Adaptive Data Selection-Based Machine Learning Algorithm for Prediction of Component Obsolescence.

Product obsolescence occurs in the manufacturing industry as new products with better performance or improved cost-effectiveness are developed. A proactive strategy for predicting component obsolescence can reduce manufacturing losses and lead to customer satisfaction. In this study, we propose a machine learning algorithm for a proactive strategy based on an adaptive data selection method to forecast the obsolescence of electronic diodes. Typical machine learning algorithms construct a single model for a dataset. By contrast, the proposed algorithm first determines a mathematical cover of the dataset via unsupervised clustering and subsequently constructs multiple models, each of which is trained with the data in one cover. For each data point in the test dataset, an optimal model is selected for regression. Results of empirical experiments show that the proposed method improves the obsolescence prediction accuracy and accelerates the training procedure. A novelty of this study is that it demonstrates the effectiveness of unsupervised clustering methods for improving supervised regression algorithms.

Forecasting Obsolescence of Components by Using a Clustering-Based Hybrid Machine-Learning Algorithm.

Product obsolescence occurs in every production line in the industry as better-performance or cost-effective products become available. A proactive strategy for obsolescence allows firms to prepare for such events and reduces the manufacturing loss, which eventually leads to positive customer satisfaction. We propose a machine learning-based algorithm to forecast the obsolescence date of electronic diodes, which has a limitation on the amount of data available. The proposed algorithm overcomes these limitations in two ways. First, an unsupervised clustering algorithm is applied to group the data based on their similarity and build independent machine-learning models specialized for each group. Second, a hybrid method including several reliable techniques is constructed to improve the prediction accuracy and overcome the limitation of the lack of data. It is empirically confirmed that the prediction accuracy of the obsolescence date for the electrical component data is improved through the proposed clustering-based hybrid method.

Metal work-function-dependent barrier height of Ni contacts with metal-embedded nanoparticles to 4H-SiC.

Metal, typically gold [Au], nanoparticles [NPs] embedded in a capping metal contact layer onto silicon carbide [SiC] are considered to have practical applications in changing the barrier height of the original contacts. Here, we demonstrate the use of silver [Ag] NPs to effectively lower the barrier height of the electrical contacts to 4H-SiC. It has been shown that the barrier height of the fabricated SiC diode structures (Ni with embedded Ag-NPs) has significantly reduced by 0.11 eV and 0.18 eV with respect to the samples with Au-NPs and the reference samples, respectively. The experimental results have also been compared with both an analytic model based on Tung's theory and physics-based two-dimensional numerical simulations.

Nano-structure fabrication of GaAs using AFM tip-induced local oxidation method: different doping types and plane orientations.

In this study, we have fabricated nano-scaled oxide structures on GaAs substrates that are doped in different conductivity types of p- and n-types and plane orientations of GaAs(100) and GaAs(711), respectively, using an atomic force microscopy (AFM) tip-induced local oxidation method. The AFM-induced GaAs oxide patterns were obtained by varying applied bias from approximately 5 V to approximately 15 V and the tip loading forces from 60 to 180 nN. During the local oxidation, the humidity and the tip scan speed are fixed to approximately 45% and approximately 6.3 µm/s, respectively. The local oxidation rate is further improved in p-type GaAs compared to n-type GaAs substrates whereas the rate is enhanced in GaAs(100) compared to and GaAs(711), respectively, under the identical conditions. In addition, the oxide formation mechanisms in different doping types and plane orientations were investigated and compared with two-dimensional simulation results.

Fabrication and characterization of nano-channel field effect transistors patterned by AFM anodic oxidation.

We report a top-down approach based on atomic force microscope (AFM) local anodic oxidation (LAO) for the fabrications of the nanowire and nano-ribbon field effect transistors (FETs). In order to investigate the transport characteristics of nano-channel, we fabricated simple FET structures with channel width W approximately 300 nm (nanowire) and 10 microm (nano-ribbon) on 20 nm-thick silicon-on-insulator (SOL) wafers. In order to investigate the transport behavior in the device with different channel geometries, we have performed detailed two-dimensional simulations of nanowire and reference nano-ribbon FETs with a fixed channel length L and thickness t but varying channel width W from 300 nm to 10 microm. By evaluating the charge distributions, we have shown that the increase of 'on state' conduction current in SiNW channel is a dominant factor, which consequently result in the improved on/off current ratio of the nanowire FET.