HP Inc.

Production performance prediction and optimization play an important role in securing smooth production and maintaining great efficiency. Traditional methods suffer from tardy and inflexible adjustments, leading to a manufacturing system with low-level responsiveness and adaptability. To pave the way for this, a digital twin (DT) emulator is proposed and driven by the collaboration of continuous prediction and iterative optimization. An overall system architecture is first presented to provide a generic reference. A multi-scale one-dimensional convolutional neural network (1DCNN) ensemble model is designed for continuous prediction, which integrates multi-scale kernels and ensemble structure to boost performance. A surrogate model of multivariate adaptive regression spline (MARS) is utilized to explicitly fit the relationship between production system variables and serve as the fitness function of optimization for appropriate parameter value searching. An industrial company collaboration case of production and assembly lines is used to demonstrate the effectiveness and feasibility of the proposed approach.

This study investigates the application of a drop‐on‐demand (DOD) thermal inkjet (TIJ)‐based bioprinting system for the fabrication of cell‐laden hydrogel microparticles (HMPs) with tunable sizes. The TIJ bioprinting technique involves the formation of vapor bubbles within the print chamber through thermal energy, expelling small droplets of bio‐ink onto a substrate. The study employs a heat‐treated saponified gelatin‐based bio‐ink, HSP‐GelMA. This bio‐ink is modified through methacrylic anhydride functionalization and undergoes subsequent saponification and heat treatment processes. Various concentrations of SPAN 80 surfactant in mineral oil were evaluated to assess their influence on HMP size and stability. The results indicate a direct correlation, with higher SPAN 80 concentrations resulting in smaller and more stable HMPs. The study further investigates the influence of jetting volume on HMP size distribution, revealing that larger jetting volumes lead to increased HMP sizes, attributed to droplet coalescence. This is supported by our further study via a Monte Carlo simulation, which shows that the mean droplet diameter grows approximately linear with the number of dispensed droplets. In addition, the study demonstrates the capability of the TIJ bioprinting system to achieve multimaterial encapsulation within HMPs, exemplified by staining living cells with distinct cytoplasmic membrane dyes. The presented approach provides insights into the controlled fabrication of cell‐laden HMPs, highlighting the versatility of the TIJ bioprinting system for potential applications in tissue engineering and drug delivery.

In this study, we adapted an HP D100 Single Cell Dispenser – a novel low-cost thermal inkjet (TIJ) platform with impedance-based single cell detection – for dispensing of individual cells and one-pot sample preparation. We repeatedly achieved label-free identification of up to 1,300 proteins from a single cell in a single run using an Orbitrap Fusion Lumos Mass Spectrometer coupled to either an Acquity UPLC M-class system or a Vanquish Neo UHPLC system. The developed sample processing workflow is highly reproducible, robust, and applicable to standardized 384- and 1536-well microplates, as well as glass LC vials. We demonstrate the applicability of the method for proteomics of single cells from multiple cell lines, mixed cell suspensions, and glioblastoma tumor spheroids. As additional proof of robustness, we monitored the results of genetic manipulations and the expression of engineered proteins in individual cells. Our cost-effective and robust single-cell proteomics workflow can be transferred to other labs interested in studying cells at the individual cell level.

The real-time monitoring of electromigration in ball-grid-array solder joints is limited to measuring the electrical resistance increase of the solder joints. Tracking the electromigration induced microstructural changes in solder balls requires cross sectioning which is a destructive technique. A novel planar solder geometry was invented and described here that allows real-time, non-destructive monitoring of microstructural changes and the rate of elemental segregation at the anode while simultaneously tracking the extent of electromigration by electrical resistance means. Electromigration in planar geometry tin-bismuth eutectic solder was studied by two means, (a) by the rate of Bi segregation at the anode and (b) by the rate of increase of electrical resistance of the solder, as a function of joint length, solder temperature and electrical current density. At low temperature and low electrical current density there was an extended initial period during which the joint resistance decreased before it increased. At higher temperatures and electrical current densities this initial period of decreasing resistance became less pronounced and at much higher temperature and current density stressing it became non-existent. The rate of bismuth segregation at the anode was somewhat proportional to the solder joint length indicating a probable Blech back-stress effect. Electromigration results from the rate of Bi segregation and the rate of increase of solder joint resistance were summarized using Arrhenius plots. The two plots gave similar electromigration activation energies of 0.7 eV from the electrical measurements and 0.75 eV from the Bi segregation measurements. The Arrhenius plot based on resistance rate increase was also used to predict the electromigration life of Sn-Bi solder joints under typical application conditions.

