Search

Recent advances in sensor technology and computing capabilities have enabled the creation of data-driven models that can support real-time decision making. Such a decision aid can allow for predictive maintenance (PdM) to be undertaken on a much greater scale in manufacturing plants. PdM includes data-driven prognostics and health management (PHM). A key element in developing prognostic models involves the acquisition of high-quality data, traditionally achieved through feature extraction methods to distill meaningful insights from extensive and noisy datasets. However, such methods may not handle noisy data well or address measurement errors adequately, potentially resulting in extracted features that inadequately represent the degradation process as a machine approaches failure or fault. Also, effects of sensor types on the feature extraction and prediction model have not been much explored yet. To overcome this limitation, we proposed a solution which involves dynamic feature extraction where a statistical penalty is introduced to mitigate the influence of noisy statistical features within a monotonic trend. Subsequently, the features extracted using this method are utilized to construct a health indicator (HI). Leveraging historical HI values, a probabilistic regression model may be used to forecast the time to failure (TTF) of rotating machinery with uncertainty propagation. To validate the proposed method, acceleration data were collected from rotating machinery for several run-to-failure cases. The proposed method is demonstrated to provide excellent forecasts of TTF for both accelerometer types. [ABSTRACT FROM AUTHOR]

The near-infrared spectra technique is an effective and non-destructive analysis method for polymer identification, and it has been applied in waste plastics classification. Since the near-infrared spectra are easily affected by instrumental noise, baseline drift and light scattering, spectral data preprocessing is indispensable for classification tasks. In this paper, a classification method is proposed to improve the accuracy of waste smartphone plastics classification. Savitzky–Golay smoothing, moving average smoothing, first-derivative, second-derivative, standard normal variate, and multiplicative scattering correction were used for noise reduction, baseline correction, and scattering correction. These algorithms were combined into five strategies for the spectral data preprocessing, and the optimal spectral data preprocessing strategy was selected based on the accuracy of mixed plastics classification. Successive projections algorithm and competitive adaptive reweighted sampling were used to extract the spectral feature and also the optimal spectral feature extraction algorithm was selected according to the accuracy of mixed plastics classification. The NIR spectra data of four waste smartphone plastics: polyamide, polycarbonate, acrylonitrile butadiene styrene, and polycarbonate/acrylonitrile butadiene styrene blend were analyzed to illustrate the performance of the proposed method. The results show that the accuracy of the proposed method is improved by 11.2% on average compared with the method without optimization. [ABSTRACT FROM AUTHOR]

Predictive maintenance (PdM) is an advanced technique to predict the time to failure (TTF) of a system. PdM collects sensor data on the health of a system, processes the information using data analytics, and then establishes data-driven models that can forecast system failure. Deep neural networks are increasingly being used as these data-driven models owing to their high predictive accuracy and efficiency. However, deep neural networks are often criticized as being "black boxes," which owing to their multi-layered and non-linear structure provide little insight into the underlying physics of the system being monitored and that are nontransparent and untraceable in their predictions. In order to address this issue, the layer-wise relevance propagation (LRP) technique is applied to analyze a long short-term memory (LSTM) recurrent neural network (RNN) model. The proposed method is demonstrated and validated for a bearing health monitoring study based on vibration data. The obtained LRP results provide insights into how the model "learns" from the input data and demonstrate the distribution of contribution/relevance to the neural network classification in the input space. In addition, comparisons are made with gradient-based sensitivity analysis to show the power of LRP in interpreting RNN models. The LRP is proved to have promising potential in interpreting deep neural network models and improving model accuracy and efficiency for PdM. [ABSTRACT FROM AUTHOR]

