Publications Lists     [ all BibTeX ]  

This paper aims to delineate a comprehensive method that integrates machine vision and deep learning for quality control within an industrial setting. The innovative approach proposed in this solution leverages a microservice architecture that ensures adaptability and flexibility to different scenarios while focusing on the employment of affordable, compact hardware, and achieves exceptionally high accuracy in performing the quality control task keeping a minimal computational time. Consequently, the developed system operates entirely on a portable smart camera, eliminating the need for additional sensors like photocells and external computation, which simplifies setup and commissioning phases and reduces the overall impact on the production line. By leveraging the integration of the embedded system with the machinery, this approach offers real-time monitoring and analysis capabilities, facilitating swift detection of defects and deviations from desired standards. Moreover, the low-cost nature of the solution makes it accessible to a wider range of manufacturing enterprises, democratizing quality processes in Industry 5.0. The system has been successfully implemented and is fully operational in a real industrial environment and the experimental results obtained from this implementation are also presented in the work.

Data-driven Anomaly Detection approaches have received increasing attention in  many other application areas in the past fewyears as a tool for monitoring of complex systems in addition to classical univariatecontrol charts. Tree-based approaches have proven to be particularly effective when dealing with high-dimensional Anomaly Detection problems and with underlying non-gaussian data distributions. The most popular approach in this family is the Isolation Forest, which is currently one of the mostpopular choices for scientists and practitioners when dealing with Anomaly Detection tasks. The Isolation Forest represents a seminal algorithm upon which many extended approaches have been presented in the past years aiming at improving the original method or at dealing with peculiar applicationscenarios. In this work, we revise some of the most popular and powerful Tree-based approaches to Anomaly Detection (extensions of the Isolation Forest and other approaches), considering both batch and streaming datascenarios. This work will review several relevant aspects of the methods, like computational costs and interpretability traits. To help practitioners, we also report available relevant libraries and open implementations and we review real-world industrial applications of the considered approaches.

The monitoring of rotating machinery is an essential task in today's production processes. Currently, several machine learning and deep learning-based modules have achieved excellent results in fault detection and diagnosis. Nevertheless, to further increase user adoption and diffusion of such technologies, users and human experts must be provided with explanations and insights by the modules. Another issue is related, in most cases, with the unavailability of labeled historical data that makes the use of supervised models unfeasible. Therefore, a new approach for fault detection and diagnosis in rotating machinery is here proposed. The methodology consists of three parts: feature extraction, fault detection and fault diagnosis. In the first part, the vibration features in the time and frequency domains are extracted. Secondly, in the fault detection, the presence of fault is verified in an unsupervised manner based on anomaly detection algorithms. The modularity of the methodology allows different algorithms to be implemented. Finally, in fault diagnosis, Shapley Additive Explanations (SHAP), a technique to interpret black-box models, is used. Through the feature importance ranking obtained by the model explainability, the fault diagnosis is performed. Two tools for diagnosis are proposed, namely: unsupervised classification and root cause analysis. The effectiveness of the proposed approach is shown on three datasets containing different mechanical faults in rotating machinery. The study also presents a comparison between models used in machine learning explainability: SHAP and Local Depth-based Feature Importance for the Isolation Forest (Local- DIFFI). Lastly, an analysis of several state-of-art anomaly detection algorithms in rotating machinery is included.

The diffusion of the Industry 4.0 paradigm lead to the creation and collection of huge manufacturing datasets; such datasets contain for example measurements coming from physical sensors located in different equipment or even in different productive manufacturing organizations. Such large and heterogeneous datasets represent a challenge when aiming for developing data-driven approaches like Anomaly Detection or Predictive Maintenance. In this work we present a new approach for performing Anomaly Detection that is able to handle heterogeneous data coming from different equipment, work centers or production sites.

Deep Learning approaches have revolutionized in the past decade the field of Computer Vision and, as a consequence, they are having a major impact in Industry 4.0 applications like automatic defect classification. Nevertheless, additional data, beside the image/video itself, is typically never exploited in a defect classification module: this aspect, given the abundance of data in data-intensive manufacturing environments (like semiconductor manufacturing) represents a missed opportunity. In this work we present a use case related to Scanning Electron Microscope (SEM) images where we exploit a  Bayesian approach to improve defect classification. We validate our approach on a real-world case study and by employing modern Deep Learning architectures for classification.

