Multitask learning for health condition identification and remaining useful life prediction: deep convolutional neural network approach

Predicting remaining useful life (RUL) is crucial for system maintenance. Condition monitoring makes not only degradation data available for RUL estimation but also categorized health status data for health state identification. However, RUL prediction has been treated as an independent process in most cases even though potential relevance exists with health status detection process. In this paper, we propose a convolution neural network based multi-task learning method to reflect the relatedness of RUL estimation with health status detection process. The proposed method applied to the C-MAPSS dataset for aero-engine unit prognostics supported superior performances to existing baseline models.

     

A run-to-run controller for a chemical mechanical planarization process using least squares generative adversarial networks

Achieving high processing quality for chemical mechanical planarization (CMP) in semiconductor manufacturing is difficult due to the distinct process variations associated with this method, such as drift and shift. Run-to-run control aims to maintain the targeted process quality by reducing the effect of process variations. The goal of controller learning is to infer an underlying output–input reverse mapping based on input–output samples considering the process variations. Existing controllers learn reverse mapping by minimizing the total mapping error for sample data. However, this approach often fails to generate inputs for unseen target outputs because conditional input distributions on target outputs are not captured in the learning. In this study, we propose a controller based on a least squares generative adversarial network (LSGAN) that can capture the input distributions. GANs are deep-learning architectures composed of two neural nets: a generator and a discriminator. In the proposed model, the generator attempts to produce fake input distributions that are similar to the real input distributions considering the process variation features extracted using convolutional layers, while the discriminator attempts to detect the fake distributions. Competition in this game drives both networks to improve their performance until the generated input distributions are indistinguishable from the real distributions. An experiment using the data obtained from a work-site CMP tool verified that the proposed model outperformed the comparison models in terms of control accuracy and computation time.

     

Customer entrepreneurship on digital platforms: Challenges and solutions for platform business models

Drawing on the mixed methods of qualitative research and agent‐based simulation, this study examines: (a) how end‐users use digital platforms to become customer–entrepreneurs undertaking commercial activities on platforms; and (b) how platform providers can convert this customer entrepreneurship into a revenue stream. Considering that end‐users have traditionally been defined as passive and uncharged actors in platform business models, an in‐depth understanding of their commercial activities and the viable revenue model to monetize this emerging customer practice is warranted. Our qualitative study reveals that customer–entrepreneurs make substantial use of platform offerings to advertise their products; communicate with end‐consumers; and accept payments. These commercial activities are largely exercised for free on platforms, even though they could otherwise serve as a source of revenue. On this point, our simulation results identify two pricing models achieving the generation of nearly identical revenues over time. First, platform providers may charge both advertising and transaction fees, which maximize the survival of professional customer–entrepreneurs. Second, platform businesses may levy advertising fees only, which maximizes the survival of informal customer–entrepreneurs operating on a micro‐scale and part‐time basis. This study offers theoretical, methodological, and managerial implications for platform studies.     

Creativity in non‐routine jobs: The role of transformational leadership and organizational identification

An increasing number of individuals work in jobs with little standardization and repetition, that is, with high levels of job non‐routinization. At the same time, demands for creativity are high, which raises the question of how employees can use job non‐routinization to develop creativity. Acknowledging the importance of social processes for creativity, we propose that transformational leaders raise feelings of organizational identification in followers and that this form of identification then helps individuals to develop creativity in jobs with little routinization. This is because organizational members evaluate and promote those ideas as more creative, which are in line with a shared understanding of creativity within the organization. To investigate these relationships, we calculated a mediated moderation model with 173 leader–follower dyads from China. Results confirm our hypotheses that transformational leadership moderates the relationship between job non‐routinization on employee creativity through organizational identification. We conclude that raising feelings of social identity is a key task for leaders today, especially when working in uncertain and fast developing environments with little repetition and the constant need to develop creative ideas.     

An investigation of the effect of online consumer trust on expectation, satisfaction, and post-expectation

This study proposes a framework regarding the relationship between consumer trust, satisfaction, expectation, and post-expectation in the context of electronic commerce. In particular, the framework draws together from three theories: social exchange theory, expectation-confirmation theory, and post-acceptance model of IS continuance. Following the longitudinal pre-purchase and post-purchase stages, this study provides a theoretical framework combining trust, expectation, satisfaction, and post-expectation (i.e., perceived usefulness) and tests the proposed model empirically using Internet consumer behavior data collected via two rounds of Web surveys. The empirical findings suggest that both consumer’s trust and expectation have positive influences on consumer’s satisfaction; a significant and positive relationship is detected between consumer’s trust and expectation; customer’s satisfaction and perceived usefulness as post-expectation belief are important predictors of repurchase intention. In consequence, the study provides a framework explaining the subsequent relationships of trust, expectation, confirmation, satisfaction, post-expectation, and repurchase intention (i.e., consumer trust → expectation → confirmation → satisfaction → post-expectation → repurchase intention) across pre-purchase and post-purchase stages. Theoretical and practical implications are discussed.     