Greenhouse gas emissions have become one of the most prominent global concerns of sustainable development. To reduce product life cycle carbon footprint, planning should begin at embodiment design phase. The accurate assessment of carbon footprint is the foundation of carbon footprint reduction. However, existing carbon footprint models cannot be applied to embodiment design phase due to incomplete and limited design information. With this in mind, this paper proposes a carbon footprint model for embodiment design based on macro-micro design features. First, a Function-Structure-Feature (FSF) model for embodiment design is established to convey the design information. The concept of design features is introduced (at both macro and micro levels). The macro design feature denotes the different operational states of the product and the constraint relationships between parts. The micro design feature denotes the specific properties of parts. Then, a model of product carbon footprint based on design features is presented through the analysis of the relationships between macro-micro design features and product carbon footprint. The feasibility of the proposed method is demonstrated through a gear hobbing machine. The product carbon footprint model allows quantitative evaluation of product carbon footprint during embodiment design phase, and the amount of carbon footprint from each type of design feature is predicted. Based on evaluation result, the design features can be improved to reduce product carbon footprint. Case study results show that the carbon footprint is decreased by 10.96% after improving design features. [ABSTRACT FROM AUTHOR]

Prediction of manufacturing equipment's energy consumption plays an important role in selecting appropriate process parameters for energy saving. However, it is difficult to model the energy consumption of metal forming equipment during the drawing process characterized by variable process parameters and dynamic loads. In this paper, a model was developed to quantify the energy consumption of a typical hydraulic press during the drawing process under a range of operating conditions. The hydraulic press studied consists of two different circuits and two controllable parameters, i.e., punch velocity and blank holder force can be set for drawing processes performed. To start, the energy flow during the drawing process was analyzed by using the energy conversion mechanism and components' specifications to understand the detailed energy characteristics of each circuit. Then, orthogonal experiments including these two parameters at three levels were designed and carried out to find significant parameters. Finally, the contribution of each parameter, which was obtained from the analysis of variance (ANOVA) of the experimental results, was used to simplify energy flow modeling efforts. Consequently, a model of the press was established and used to predict the energy consumption of the drawing processes with different parameters. Good agreement with the experimental results was observed. The model can be used to identify parameters for minimal energy consumption, while the approach could be adapted to develop an energy consumption model for different hydraulic equipment. [ABSTRACT FROM AUTHOR]

The application of cutting-edge technologies such as AI, smart sensors, and IoT in factories is revolutionizing the manufacturing industry. This emerging trend, so called smart manufacturing, is a collection of various technologies that support decision-making in real-time in the presence of changing conditions in manufacturing activities; this may advance manufacturing competitiveness and sustainability. As a factory becomes highly automated, physical asset management comes to be a critical part of an operational life-cycle. Maintenance is one area where the collection of technologies may be applied to enhance operational reliability using a machine condition monitoring system. Data-driven models have been extensively applied to machine condition data to build a fault detection system. Most existing studies on fault detection were developed under a fixed set of operating conditions and tested with data obtained from that set of conditions. Therefore, variability in a model's performance from data obtained from different operating settings is not well reported. There have been limited studies considering changing operational conditions in a data-driven model. For practical applications, a model must identify a targeted fault under variable operational conditions. With this in mind, the goal of this paper is to study invariance of model to changing speed via a deep learning method, which can detect a mechanical imbalance, i.e., targeted fault, under varying speed settings. To study the speed invariance, experimental data obtained from a motor test-bed are processed, and time-series data and time–frequency data are applied to long short-term memory and convolutional neural network, respectively, to evaluate their performance. [ABSTRACT FROM AUTHOR]

