified defect images to the different training models that TYNE deploys on the server based on different industry needs through TYNE ADC System to build new models. The new model can automatically determine the classification of defects, identify the location of defects, and determines if automatic reworking or repair is needed.

Step 1

Organize images that need defect identification on the work station and classification to the catalogues named after each defect.

Step 2

Load the images to ADC System to build the model. After one or multiple training processes, the best model for the work station can be built.

Step 3

Use the best model for prediction and classification of defect images, and, if necessary, repeat Step 2 to gradually increase the accuracy of prediction.

Automatic Scheduling

Users can build multiple models at the same time; the system also allows multiple users to build models at the same time. TYNE ADC System automatically carries out model training and modeling in an orderly manner. Users are freed with more time to do more valuable tasks.

Reload enables fast and efficient model building

For new defect images that need to be added to the original model for remodeling, TYNE ADC System allows direct addition of the new images to the original model for modeling training. A new best model can be built without loading all the images used during the training of the original modeling. This time-saving and effective approach for building new models helps respond to the needs of manufacturing process in real time.

Release enables fast and efficient release of ADC models to the designated process

When a model is built and verified, users can release the model to any designated process, machine, and data file on the production line, preventing possible mistakes from manual data storage and increasing the effectiveness and accuracy of real-time inspection of that station.

Share Model

Users can decide whether to share and apply an established model to another designated process or machine. With the simplified modeling process, users are freed with more time to do more valuable tasks.

Smart and Automatic Debugging and Improvement

Integrating TYNE ADC System and TYNE EDA/YMS System allows rapid defect classification and location identification while identifying the root cause of the defect. Impacted batch can also be tracked to limit potential risk. The system provides automatic holistic health checks on causes related to defects and analyzes its correlation with yield rates for fundamental improvement and yield rate and production efficiency enhancement. The analytical process of defect cause and improvement is embedded as its unique expert system and can be employed by the company internally across facilities and borders, bringing maximum benefits to knowledge management.

Tremendous Return on Investment (ROE)

Defect issues can be identified and dealt with in real time by TYNE ADC System, improving yield rates and reducing client complaints. AOI and AVI inspection of production lines save at least 80% of labor cost. When used combined, TYNE ADC System and TYNE AYEDAS System can rapidly identify the root cause of defects for fundamental improvement, greatly enhancing yield rates and production efficiency.

Low Maintenance Cost

With just a URL and an authorization, users can build or access models and share the functions of the system remotely from all over the world. Convenience is ensured across borders, facilities, and departments. The system is easy to deploy, easy to maintain, and easy to expand; and with integrated functions such as access management, TYNE ADC System meets the needs of IT managers in all aspects.

An Easy-to-Manage System

A comprehensive user access management mechanism is provided for administrators to easily add/delete/modify user accounts, groups, accessible data and functions. No more headaches when there is an organizational change. Users can set the schedule of modeling and produce ADC Models for new products and new processes on their own without going through the IT department, making most of GPU up time.