Visual inspection has recently gained increasing importance in the manufacturing industry and is often addressed by means of learning methodologies applied to data obtained from specific lighting and camera system setups. The industrial scenario becomes particularly challenging when the inspection regards reflective objects, which may affect both the data acquisition and the classification decision process, thus limiting the overall performance. In this context, we observe that the dynamics of the reflected light is the key aspect to characterize these surfaces and needs to be accurately exploited to improve the performances of the learning algorithms. To this aim, we propose a combined model-based and data-driven approach designed to detect defects on the reflective surfaces of industrial products, captured as video sequences under coaxial structured illumination. Specifically, a tunable spatial-temporal descriptor of the evolution of the reflected light (Dynamic Evolution of the Light, DEL) is designed and employed within a Hybrid Learning (HL) framework, where the learning process of a Convolutional Neural Network (CNN) is driven by the model-based descriptor. This approach is also extended by adopting the similar in nature descriptor Dynamic Image. The proposed HL solutions are validated against a whole spectrum of state-of-the-art learning procedures and different descriptors. Experiments run on a dataset coming from an actual industrial scenario confirm the ability of DEL to accurately characterize reflective surfaces and the validity of the HL method, which shows remarkably better performance in fault detection even with respect to modern 3D- CNNs with comparable computational effort.

The advent of the Industrial Internet of Things (IIoT) has led to the availability of huge amounts of data, that can be used to train advanced Machine Learning algorithms to perform tasks such as Anomaly Detection, Fault Classification and Predictive Maintenance. Even though not all pieces of equipment are equipped with sensors yet, usually most of them are already capable of logging warnings and alarms occurring during operation. Turning this data, which is easy to collect, into meaningful information about the health state of machinery can have a disruptive impact on the improvement of efficiency and up-time. The provided dataset consists of a sequence of alarms logged by packaging equipment in an industrial environment. The collection includes data logged by 20 machines, deployed in different plants around the world, from 2019-02-21 to 2020-06-17. There are 154 distinct alarm codes, whose distribution is highly unbalanced. This data can be used to address the following tasks: 

1. Next alarm forecasting: this problem can be framed as a supervised multi-class classification task, or a binary classification task when a specific alarm code is considered.
2. Predicting alarms occurring in a future time frame: here the goal is to forecast the occurrence of certain alarm types in a future time window. Since many alarms can occur, this is a supervised multi-label classification.
3. Future alarm sequence prediction: here the goal is predicting an ordered sequence of future alarms, in a sequence-to-sequence forecasting scenario.
4. Anomaly Detection: the task is to detect abnormal equipment conditions, based on the pattern of alarms sequence. This task can be either unsupervised, if only the input sequence is considered, or supervised if future alarms are taken into account to assess whether or not there is an anomaly.

 All of the above tasks can also be studied from a continual learning perspective. Indeed, information about the serial code of the specific piece of equipment can be used to train the model; however, a scalable model should also be easy to apply to new machines, without the need of a new training from scratch. The collection and release of this dataset has been supported by the Regione Veneto project PreMANI (MANIFATTURA PREDITTIVA: progettazione, sviluppo e implementazione di soluzioni di Digital Manufacturing per la previsione della Qualita e la Manutenzione Intelligente - PREDICTIVE MAINTENANCE: design, development and implementation of Digital Manufacturing solutions for the intelligent quality and maintenance systems).

The Industrial IoT (IIoT) paradigm represents an attractive opportunity for new generation measurement applications, which are increasingly based on efficient and reliable communication systems to allow the extensive availability of continuous data from instruments and/or sensors, thus enabling real-time measurement analysis. Nevertheless, different communication systems and heterogeneous sensors and acquisition systems may be found in an IIoT-enabled measurement application, so that solutions need to be defined to tackle the issue of seamless, effective, and low-latency interoperability. A significant and appropriate solution is the Open Platform Communications (OPC) Unified Architecture (UA) protocol, thanks to its object–oriented structure that allows a complete contextualization of the information. The intrinsic complexity of OPC UA, however, imposes a meaningful performance assessment to evaluate its suitability in the aforementioned context. To this aim, this paper presents the design of a general yet accurate and reproducible measurement setup that will be exploited to assess the performance of the main open source implementations of OPC UA. The final goal of this work is to provide a characterization of the impact of this protocol stack in an IIoT-enabled Measurement System, in particular in terms of both the latency introduced in the measurement process and the power consumption.