Restaurants in China have recently begun offering group-buying (GB) options on the internet as a marketing and advertising tool to attract customers. GB coupon is available to parties of any size and the groups do not need to go together to the restaurants. Such restaurants serve two types of customers: GB customers with GB coupons and regular customers without coupons. One common practice for these restaurants is to set a maximum number of tables for each type of customer in advance and ask customers to wait in separate queues when all the tables for that customer type are occupied. As a result, restaurants are interested in finding optimal table allocation to serve the two types of customers to maximize profits. Since customers come follow a stochastic and discrete process, eat on a table that can be regarded as being served by a server in a queueing system, and wait in the queue when the restaurant is full, the dining process of customers and operating process of the restaurant is suitable to be described by queueing system. So, this study applies queueing theory to examine the table allocation problem. The effects of customer related parameters such as arrival rate and patience degree on the optimal allocation are discussed. The simulation model is extended to consider customers arriving in parties of different and serving tables of different sizes. We find that for a specific type of customer, if the arrival rate increases, the number of tables allocated for them increases. Patience degree has opposite influences on table allocation for the two types of customers: if regular customers are not patient, more tables should be allocated to them; while if GB customers are not patient, less tables should be allocated to them. If considering different customer party sizes and table sizes, as the arrival rate of regular customer increases, number of GB table decreases, number of tables for large (small) regular party first decreases (increases) and then increases (decreases). [ABSTRACT FROM AUTHOR]

The nature of the first genetic polymer is the subject of major debate1. Although the 'RNA world' theory suggests that RNA was the first replicable information carrier of the prebiotic era—that is, prior to the dawn of life2,3—other evidence implies that life may have started with a heterogeneous nucleic acid genetic system that included both RNA and DNA4. Such a theory streamlines the eventual 'genetic takeover' of homogeneous DNA from RNA as the principal information-storage molecule, but requires a selective abiotic synthesis of both RNA and DNA building blocks in the same local primordial geochemical scenario. Here we demonstrate a high-yielding, completely stereo-, regio- and furanosyl-selective prebiotic synthesis of the purine deoxyribonucleosides: deoxyadenosine and deoxyinosine. Our synthesis uses key intermediates in the prebiotic synthesis of the canonical pyrimidine ribonucleosides (cytidine and uridine), and we show that, once generated, the pyrimidines persist throughout the synthesis of the purine deoxyribonucleosides, leading to a mixture of deoxyadenosine, deoxyinosine, cytidine and uridine. These results support the notion that purine deoxyribonucleosides and pyrimidine ribonucleosides may have coexisted before the emergence of life5. A prebiotic synthesis of the purine DNA nucleosides (deoxyadenosine and deoxyinosine) in which the pyrimidine RNA nucleosides (cytidine and uridine) persist has implications for the coexistence of DNA and RNA at the dawn of life. [ABSTRACT FROM AUTHOR]

Tool wear is one of the consequences of a machining process. Excessive tool wear can lead to poor surface finish, and result in a defective product. It can also lead to premature tool failure, and may result in process downtime and damaged components. With this in mind, it has long been desired to monitor tool wear/tool condition. Kernel principal component analysis (KPCA) is proposed as an effective and efficient method for monitoring the tool condition in a machining process. The KPCA-based method may be used to identify faults (abnormalities) in a process through the fusion of multi-sensor signals. The method employs a control chart monitoring approach that uses Hotelling's T2-statistic and Q-statistic to identify the faults in conjunction with control limits, which are computed by kernel density estimation (KDE). KDE is a non-parametric technique to approximate a probability density function. Four performance metrics, abnormality detection rate, false detection rate, detection delay, and prediction accuracy, are employed to test the reliability of the monitoring system and are used to compare the KPCA-based method with PCA-based method. Application of the proposed monitoring system to experimental data shows that the KPCA based method can effectively monitor the tool wear. [ABSTRACT FROM AUTHOR]

In a highly automated manufacturing plant, the reliability of manufacturing equipment is critical for normal operation. A sudden machine breakdown can bring unexpected downtime, shorter lifespan of equipment, and lower operational efficiency. Breakdowns can lead to defective parts and consume extra energy—issues that are undesirable from an environmental standpoint—and also erode productivity and increase costs. To improve machine tool reliability, a machine may be continuously monitored to track its health condition. Monitoring a machine often provides large amounts of data that must be processed to distill useful information. Electric motors are found in many pieces of common manufacturing equipment. Deep learning methods can be combined with data collected on motors, e.g., acceleration time-frequency data, to identify motor condition. In this paper, three state-of-the-art deep learning architectures are evaluated for their ability to effectively monitor motor condition. Experiments are performed on a lab-scale motor test bed to secure condition data for several common motor faults. Tri-axial acceleration data are collected and converted into 2D images (spectrograms) using the power spectral density function. Some of these experiments are used to tune the deep learning algorithms, and others are used to test the proposed monitoring methods. The relative performances of the architectures are assessed, and it is demonstrated that the use of time-frequency images within a deep learning context can efficiently handle large amounts of data and effectively monitor the motor condition. [ABSTRACT FROM AUTHOR]

