Supply chain data analytics for predicting supplier disruptions: a case study in complex asset manufacturing

Although predictive machine learning for supply chain data analytics has recently been reported as a significant area of investigation due to the rising popularity of the AI paradigm in industry, there is a distinct lack of case studies that showcase its application from a practical point of view. In this paper, we discuss the application of data analytics in predicting first tier supply chain disruptions using historical data available to an Original Equipment Manufacturer (OEM). Our methodology includes three phases: First, an exploratory phase is conducted to select and engineer potential features that can act as useful predictors of disruptions. This is followed by the development of a performance metric in alignment with the specific goals of the case study to rate successful methods. Third, an experimental design is created to systematically analyse the success rate of different algorithms, algorithmic parameters, on the selected feature space. Our results indicate that adding engineered features in the data, namely agility, outperforms other experiments leading to the final algorithm that can predict late orders with 80% accuracy. An additional contribution is the novel application of machine learning in predicting supply disruptions. Through the discussion and the development of the case study we hope to shed light on the development and application of data analytics techniques in the analysis of supply chain data. We conclude by highlighting the importance of domain knowledge for successfully engineering features.     

Multifunctional Ir–Ru alloy catalysts for reversal-tolerant anodes of polymer electrolyte membrane fuel cells

To address the problem of fuel starvation in fuel-cell electric vehicles,which causes cell voltage reversal and results in cell failure when repeated continuously,we developed a reversal-tolerant anode(RTA) to promote water oxidation in preference to carbon corrosion.Graphitized carbon-supported Ir-Ru alloys with different compositions are employed as RTA catalysts in an acidic polyol solution and are shown to exhibit composition-dependent average crystallite sizes of <5.33 nm.The adopted approach allows the generation of relatively well-dispersed Ir-Ru alloy nanoparticles on the carbon support without severe agglomeration.The activity of IrRu₂/C for the hydrogen oxidation reaction is 1.10 times that of the stateof-the-art Pt/C catalyst.Cell reversal testing by simulation of fuel starvation reveals that the durability of IrRu₂/C(~7 h) significantly exceeds that of the conventional Pt/C catalyst(~10 min) and is the highest value reported so far.Thus,the developed Ir-Ru alloy catalyst can be used to fabricate practical RTAs and replace Pt catalysts in the anodes of polymer electrolyte membrane fuel cells.     

A Multiwell Microfluidic Device for Analyzing and Screening Nonhormonal Contraceptive Agents

Birth control and family planning play pivotal roles in the economic growth and reduction of maternal, infant, and child mortality. Current contraceptives, such as hormonal agents and intrauterine devices, target only a small subset of reproductive processes and can have serious side effects on the health of women. To develop novel contraceptive agents, a scalable microfluidic device is established for analyzing and screening the effects of potential contraceptive agents on the maturation of the cumulus–oocyte complex. The microfluidic device performs on‐chip incubation for studying oocyte maturation and cumulus expansion and isolates the microwells by oil–water interfaces to avoid crosstalk between the wells. A filter membrane is incorporated in the device to simplify incubation, medium exchange, washing, and fluorescence staining of oocytes. Cumulus expansion can be monitored directly in the device and oocyte maturation can be examined after enzymatic removal of cumulus cells and on‐chip fluorescence staining. The performance of the device is evaluated by studying the influence of three drugs known to block oocyte maturation and/or cumulus expansion.     

A study on the citation situation within the citing paper: citation distribution of references according to mention frequency

Citation impact indicators play a significant role in evaluating the scientific research activity. Most of citation impact indicators are based on the citation count that the publication is cited as a reference in the other publications, but the difference between each citation situation was not considered. Normally, the number of citations that a publication is cited in the other publications may represent the formal quality of the publication. Similarly, the number of times that a publication is really mentioned within the citing publication, it may also represent the formal quality of the citation. We have examined about how many times each reference was really mentioned within the citing publications and studied about the citation situation within the citing publications. We verified that the citation distribution of references according to the mention frequency follows the Generalized Pareto distribution. The results showed that about 20% of total references were mentioned three and more times, and the number of citation mentions for the about 50% of total references were from about 20% of the total references in the citing publications.     

High Performance Classification of Android Malware Using Ensemble Machine Learning

Although Android becomes a leading operating system in market, Android users suffer from security threats due to malwares. To protect users from the threats, the solutions to detect and identify the malware variant are essential. However, modern malware evades existing solutions by applying code obfuscation and native code. To resolve this problem, we introduce an ensemble-based malware classification algorithm using malware family grouping. The proposed family grouping algorithm finds the optimal combination of families belonging to the same group while the total number of families is fixed to the optimal total number. It also adopts unified feature extraction technique for handling seamless both bytecode and native code. We propose a unique feature selection algorithm that improves classification performance and time simultaneously. 2-gram based features are generated from the instructions and segments, and then selected by using multiple filters to choose most effective features. Through extensive simulation with many obfuscated and native code malware applications, we confirm that it can classify malwares with high accuracy and short processing time. Most existing approaches failed to achieve classification speed and detection time simultaneously. Therefore, the approach can help Android users to keep themselves safe from various and evolving cyber-attacks very effectively.     

Clustering of platinum atoms into nanoscale particle and network on NaY zeolite

Platinum has been supported on NaY zeolite by ion exchange of Pt(NH₃)₄ ²⁺. Clustering of Pt atoms into a 1 nm cluster, and three-dimensional quantum-size wire (3 nm thick) network (greater than 100 nm) from reduction of the Pt species has been investigated by extended X-ray absorption fine structure transmission electron microscopy and xenon adsorption measurements.     

On the lateral instability of a thin beam under periodic bending loads

The lateral instability of a thin beam under periodic bending loads was investigated. Physical evidences of the instability were observed previously by experiments. But an analytical study has not been reported. The object of this study is to demonstrate the nature and existence of dynamic lateral instability. The harmonic balance method is applied to bifurcation modes which result from the stability change of torsional mode of a beam and then compared with numerical simulations. It is found, in a certain frequency range, that a small bending load results in the lateral instability when damping is small. Inha University     

On the modal curve of nonlinear normal modes

In a nonlinear holonomic conservative system having two-degree-of-freedom, the modal curves of normal mode vibrations are investigated investimgated by the harmonic balance method. The general procedure to compute the modal curve is suggested. Even if the linearized frequencies of the system are satisfied with the commensurability condition under which the approaches using the perturbation method have the problem of small divisor, the modal curve can be obtained by this method, provided that the fundamental harmonics are dominant when the normal modes are expanded in Fourier series in time domain. As an example, in a system with cubic nonlinearity the modal curves are computed analytically and numerically to compare both results.     

FVF LDO regulator with dual dynamic-load composite gain stage

A low-voltage output-capacitorless low-dropout regulator using dual dynamic-load composite gain stage for flipped voltage follower topology is presented. It also incorporates a delay discharge circuit which aims to reduce the long discharge time arising from the large capacitive load, thus achieving the overshoot time reduction and sustaining fast transient characteristic when driving low-power digital system with internal heavy capacitive load requirement. The regulator can support a minimum of 0.75 V input voltage with 0.5 V output voltage. It consumes 49.4 µA whilst maintaining the stability for a capacitance load range from 470 pF to 10 nF. For a current load transient from 0 to 10 mA with 200 ps edge time, the settling time is 0.38 µs for the load capacitance of 3 nF. The obtained transient figure-of-merit is 0.42 mV. This transient metric outperforms the representative prior-art reported works.