One of the main challenges in developing Machine Learning-based solutions for Semiconductor Manu-facturing is the high number of machines in the production and their differences, even when consideringchambers of the same machine; this poses a challenge in the scalability of Machine Learning-based so-lutions in this context, since the development of chamber-specific models for all equipment in the fab isunsustainable. In this work, we present a domain adaptation approach for Virtual Metrology (VM), one ofthe most successful Machine Learning-based technology in this context. The approach provides a commonVM model for two identical-in-design chambers whose data follow different distributions. The approach isbased on Domain-Adversarial Neural Networks and it has the merit of exploiting raw trace data, avoidingthe loss of information that typically affects VM modules based on features. The effectiveness of theapproach is demonstrated on real-world Etching.

Machine Learning-based Anomaly Detection (AD) approaches are efficient tools to monitor complexprocesses. One of the advantages of such approaches is that they provide a unique anomaly indicator,a quantitative index that captures the degree of ’outlierness’ of the process at hand considering possiblyhundreds or more variables at the same time, the typical scenario in semiconductor manufacturing. Oneof the drawback of such approaches is that Root Cause Analysis is not guided by the system itself. Inthis work, we show the effectiveness of a method, called DIFFI, to equip Isolation Forest, one of themost popular AD algorithms, with interpretability traits that can help corrective actions and knowledgeunderstanding. Such approach is validated on real world semiconductor manufacturing data related to aChemical Vapor Deposition process.

Solutions for the quality control of metallic surfaces are proposed. Specifically, we study a deflectometric apparatus based on coaxial structured light and the related algorithmic procedure, which is able to detect the faulty surface of a sample captured by a video sequence. First, by considering the metallic surface a dynamic scene illuminated under different light conditions, we develop the descriptor residuals of linear evolution of light (RLEL) that extracts the defectiveness information starting from the movement of the object without explicitly considering the physical characteristics of the light structure. Then, leveraging on RLEL, we present a hybrid learning (HL) technique capable of overcoming the data-driven approach used in classic deep learning (DL). By exploiting a multisteps training process, we combine the model-based descriptor RLEL and a classical data-driven convolutional neural network (CNN) to obtain an unconventional gray-box CNN, which exceeds the performance of popular DL solutions such as 3-D-inception and 3-D-residual DL networks. Remarkably, HL also shows its validity in comparing the performance of the same network structures trained not in a hybrid way, namely without the injection of the model-based information given by RLEL.

We propose a Deep Learning (DL)-based approach for production performance forecasting in fresh products packaging. On the one hand, this is a very demanding scenario where high throughput is mandatory; on the other, due to strict hygiene requirements, unexpected downtime caused by packaging machines can lead to huge product waste. Thus, our aim is predicting future values of key performance indexes such as Machine Mechanical Efficiency (MME) and Overall Equipment Effectiveness (OEE). We address this problem by leveraging DL-based approaches and historical production performance data related to measurements, warnings and alarms. Different architectures and prediction horizons are analyzed and compared to identify the most robust and effective solutions. We provide experimental results on a real industrial case, showing advantages with respect to current policies implemented by the industrial partner both in terms of forecasting accuracy and maintenance costs. The proposed architecture is shown to be effective on a real case study and it enables the development of predictive services in the area of Predictive Maintenance and Quality Monitoring for packaging equipment providers.

In modern manufacturing scenarios, detecting anomalies in production systems is pivotal to keep high-quality standards and reduce costs. Even in the Industry 4.0 context, realworld monitoring systems are often simple and based on the use of multiple univariate control charts. Data-driven technologies offer a whole range of tools to perform multivariate data analysis that allow to implement more effective monitoring procedures. However, when dealing with complex data, common data-driven methods cannot be directly used, and a feature extraction phase must be employed. Feature extraction is a particularly critical operation, especially in anomaly detection tasks, and it is generally associated with information loss and low scalability. In this paper we consider the task of Anomaly Detection with twodimensional, image-like input data, by adopting a Deep Learningbased monitoring procedure, that makes use of convolutional autoencoders. The procedure is tested on real Optical Emission Spectroscopy data, typical of semiconductor manufacturing. The results show that the proposed approach outperforms classical feature extraction procedures.