An important production planning problem is how to best schedule jobs (or lots) when each job consists of a large number of identical parts. This problem is often approached by breaking each job/lot into sublots (termed lot streaming). When the total number of transfer sublots in lot streaming is large, the computational effort to calculate job completion time can be significant. However, researchers have largely neglected this computation time issue. To provide a practical method for production scheduling for this situation, we propose a method to address the n-job, m-machine, and lot streaming flow-shop scheduling problem. We consider the variable sublot sizes, setup time, and the possibility that transfer sublot sizes may be bounded because of capacity constrained transportation activities. The proposed method has three stages: initial lot splitting, job sequencing optimization with efficient calculation of the makespan/total flow time criterion, and transfer adjustment. Computational experiments are conducted to confirm the effectiveness of the three-stage method. The experiments reveal that relative to results reported on lot streaming problems for five standard datasets, the proposed method saves substantial computation time and provides better solutions, especially for large-size problems. [ABSTRACT FROM AUTHOR]

Manufacturing processes, such as machining, transform raw materials into finished goods, and these processes consume significant energy. There is an increasing concern about the energy required for such processes and the environmental consequences attributable to the generation of the energy. Reducing the energy required to perform machining operations will not only reduce the environmental footprint, but also provide economic benefits. To that end, the effects of cutting conditions (e.g., feed and speed) and tool geometry (diameter and number of teeth) on the power required for an end milling operation are investigated experimentally. Experimental results are presented from a cutting mechanism perspective with the goal of understanding the role of the process variables. The specific cutting energy (SCE) is found decreasing when material removal rate increases, but there is substantial variation about the general trend. In essence, the cutting parameters and the tool geometry influenced the changes of average chip thickness and cutting speed, which cause the shear deformation energy changes and eventually collectively influence the SCE's change. Based on the experiments, suggestions on selecting process parameters are provided to improve milling energy efficiency. [ABSTRACT FROM AUTHOR]

When considering life's aetiology, the first questions that must be addressed are "how?" and "where?" were ostensibly complex molecules, considered necessary for life's beginning, constructed from simpler, more abundant feedstock molecules on primitive Earth. Previously, we have used multiple clues from the prebiotic synthetic requirements of (proto)biomolecules to pinpoint a set of closely related geochemical scenarios that are suggestive of flow and semi-batch chemistries. We now wish to report a multistep, uninterrupted synthesis of a key heterocycle (2-aminooxazole) en route to activated nucleotides starting from highly plausible, prebiotic feedstock molecules under conditions which mimic this scenario. Further consideration of the scenario has uncovered additional pertinent and novel aspects of prebiotic chemistry, which greatly enhance the efficiency and plausibility of the synthesis. [ABSTRACT FROM AUTHOR]

Circular economy is a promising business model that promotes sustainable development by closing material loops. Making progress toward a circular economy requires the recovery of valuable materials and components from end-of-use products and subsequent reuse of them in some form, thus maximizing the utility of components and materials. Currently, end-of-use products value recovery is carried out without a rational planning, causing the loss of the recoverable value embedded in material and components. To address this problem, dismantling planning and appropriate technologies should be employed to improve the economic performance of end-of-use products value recovery. In this paper, a two-stage dismantling planning method is proposed to find a profitable end-of-use strategy. In the first stage of this method, disassembly optimization model is constructed and can be executed to obtain the optimal disassembly plan allowing maximum preservation of component function value, in a preservative disassembly scenario. To speed up the modeling, a method for automatic generation of AND/OR graph-a structure of incorporating all possible disassembly operations and associated subassemblies, is presented. In the second stage of the method, in order to increase profitability, Pareto analysis is employed to identify bottlenecks to disassembly and automated/destructive technologies are considered to remove the bottlenecks. A hard disk drive serves as a case study to illustrate the suggested method. [ABSTRACT FROM AUTHOR]