Freshwater scarcity is a pressing issue worldwide, and solar steam generators (SSGs) have emerged as a promising device for seawater desalination, harnessing renewable solar energy to facilitate sustainable water evaporation. The facile fabrication approach for SSG with complex topologies to achieve high water evaporation efficiency remains a challenge. Herein, a MIL-101 (Fe)-derived C@Fe3O4 ink was employed to multi-jet fusion (MJF) printing of polymeric porous SSGs with specific topologies. The optimized porous structure endows the printed SSGs with capillary force, greatly promoting water transport. The tree-like topology enables high water evaporation rates under various simulated solar radiation conditions. A finite element model was built to fully understand the light-to-thermal energy conversion and water evaporation processes. Moreover, the MJF-printed SSGs exhibit self-cleaning properties and can automatically remove accumulated salt on their surfaces, enabling sustainable desalination. During prolonged testing, the water evaporation rate of the SSGs remained relatively stable and reached as high as 1.55 kg m⁻² h⁻¹. Additionally, the desalinated water met the standards for direct drinking water. This study presents a state-of-the-art technology for producing efficient SSGs for desalination and introduces a novel method for MJF printing of functional nanocomposites.

Fueled by an unprecedented adoption of artificial intelligence, data centers are becoming the largest growing consumers of energy. This results in both an opportunity and necessity to reinvent a tighter relationship between IT and sustainable energy supplies.

Spiropyran is a dynamic organic compound that is distinguished by its reversible conversion between two forms: the colorless closed spiropyran (SP) form and the purple open merocyanine (MC) form. Typically triggered by UV light and reversed by visible light, spiropyran-functionalized surfaces offer reversible conversion in properties including color, polarity, reactivity, and fluorescence, making them applicable to diverse applications in chemical sensors, biosensors, drug delivery, and heavy metal extraction. While spiropyran has been successfully incorporated into various material platforms with SiO2 surfaces, its application on flat surfaces has been limited due to surface area constraints and a lack of standardized evaluation methods, which largely depend on the integration approach and substrate type used. In this study, we systematically review the existing literature and categorize integration methods and substrate types first and then report on our experimental work, in which we developed a streamlined three-step immobilization protocol, which includes surface activation, amination with (3-aminopropyl) triethoxysilane (APTES), and subsequent functionalization with carboxylic spiropyran (SP-COOH). Using SiO2 surfaces as a demonstration, we have also established a robust characterization protocol, consisting of contact angle measurements, X-ray photoelectron spectroscopy (XPS), ellipsometry, and fluorometric analysis. Our results evaluate the newly developed immobilization protocol, demonstrating effective activation and optimal amination using a 2% APTES solution, achieved in 5 min at room temperature. Fluorescence imaging provided clear contrast between the SP and the MC forms. Furthermore, we discuss limitations in the surface density of functional groups and steric hindrance and propose future improvements. Our work not only underscores the versatility of spiropyran in surface patterning but also provides optimized protocols for its immobilization and characterization on SiO2 surfaces, which may be adapted for use on other substrates. These advancements lay the groundwork for on-chip sensing technologies and other applications.

This paper considers the problem of making inference for a multicomponent stress–strength (MSS) model under Type-II censoring. It is assumed that stress–strength components of the multicomponent system follow unit generalized Rayleigh (UGR) distributions. Inferences are derived when the strength and stress have a common unknown UGR parameters. We derive maximum likelihood estimator of MSS reliability based on observed censored data. In sequel, interval estimator is evaluated using delta method and asymptotic normality property. Pivotal quantities based inference upon MSS reliability are derived as well. In addition, we further consider the case when all the parameters of stress–strength model are unknown and obtain various inferences for the reliability. Equivalence of model parameters is considered based on likelihood ratio test. Assessment of different methods is evaluated from Monte Carlo simulations and remarks are presented for further discussion. Three numerical examples including a petroleum reservoirs data are analyzed for illustration purposes.

