Search

Designing and installing jigs and fixtures for new products are very costly tasks. In order to rapidly respond to frequent product changes in a manufacturing system, we developed a transformable pin array fixture system including reconfiguration planning software that finds the optimal position for mounting an assembly part and automatically generates a pin height control code. We first classified main fixturing quality factors such as type of contact, distance between a pin and a joining point, and part shape. We then formulate the part positioning problem as an alignment problem between the joining point set and the pin point set in the xy plane in consideration of the fixturing quality factors. A car door trim assembly module was used to verify the proposed method. Finite-element analysis results showed that in a car door trim assembly process such as ultrasonic welding and screwing, relocating an assembly part at the optimal position reduces its deformation.

Monolithic 3-dimensional integration (M3D) requires heat treatment techniques that should not degrade the device performance of the lower layers when applied to device fabrication in the upper layers. Deep ultraviolet (DUV) laser annealing was implemented in the crystallization of amorphous silicon and source/drain junction activation to fabricate polycrystalline silicon (poly-Si) metal-oxide-semiconductor field-effect-transistors (MOSFETs) for upper-layer devices. The poly-Si MOSFET devices were fabricated successfully on top of the bottom electronic circuit layer (isolated by a thick silicon dioxide layer). Devices on the upper and lower layers were connected through interlayer vias to form a current-starved ring oscillator. The current-to-frequency converter on the bottom layer showed a reasonable increase in frequency as the current of the poly-Si MOSFET on the upper layer increased with increasing silicon grain size, confirming that DUV laser annealing did not degrade the performance of the bottom layer devices. This opens more opportunities for using DUV laser annealing in the M3D integration.

Approaches for predicting financial markets, including conventional statistical methods and recent deep learning methods, have been investigated in many studies. However, financial time series data (e.g., daily stock market index) contain noises that prevent stable predictive model learning. Using these noised data in predictions results in performance deterioration and time lag. This study proposes padding-based Fourier transform denoising (P-FTD) that eliminates the noise waveform in the frequency domain of financial time series data and solves the problem of data divergence at both ends when restoring to the original time series. Experiments were conducted to predict the closing prices of S&P500, SSE, and KOSPI by applying data, from which noise was removed by P-FTD, to different deep learning models based on time series. Results show that the combination of the deep learning models and the proposed denoising technique not only outperforms the basic models in terms of predictive performance but also mitigates the time lag problem.

The lateral distributions of free carriers and defects in a channel of top gate In-Ga-Zn-O thin-film transistors were extracted using a single pulse charge pumping method. A square pulse was applied to the gate, while the charge transport current was monitored at the source–drain. A transmission line model was used to confirm the charging mechanism and extract the effective charge length. The time dependence of the trapped charge density was determined by comparing the charging and discharging transient characteristics. The lateral profiles of the free carriers and defects were decoupled successfully using the time dependence of the effective charge length, free charges, and trapped charges.

Despite the recent advances in manufacturing automation, the role of human involvement in manufacturing systems is still regarded as a key factor in maintaining higher adaptability and flexibility. In general, however, modeling of human operators in manufacturing system design still considers human as a physical resource represented in statistical terms. In this paper, we propose a human in the loop (HIL) approach to investigate the operator's choice complexity in a mixed model assembly line. The HIL simulation allows humans to become a core component of the simulation, therefore influencing the outcome in a way that is often impossible to reproduce via traditional simulation methods. At the initial stage, we identify the significant features affecting the choice complexity. The selected features are in turn used to build a regression model, in which human reaction time with regard to different degree of choice complexity serves as a response variable used to train and test the model. The proposed method, along with an illustrative case study, not only serves as a tool to quantitatively assess and predict the impact of choice complexity on operator's effectiveness, but also provides an insight into how complexity can be mitigated without affecting the overall manufacturing throughput.