We consider the problem of scheduling jobs on a single machine to minimize the total electricity cost of processing these jobs under time-of-use electricity tariffs. For the uniform-speed case, in which all jobs have arbitrary power demands and must be processed at a single uniform speed, we prove that the non-preemptive version of this problem is inapproximable within a constant factor unless $$\text {P} = \text {NP}$$ . On the other hand, when all the jobs have the same workload and the electricity prices follow a so-called pyramidal structure, we show that this problem can be solved in polynomial time. For the speed-scalable case, in which jobs can be processed at an arbitrary speed with a trade-off between speed and power demand, we show that the non-preemptive version of the problem is strongly NP-hard. We also present different approximation algorithms for this case, and test the computational performance of these approximation algorithms on randomly generated instances. In addition, for both the uniform-speed and speed-scaling cases, we show how to compute optimal schedules for the preemptive version of the problem in polynomial time. [ABSTRACT FROM AUTHOR]

Many different organic compounds may be converted by microbial biotransformation to high-value products for the chemical and pharmaceutical industries. This review summarizes the use of strains of Aspergillus niger, a well-known filamentous fungus used in numerous biotechnological processes, for biochemical transformations of organic compounds. The substrates transformed include monocyclic, bicyclic, and polycyclic aromatic hydrocarbons; azaarenes, epoxides, chlorinated hydrocarbons, and other aliphatic and aromatic compounds. The types of reactions performed by A. niger, although not unique to this species, are extremely diverse. They include hydroxylation, oxidation of various functional groups, reduction of double bonds, demethylation, sulfation, epoxide hydrolysis, dechlorination, ring cleavage, and conjugation. Some of the products may be useful as new investigational drugs or chemical intermediates. [ABSTRACT FROM AUTHOR]

Steroids are derivatives of the triterpenoid squalene, containing three fused cyclohexane rings and a cyclopentane ring, and flavonoids are derivatives of L-phenylalanine, containing two aromatic rings joined by a three-carbon bridge that may form part of a heterocyclic ring. A great variety of steroids and flavonoids are produced by plants, and many additional steroids are produced by animals or fungi. Because these compounds have many nutritional and pharmaceutical values, and many of them cannot be produced by chemical synthesis, biotechnological processes are being developed that use cultures of Aspergillus niger and other fungi to transform steroids and flavonoids to a variety of metabolites. These biochemical reactions, including hydroxylation, dehydrogenation, O-methylation, demethylation, cleavage of rings, epoxide hydrolysis, double bond reduction, and others, may be used for the production of higher-value compounds. [ABSTRACT FROM AUTHOR]

A minimal cell can be thought of as comprising informational, compartment-forming and metabolic subsystems. To imagine the abiotic assembly of such an overall system, however, places great demands on hypothetical prebiotic chemistry. The perceived differences and incompatibilities between these subsystems have led to the widely held assumption that one or other subsystem must have preceded the others. Here we experimentally investigate the validity of this assumption by examining the assembly of various biomolecular building blocks from prebiotically plausible intermediates and one-carbon feedstock molecules. We show that precursors of ribonucleotides, amino acids and lipids can all be derived by the reductive homologation of hydrogen cyanide and some of its derivatives, and thus that all the cellular subsystems could have arisen simultaneously through common chemistry. The key reaction steps are driven by ultraviolet light, use hydrogen sulfide as the reductant and can be accelerated by Cu(I)-Cu(II) photoredox cycling. [ABSTRACT FROM AUTHOR]