Because of the rapidly changing environment and high uncertainties, the semiconductor industry is in need of appropriate forecasting technology. In particular, both the cost and time in the test process are increasing because the process becomes complicated and there are more factors to consider. In this paper, we propose a prediction model that predicts a final “good” or “bad” on the basis of preconditioning test data generated in the semiconductor test process. The proposed prediction model solves the classification and regression problems that are often dealt with in the semiconductor process and constructs a reliable prediction model. We also implemented a prediction model through various machine learning algorithms. We compared the performance of the prediction models constructed through each algorithm. Actual data of the semiconductor test process was used for accurate prediction model construction and effective test verification.

In an attempt to estimate the life projection of LED packages, IESNA published a paper regarding an LED package measurement test method in 2008, and a life projection technical document in 2011, to be used for LED life estimation. IESNA’s publications regarding LED package measurement methods were functional, but they were not internationally standardized before 2017. In order to develop a standardized method, the International Standard chose to use the LM-80 as a measurement method for LED life projection in their publication in 2017. Many projection methods have been discussed by the IEC Technical Committee 34 working group, including the method using an exponential function, which reflects lumen degradation characteristics well. This study is designed to explore alternative LED package life estimation methods using an exponential function with statistical analysis, other than the one suggested by the International Standard.

This research was analyzed thermal characteristics that was appointed disadvantage when smart LED driver ICs was packaged and we applied extracted thermal characteristics for optimal layout design. We confirmed reliability of smart LED driver ICs package without additional heat sink. If the package is not heat sink, we are possible to minimize package. For extracting thermal loss due to overshoot current, we increased driver current by two and three times. As a result of experiment, we obtained 22 mW and 49.5 mW thermal loss. And we obtained optimal data of 350 mA driver current. It is important to distance between power MOSFET and driver ICs. If thhe distance was increased, the temperature of package was decreased. And so we obtained optimal data of 3.7 mm distance between power MOSFET and driver ICs. Finally, we fabricated real package and we analyzed the electrical characteristics. We obtained constant 35 V output voltage and 80% efficiency.

This research was analyzed thermal characteristics that was appointed disadvantage when smart LED driver ICs was packaged and we applied extracted thermal characteristics for optimal layout design. We confirmed reliability of smart LED driver ICs package without additional heat sink. If the package is not heat sink, we are possible to minimize package. For extracting thermal loss due to overshoot current, we increased driver current by two and three times. As a result of experiment, we obtained 22 mW and 49.5 mW thermal loss. And we obtained optimal data of 350 mA driver current. It is important to distance between power MOSFET and driver ICs. If thhe distance was increased, the temperature of package was decreased. And so we obtained optimal data of 3.7 mm distance between power MOSFET and driver ICs. Finally, we fabricated real package and we analyzed the electrical characteristics. We obtained constant 35 V output voltage and 80% efficiency.

Recently, remote phosphor is reported for white LED enhancing of phosphor efficiency compared with conventional phosphor-based W-LED. In this study, Remote phosphor was produced by screen printing coating on glass substrate with phosphor contents rated paste and heat treatment. The paste consists of phosphor, lowest softening glass frit and organic binders. Remote phosphor can be well controlled by varying the phosphor content rated paste. After mounting remote phosphor on top of blue LED chip, CCT, CRI, and luminance efficiency were measured. The measurement results showed that CCT, CRI. and luminance efficiency were 6,645, 68, and 1,161 n/W in phosphor 80 wt.% remote phosphor sintered at 600℃.

We developed a package of remote phosphor structure having blue LED chips and phosphors physically separated, and the characteristics were evaluated according to different classifications of phosphor coatings. Remote phosphor was produced by screen printing coating on glass substrate with phosphor content rated paste and heat treatment. After mounting Remote phosphor, which has been classified according to number of coatings, on top of blue LED chips, luminous flux, luminous efficacy, CCT and CRI were measured. The measurement results showed the most suitable characteristics of white LED package as a general light source when the content rate of phosphor in Remote phosphor was 80 wt.% with 3 layers of coatings and thickness over 12 μm.