Comparison of algorithms for error prediction in manufacturing with automl and a cost-based metric

In order to manufacture products at low cost, machine learning (ML) is increasingly used in production, especially in high wage countries. Therefore, we introduce our PREFERML AutoML system, which is adapted to the production environment. The system is designed to predict production errors and to help identifying the root cause. It is particularly important to produce results for further investigations that can also be used by quality engineers. Quality engineers are not data science experts and are usually overwhelmed with the settings of an algorithm. Because of this, our system takes over this task and delivers a fully optimized ML model as a result. In this paper, we give a brief overview of what results can be achieved with a state-of-the-art classifier. Moreover, we present the results with optimized tree-based algorithms based on RandomSearchCV and HyperOpt hyperparameter tuning. The algorithms are optimized based on multiple metrics, which we will introduce in the following sections. Based on a cost-oriented metric we can show an improvement for companies to predict the outcome of later product tests. Further, we compare the results from the mentioned optimization approaches and evaluate the needed time for them.

Digital transformation of machining workshop’s manufacturing system

The article proposes the method of intellectualization workshop’s manufacturing system. The digital technology is widely developed today so intellectualization of manufacturing system is relevant today. The use of AMT-supported manufacturing process in manufacturing system contributes to obtaining advantages in material and financial terms and reducing the cost of time resources.The use of automation technology in workshop process assumes cutbacks to its funding and ensures its digitalization. The digitalization is aimed at eliminating production errors and shortcomings already at the design stage of the production process. In consequence of this technological defects of products are reduced and the technological process is well organized. The use of an innovative approach to the development of the technological process of manufacturing products allows reducing time resources and increasing technical efficiency. Actualization of the suggested approach provided with system of computer-aided planning multiproduct technologic process. This system is developed by the authors and aimed at planning the intelligent product campaign. The presented algorithmic and software support for the formation of a rational set of control-and-measuring tools provides the development of a rational technological process for manufacturing a product and automation of the production’s technological preparation.

Development of a Material Design Space for 4D-Printed Bio-Inspired Hygroscopically Actuated Bilayer Structures with Unequal Effective Layer Widths

(1) Significance of geometry for bio-inspired hygroscopically actuated bilayer structures is well studied and can be used to fine-tune curvatures in many existent material systems. We developed a material design space to find new material combinations that takes into account unequal effective widths of the layers, as commonly used in fused filament fabrication, and deflections under self-weight. (2) For this purpose, we adapted Timoshenko’s model for the curvature of bilayer strips and used an established hygromorphic 4D-printed bilayer system to validate its ability to predict curvatures in various experiments. (3) The combination of curvature evaluation with simple, linear beam deflection calculations leads to an analytical solution space to study influences of Young’s moduli, swelling strains and densities on deflection under self-weight and curvature under hygroscopic swelling. It shows that the choice of the ratio of Young’s moduli can be crucial for achieving a solution that is stable against production errors. (4) Under the assumption of linear material behavior, the presented development of a material design space allows selection or design of a suited material combination for application-specific, bio-inspired bilayer systems with unequal layer widths.

Evaluation of Production Lots of a Rapid Point-of-Care Lateral Flow Serological Test Intended for Identification of IgM and IgG against the N-Terminal Part of the Spike Protein (S1) of SARS-CoV-2

The potential of rapid point-of-care (POC) tests has been subject of doubt due to an eventual risk of production errors. The aim was therefore to evaluate the two separate production lots of a commercial POC lateral flow test, intended for the detection of IgM and IgG against the SARS-CoV-2 spike protein (S1). Control samples consisted of serum from individuals with confirmed SARS-CoV-2 infection and pre-COVID-19 negative sera gathered from a biobank. The presence of anti-S1 IgM/IgG in the sera was verified by an in-house Luminex-based serological assay (COVID-19 SIA). One hundred samples were verified as positive for anti-S1 IgG and 74 for anti-S1 IgM. Two hundred samples were verified as negative for anti-S1 IgM/IgG. For the two lots of the POC-test, the sensitivities were 93.2% and 87.8% for IgM and 93.0% and 100% for IgG. The specificities were 100% for IgM and 99.5% for IgG. The positive predictive value was 100% for IgM and 98.9% and 99.0% for IgG. The negative predictive value was 97.6% and 95.7% for IgM, and 96.6% and 100% for IgG. The evaluated POC-test is suitable to assess anti-SARS-CoV-2 S1 IgM and IgG, as a measure of previous virus exposure on an individual level. The external validation of separate lots of rapid POC-tests is encouraged to ensure high sensitivity before market introduction.

Effects of Verb Network Strengthening Treatment on Sound-Level and Speech Production Errors in Individuals With Aphasia and Acquired Apraxia of Speech

Purpose This retrospective pilot study investigated whether sound-level and speech production errors decreased in confrontation naming following Verb Network Strengthening Treatment (VNeST) for four participants with acquired apraxia of speech (A-AOS) and aphasia for whom lexical retrieval was previously reported. Specifically, we investigated a potential increase in correct number of syllables per word and posttreatment changes across three domains of speech: segmental production, fluency, and prosody. It was hypothesized that treatment shown to increase lexical retrieval in persons with aphasia and A-AOS could potentially facilitate a reduction in sound-level and speech production errors consistent with dual diagnoses of A-AOS and aphasia. Method Naming responses from four participants with aphasia and A-AOS who previously participated in VNeST studies were investigated for correct number of syllables per word and measures of segmental speech, fluency, and prosody. Results Significant gains in at least one measure of speech production were reported for three of the participants. One participant demonstrated decreased segmental speech errors, two showed significant reduction on syllable segmentation, and two demonstrated a significant reduction on false starts and pauses. Significant gains in production of correct number of syllables were limited to one participant, and one participant did not demonstrate increased accuracy on any measure of speech production. Conclusions While speech production errors consistent with motor speech impairment cannot always be definitively distinguished from the confound of aphasia, two participants produced significantly decreased segmentation of syllables, a characteristic unique to A-AOS. The sound-level and speech production changes recorded may be attributed to a combination of interacting motor and language processes and resource allocation. In addition, specific components of VNeST may have contributed to speech production changes. Future work will focus on a prospective study of effects of language therapy (e.g., VNeST) on measures of speech production with investigation beyond the single-word level.