With the rise of Industry 4.0 and of the Industrial Internet, the computing and communication infrastructures achieved an essential role within process and factory automation, and cyberphysical systems in general. In this scenario, the OPC UA standard is currently becoming a widespread opportunity to enable interoperability among heterogeneous industrial systems. Nonetheless, OPC UA is characterized by a complex protocol architecture, that may impair the scalability of applications and may represent a bottleneck for its effective implementation in resource-constrained devices, such as low-cost industrial embedded systems. Several different OPC UA implementations are available, which in some significant cases are released under an open source license. In this context, the aim of this paper is to provide an assessment of the performance provided by some of these different OPC UA implementations, focusing specifically on potential development and resource bottlenecks. The analysis is carried out through an extensive experimental campaign explicitly targeting general purpose low-cost embedded systems. The final goal is to provide a comprehensive performance comparisons to allow devising some useful practical guidelines.

Industry 4.0 encapsulates methods, technologies, and procedures that transform data into informed decisions and added value inan industrial context. In this regard, technologies such as Virtual Metrology or Soft Sensing have gained much interest in thelast two decades due to their ability to provide valuable knowledge for production purposes at limited added expense. However,these technologies have struggled to achieve wide-scale industrial adoption, largely due to the challenges associated with handlingcomplex data structures and the feature extraction phase of model building. This phase is generally hand-engineered and basedon specific domain knowledge, making it time consuming, difficult to automate, and prone to loss of information, thus ultimatelylimiting portability. Moreover, in the presence of complex data structures, such as 2-dimensional input data, there are no establishedprocedures for feature extraction. In this paper, we present a Deep Learning approach for Virtual Metrology, called DeepVM,that exploits semi-supervised feature extraction based on Convolutional Autoencoders. The proposed approach is demonstratedusing a real world semiconductor manufacturing dataset where the Virtual Metrology input data is 2-dimensional Optical EmissionSpectrometry data. The feature extraction method is tested with different types of state-of-the-art autoencoder.

In the past recent years, Machine Learning methodologies have been applied in countless application areas. In particular, they play a key role in enabling Industry4.0. However, one of the main obstacles to the diffusion of Machine Learning-based applications is related to the lack of interpretability of most of these methods. In this work, we propose an approach for defining a ‘feature importance’ in Anomaly Detection problems. Anomaly Detection is an important Machine Learning task that has an enormous applicability in industrial scenarios. Indeed, it is extremely relevant for the purpose of quality monitoring. Moreover, it is often the first step towards the design of a Machine Learning based smart monitoring solutions because Anomaly Detection can be implemented without the need of labelled data. The proposed feature importance evaluation approach is designed for Isolation Forest, one of the most commonly used algorithm for Anomaly Detection. The efficacy of the proposed method is tested on synthetic and real industrial datasets.

With this work we report on the design, development and testing of near UV LED-based systems for oxygen gas sensing. The design and developed system is an optoelectronic setup based on 405 nm LEDs which excites and measures the photoluminescence emitted from a porphyrin based luminophor. By means of an accurate optical and optoelectronic setup, the system is able to operate without the need of avalanche photodiodes, thus resulting in a compact and low energy structure. The optical setup is specifically designed to maximize both the LED light exciting the luminophor and converted light acquired from the sensor.

Wireless networks are ever more deployed in industrial automation systems in various types of applications. A significant example in this context is represented by the transmission of safety data that, traditionally, was accomplished by wired systems. In this paper we propose an implementation of the Fail Safe over EtherCAT (FSoE) protocol on the top of IEEE 802.11 WLAN. The paper, after a general introduction of FSoE, focuses on the implementation of such protocol on commercial devices running UDP at the transport layer and connected via the IEEE 802.11 Wireless LAN. Then the paper presents some experimental setups and the tests that have been carried out on them. The obtained results are encouraging, since they show that good safety performance can be achieved even in the presence of wireless transmission media.

Exploiting the huge amount of data collected by industries is definitely one of the main challenges of the so-called Big Data era. In this sense, Machine Learning has gained growing attention in the scientific community, as it allows to extract valuable information by means of statistical predictive models trained on historical process data. In Semiconductor Manufacturing, one of the most extensively employed data-driven applications is Virtual Metrology, where a costly or unmeasurable variable is estimated by means of cheap and easy to obtain measures that are already available in the system. Often, these measures are multi-dimensional, so traditional Machine Learning algorithms cannot handle them directly. Instead, they require feature extraction, that is a preliminary step where relevant information is extracted from raw data and converted into a design matrix. Features are often hand-engineered and based on specific domain knowledge. Moreover, they may be difficult to scale and prone to information loss, affecting the effectiveness and maintainability of machine learning procedures. In this paper, we present a Deep Learning method for semi-supervised feature extraction based on Convolutional Autoencoders that is able to overcome the aforementioned problems. The proposed method is tested on a real dataset for Etch rate estimation. Optical Emission Spectrometry data, that exhibit a complex bi-dimensional time and wavelength evolution, are used as input.

