EVIS delivers services to transform your data to information.

The Semiconductor Industry uses Data Mining techniques to extract knowledge for use in predictive maintenance, fault detection design, production, quality assistance, scheduling, decision support systems, control, and CRM. EVIS can assist you in selecting and implementing those Data Mining techniques that are most suitable for your problem.

In the following paragraphs, we briefly describe some of the above-mentioned areas.

Yield enhancement Semiconductor companies gather more and more data, but it is still very difficult to identify the most important parameters required for yield modeling and prediction. It is said that “the amount of data generated is exceeding the yield’s engineer’s ability to effectively monitor and control unexpected trends and excursions”.  By using sophisticated data management and data mining tools it is possible to explain and predict yield excursions: Pattern extraction from wafer bin map (the result of circuit probe inspection of dies on a wafer at the end of fabrication) is performed in

• Manufacturing process monitoring and defect diagnosis in order to remove assignable causes and thus improve the yield; identification of the causal relationships between the machines of specific process and yield rate. 
• Visualization of massive mixed-type Semiconductor Data allows you to explicitly show the relations between variables and causes behind yield problems. 

Critical Product Performance in Manufacturing

In the manufacturing of the critical components of a product, it is important to ascertain the performance and behavior of the part components being produced before assembly. The acceptance tests of these components might be costly and may affect the cycle time production.  Intelligent data analysis uncovers the characteristics of the part components performance using past acceptance test data. This historical information it is used as a basis for monitoring and diagnosis.

Fault Detection and quality improvement

Examination of what happened in the past allows you to gain greater insight in your process, which can be used to predict the future system performance. Data Mining can determine the factors that influence the success or failure of a process. In addition, it can also be used to reduce the numbers of expensive tests while meeting the performance criteria, to diagnose accurately faults and determine types of faults and failures, to identify the failure mechanism of mechanical components, to inspect dynamic characteristic of the machinery. 

For more information, please contact us.