Single-parameter and parameter combination errors in L2 productions of Swiss German Sign Language

This article presents a study of errors committed by hearing adult L2 learners of Swiss German Sign Language (Deutschschweizerische Gebärdensprache, DSGS). As part of a statistical analysis of single-parameter errors, movement was found to be the parameter most susceptible to errors, followed by location, orientation, and handshape. An analysis of production errors with respect to combinations of manual parameters was also conducted, something that previously has not been undertaken. The parameter combination most frequently involved in errors was movement with location. Possible aspects contributing to the higher error rate for movement are suggested, among which are the inherent complexity of the movement parameter. Finally, the article discusses factors influencing the judging of errors.

Wind resource prospecting: Predicting winds aloft using surface winds and the position of the sun

The sun’s position in the sky appears to strongly influence the vertical wind gradient within the atmospheric boundary layer. This study uses a quantitative research design that combines the position of the sun along with surface wind speeds to estimate the wind gradient. The model applies to uncluttered simple terrain and requires only 10 m wind speed data to extrapolate wind speeds and wind turbine energy production to heights of 60 m or more. The average daytime and nighttime model wind speed errors are 1.7% and 2.4%, respectively. The average daytime and nighttime model turbine energy production errors are 3.8% and 6.1%, respectively. The model offers a practical and low-cost alternative to tall tower systems to assess wind resources, especially for remote sites.

KESALAHAN PRODUKSI KALIMAT PADA LOMBA DEBAT KONSTITUSI MAHASISWA TAHUN 2018

This research aims to describe sentence production errors especially pause and tongue slips in the 2018 Student Constitutional Debate Competition. The benefits of research for the researcher are to deepen knowledge about language, especially errors in the production of sentences, and for the reader to give knowledge to the public about the study of psycholinguistics in producing a sentence and broaden the insight into language about sentence production. Descriptive qualitative method was used in this study. This research used source data of 2018 Student Constitutional Debate Competition semifinal one and final one video. In this study the tapping technique was used for the data collection technique. Based on the results of data management and analysis, the sentence production errors contained in the 2018 Student Constitution Debate Contest video are (1) Breathing pause, (2) Doubt pause, (3) Malaproprism, (4) Transposition tongue slip, (5) Anticipation tongue slip, (6) Perseveration tongue slip. Abstrak Tujuan penelitian ini adalah untuk mendeskripsikan kesalahan produksi kalimat terutama senyapan dan kilir lidah pada Lomba Debat Konstitusi Mahasiswa Tahun 2018. Manfaat penelitian bagi peneliti adalah untuk memperdalam pengetahuan mengenai bahasa khususnya kesalahan produksi kalimat dan bagi pembaca yaitu untuk memberi pengetahuan kepada masyarakat tentang kajian Psikolinguistik dalam memproduksi suatu kalimat serta memperluas wawasan kebahasaan mengenai produksi kalimat. Dalam penelitian digunakan metode berupa desktiptif kualitatif. Penelitian menggunakan sumber data berupa video Lomba Debat Konstitusi Mahasiswa tahun 2018 babak semifinal satu dan final satu. Penelitian menggunakan teknik pengumpulan data berupa teknik sadap catat. Berdasarkan hasil pengelolaan dan analisis data, kesalahan produksi kalimat yang terdapat dalam video Lomba Debat Konstitusi Mahasiswa Tahun 2018 yaitu (1) Senyapan pernapasan, (2) Senyapan keraguan, (3) Kilir lidah malaproprisme, (4) Kilir lidah transposisi, (5) Kilir lidah antisipasi, (6) Kilir lidah perseverasi. Kata Kunci: Kesalahan Produksi Kalimat, Debat, Kilir Lidah, Senyapan

The Effect of Pitch Auditory Feedback Perturbations on the Production of Anticipatory Phrasal Prominence and Boundary

Purpose In this study, we investigated how the direction and timing of a perturbation in voice pitch auditory feedback during phrasal production modulated the magnitude and latency of the pitch-shift reflex as well as the scaling of acoustic production of anticipatory intonation targets for phrasal prominence and boundary. Method Brief pitch auditory feedback perturbations (±200 cents for 200-ms duration) were applied during the production of a target phrase on the first or the second word of the phrase. To replicate previous work, we first measured the magnitude and latency of the pitch-shift reflex as a function of the direction and timing of the perturbation within the phrase. As a novel approach, we also measured the adjustment in the production of the phrase-final prominent word as a function of perturbation direction and timing by extracting the acoustic correlates of pitch, loudness, and duration. Results The pitch-shift reflex was greater in magnitude after perturbations on the first word of the phrase, replicating the results from Mandarin speakers in an American English–speaking population. Additionally, the production of the phrase-final prominent word was acoustically enhanced (lengthened vowel duration and increased intensity and fundamental frequency) after perturbations earlier in the phrase, but more so after perturbations on the first word in the phrase. Conclusion The timing of the pitch perturbation within the phrase modulated both the magnitude of the pitch-shift reflex and the production of the prominent word, supporting our hypothesis that speakers use auditory feedback to correct for immediate production errors and to scale anticipatory intonation targets during phrasal production.