In semiconductor manufacturing, metrology is generally a high cost, non-value added operation that impacts significantly on cycle time. As such, reducing wafer metrology continues to be a major target in semiconductor manufacturing efficiency initiatives. Data-driven spatial dynamic sampling methodologies are here compared. Such strategies aim at minimizing the number of sites that need to be measured across a wafer surface while maintaining an acceptable level of wafer profile reconstruction accuracy. The Spatial Dynamic Sampling approaches are based on analyzing historical metrology data to determine, from a set of candidate wafer sites, the minimum set of sites that need to be monitored in order to reconstruct the full wafer profile using statistical regression techniques. Spatial Dynamic sampling is then implemented in various strategies that guarantee coverage of all the possible sites in a given set of process iteration. In this way, the risk of not detecting previously unseen process behavior is mitigated. In this work, we demonstrate the efficacy of spatial dynamic sampling methodologies using both simulation studies and metrology data from a semiconductor manufacturing process.

In the context of Smart Monitoring and Fault Detection and Isolation in industrial systems, the aim of Predictive Maintenance technologies is to predict the happening of process or equipment faults. In order for a Predictive Maintenance technology to be effective, its predictions have to be both accurate and timely for taking strategic decisions on maintenance scheduling, in a cost-minimization perspective. A number of Predictive Maintenance technologies are based on the use of “health factors” quantitative indicators associated with the equipment wear that exhibit a monotone evolution. In real industrial environment, such indicators are usually affected by measurement noise and non-uniform sampling time. In this work we present a methodology, formulated as a stochastic filtering problem, to optimally predict the evolution of the aforementioned health factors based on noisy and irregularly sampled observations. In particular, a hidden Gamma process model is proposed to capture the nonnegativity and nonnegativity of the derivative of the health factor. As such filtering problem is not amenable to a closed form solution, a numerical Monte Carlo approach based on particle filtering is here employed. An adaptive parameter identification procedure is proposed to achieve the best trade-off between promptness and low noise sensitivity. Furthermore, a methodology to identify the risk function associated to the observed equipment based on previous maintenance data is proposed. The present study is motivated and tested on a real industrial Predictive Maintenance problem in semiconductor manufacturing, with reference to a dry etching equipment.

In semiconductor manufacturing, metrology is generally a high cost nonvalue-added operation that significantly impacts on cycle time. As such, reducing wafer metrology continues to be a major target in semiconductor manufacturing efficiency initiatives. A novel data-driven spatial dynamic sampling methodology is presented that minimizes the number of sites that need to be measured across a wafer surface while maintaining an acceptable level of wafer profile reconstruction accuracy. The methodology is based on analyzing historical metrology data using forward selection component analysis (FSCA) to determine, from a set of candidate wafer sites, the minimum set of sites that need to be monitored in order to reconstruct the full wafer profile using statistical regression techniques. Dynamic sampling is then implemented by clustering unmeasured sites in accordance with their similarity to the FSCA selected sites and temporally selecting a different sample from each cluster. In this way, the risk of not detecting previously unseen process behavior is mitigated. We demonstrate the efficacy of the proposed methodology using both simulation studies and metrology data from a semiconductor manufacturing process.

Wafer metrology is an expensive and time consuming activity in semiconductor manufacturing, but is essential to support advanced process control, predictive maintenance and other quality assurance functions. Keeping metrology to a minimum is therefore desirable. In the context of spatial sampling of wafers this has motivated the development of a number of data driven methodologies for optimizing wafer sampling plans. Two such methodologies are considered in this paper. The first combines Principal Component Analysis and Minimum Variance Estimation (PCA-MVE) to determine an optimum subset of sites from historical metrology data from a larger candidate set, while the second employs Forward Selection Component Analysis (FSCA), an unsupervised variable selection technique, to achieve the same result. We investigate the relationship between these two approaches and show that under specific conditions a regularized extension of FSCA, denoted FSCA-R, and PCA-MVE are equivalent. Numerical studies using simulated data verify the equivalence conditions. Results for simulated and industrial case studies show that the improvement in wafer profile reconstruction accuracy with regularization is not statistically significant for the case studies considered, and that when PCA-MVE is implemented with a denoising step as originally proposed, it is outperformed by FSCA. Therefore, FSCA is the preferred methodology.