Visual sensing has been widely adopted for quality inspection in production processes. This paper presents the design and implementation of a smart collaborative camera system, called BubCam , for automated quality inspection of manufactured ink bags in Hewlett-Packard (HP) Inc.’s factories. Specifically, BubCam estimates the volume of air bubbles in an ink bag, which may affect the printing quality. The design of BubCam faces challenges due to the dynamic ambient light reflection, motion blur effect, and data labeling difficulty. As a starting point, we design a single-camera system which leverages various deep learning (DL)-based image segmentation and depth fusion techniques. New data labeling and training approaches are proposed to utilize prior knowledge of the production system for training the segmentation model with a small dataset. Then, we design a multi-camera system which additionally deploys multiple wireless cameras to achieve better accuracy due to multi-view sensing. To save power of the wireless cameras, we formulate a configuration adaptation problem and develop the single-agent and multi-agent deep reinforcement learning (DRL)-based solutions to adjust each wireless camera’s operation mode and frame rate in response to the changes of presence of air bubbles and light reflection. The multi-agent DRL approach aims to reduce the retraining costs during the production line reconfiguration process by only retraining the DRL agents for the newly added cameras and the existing cameras with changed positions. Extensive evaluation on a lab testbed and real factory trial shows that BubCam outperforms six baseline solutions including the current manual inspection and existing bubble detection and camera configuration adaptation approaches. In particular, BubCam achieves 1.3x accuracy improvement and 300x latency reduction, compared with the manual inspection approach.

Convolutional neural networks (CNNs) have demonstrated encouraging results in image classification tasks. However, the prohibitive computational cost of CNNs hinders the deployment of CNNs onto resource-constrained embedded devices. To address this issue, we propose EdgeCompress, a comprehensive compression framework to reduce the computational overhead of CNNs. In EdgeCompress, we first introduce dynamic image cropping (DIC), where we design a lightweight foreground predictor to accurately crop the most informative foreground object of input images for inference, which avoids redundant computation on background regions. Subsequently, we present compound shrinking (CS) to collaboratively compress the three dimensions (depth, width, and resolution) of CNNs according to their contribution to accuracy and model computation. DIC and CS together constitute a multidimensional CNN compression framework, which is able to comprehensively reduce the computational redundancy in both input images and neural network architectures, thereby improving the inference efficiency of CNNs. Further, we present a dynamic inference framework to efficiently process input images with different recognition difficulties, where we cascade multiple models with different complexities from our compression framework and dynamically adopt different models for different input images, which further compresses the computational redundancy and improves the inference efficiency of CNNs, facilitating the deployment of advanced CNNs onto embedded hardware. Experiments on ImageNet-1K demonstrate that EdgeCompress reduces the computation of ResNet-50 by 48.8% while improving the top-1 accuracy by 0.8%. Meanwhile, we improve the accuracy by 4.1% with similar computation compared to HRank. The state-of-the-art compression framework. The source code and models are available at https://github.com/ntuliuteam/edge-compress .

Metal oxide nanorods exhibit promising potential as reinforcement fillers in various polymer matrices, but their application in the Multi Jet Fusion (MJF) technique is rarely reported. In this work, surface-modified zinc oxide nanorods (SMZnO) were synthesized and incorporated into polyamide 12 (PA12) powder to enhance the mechanical properties of the MJF-printed parts. Compared to ZnO, SMZnO exhibited better dispersion, resulting in markedly enhanced mechanical performances. The ultimate tensile strength and the Young's modulus of the MJF-printed SMZnO/PA12 nanocomposites were 62.02 MPa and 2.28 GPa in the X orientation and 64.07 MPa and 2.34 GPa in the Y orientation, equivalent to 27.85%, 59.44%, 29.12%, and 54.97% increments, respectively. The flexural strength and modulus demonstrated similar improvements in the X and Y orientations, confirming the uniform mechanical enhancement effect of homogenously distributed SMZnO. This work provides a novel and facile approach for the additive manufacturing of polymeric nanocomposites with superior mechanical performance.