Optimal tracking performance of MIMO discrete‐time systems with communication constraints

The optimal tracking problem for multiple‐input multiple‐output linear‐time‐invariant discrete‐time systems with communication constraints in the feedback path is studied in this paper. The tracking performance is measured by the energy of the error signal between the output of the plant and the reference signal. The objective is to obtain an optimal tracking performance, attainable by all possible stabilizing compensators. It is shown that the optimal tracking performance consists of two parts, one depends on the nonminimum phase zeros and zero direction of the given plant, as well as the reference input signal direction, and the other depends on the nonminimum phase zeros, unstable poles, and pole direction of the given plant, as well as the bandwidth and additive white Gaussian noise of the communication channel. It is also shown that, if the constraint of the communication channel does not exist, the optimal tracking performance reduces to the existing tracking performance of the control system without communication constraints. A typical example is given to illustrate the theoretical results. Copyright © 2011 John Wiley & Sons, Ltd.     

Stability analysis and H∞ control for hybrid complex dynamical networks with coupling delays

This paper formulates and studies a model of complex dynamical networks with switching topology and coupling delays. Based on the hybrid control and Lyapunov function, the stability and robust H_∞ control of such networks with impulsive and switching effects, which have not been studied before, are addressed with some criteria derived. Examples are given to illustrate the effectiveness of the theoretical results. Copyright © 2011 John Wiley & Sons, Ltd.     

Low‐voltage and high‐transmittance blue‐phase liquid‐crystal device with slanted electrodes

Abstract— A low‐voltage (～10 V_rms) and high‐transmittance (～90%) polymer‐stabilized blue‐phase liquid‐crystal (BPLC) device with a slanted‐electrodes structure is proposed. Unlike the vertical‐field‐switching (VFS) mode in which oblique incident light and a vertical field are employed, the proposed device utilizes normal incident light and an oblique field. The slanted electrodes generate a strong and uniform oblique electric field, which contributes in obtaining low voltage and high transmittance. Moreover, no couple films or prism sheets are needed, which helps to enhance the optical efficiency and simplify of the device structure. This device has great potential application for emerging BPLC displays and photonic devices.     

Roll‐to‐roll UV embossing process applied for light bar‐based sub‐wavelength gratings for backlight

This study proposes a roll‐to‐roll process‐based sub‐wavelength grating, which is attached on a light bar to turn the side‐lit red/green/blue (620, 520, and 450 nm) incident rays into a uniformly and normally output white light with high illuminance from the light bar's surface. On the basis of the rigorous coupling wave analysis, the relationship between the first‐order transmission/reflection efficiency and the pitch of the gratings with different shapes was analyzed. The optimal design can effectively reduce the coupling length and enhance the white color balance for display applications.     

Fast full‐color reflective display via photoluminescent enhancement

Reflective displays are advantageous in applications requiring low power or daylight readability. However, there are no low‐cost reflective technologies capable of displaying bright colors. By employing photoluminescence to more efficiently use ambient light, we created a prototype display that provides bright, full color in a simple, low‐cost architecture. This prototype includes a novel electrokinetic shutter, a layer that incorporates patterned luminescent red, green, and blue sub‐pixel elements, and a novel optical out‐coupling scheme. The luminescent elements convert otherwise‐wasted portions of the incident spectrum to light in the desired color band, resulting in improved color saturation and lightness. This prototype provides a color gamut that is superior to competing reflective display technologies that utilize color filters in single‐layer side‐by‐side sub‐pixel architectures. The current prototype is capable of switching in <0.5 s; future displays based on an alternative electro‐optic shutter technology should achieve video rate operation. A transflective version of this technology has also been prototyped. The transflective version utilizes its backlight with a power efficiency that is at least three times that of a conventional liquid crystal display. These photoluminescence‐based technologies enable a host of applications ranging from low‐power mobile products and retail pricing signage to daylight readable signage for outdoor advertising segments.