Smart production monitoring is a crucial activity in advanced manufacturing for quality, control and maintenance purposes. Advanced Monitoring Systems aim to detect anomalies and trends; anomalies are data patterns that have different data characteristics from normal instances, while trends are tendencies of production to move in a particular direction over time. In this work, we compare state-of-the-art ML approaches (ABOD, LOF, onlinePCA and osPCA) to detect outliers and events in high-dimensional monitoring problems. The compared anomaly detection strategies have been tested on a real industrial dataset related to a Semiconductor Manufacturing Etching process

Advanced Monitoring Systems are fundamental in advanced manufacturing for control, quality and maintenance purposes. Nowadays, with the increasing availability of data in production and equipment, the need for high-dimensional Anomaly Detection techniques is thriving; anomalies are data patterns that have different data characteristics from normal production instances and that may be associated with faults or drifts in production. Tools for dealing with high-dimensional monitoring problems are provided by Machine Learning: in this paper, we test the performance of a state-of-the-art anomaly detection technique, called Isolation Forest, on a real industrial dataset related to Etching, one of the most important semiconductor manufacturing process. The monitoring has been performed exploiting Optical Spectroscopy Data.

Virtual Metrology is one of the most prominentAdvanced Process Control applications in SemiconductorManufacturing. The goal of Virtual Metrology is to provideestimations of quantities that are important for production andto assess process quality, but are costly or impossible to bemeasured. Virtual Metrology solutions are based on MachineLearning approaches. The bottleneck of developing VirtualMetrology solutions is generally the feature extraction phase thatcan be time-consuming, and can deeply affect the estimationperformance. In particular, in presence of data with additionaldimensions, such as time, feature extraction is typicallyperformed by means of heuristic approaches that may pickfeatures with poor predictive capabilities. In this work, wepropose the usage of modern Deep Learning approaches tobypass manual feature extraction and to provide highperformanceautomatic Virtual Metrology modules. Theproposed methodology is tested on a real industrial datasetrelated to Etching. The dataset at hand contains OpticalEmission Spectroscopy data and it is paradigmatic of the featureextraction problem under examination.

In this paper an approach to deal with Predictive Maintenance (PdM) problems with time-series data is discussed. PdM is a important approach to tackle maintenance and it is gaining an increasing attention in advanced manufacturing to minimize scrap materials, downtime, and associated costs. PdM approaches are generally based on Machine Learning tools that require the availability of historical process and maintenance data. Given the exponential growth in data logging in modern equipment, time series dataset are increasingly available in PdM applications. To exploit time series data for PdM, a functional learning methodology, namely Supervised Aggregative Feature Extraction (SAFE), is here employed on a semiconductor manufacturing maintenance problem.

In many applications, and especially thosewhere batch processes are involved, a target scalar outputof interest is often dependent on one or more time seriesof data. With the exponential growth in data logging inmodern industries such time series are increasingly availablefor statistical modeling in soft sensing applications. In orderto exploit time series data for predictive modelling, it isnecessary to summarise the information they contain as aset of features to use as model regressors. Typically thisis done in an unsupervised fashion using simple techniquessuch as computing statistical moments, principalcomponents or wavelet decompositions, often leading tosignificant information loss and hence suboptimal predictivemodels. In this paper, a functional learning paradigm isexploited in a supervised fashion to derive continuous,smooth estimates of time series data (yielding aggregatedlocal information), while simultaneously estimating a continuousshape function yielding optimal predictions. Theproposed Supervised Aggregative Feature Extraction (SAFE)methodology can be extended to support nonlinear predictivemodels by embedding the functional learning framework ina Reproducing Kernel Hilbert Spaces setting. SAFE has anumber of attractive features including closed form solutionand the ability to explicitly incorporate first and secondorder derivative information. Using simulation studies and apractical semiconductor manufacturing case study we highlightthe strengths of the new methodology with respect tostandard unsupervised feature extraction approaches.

In this paper a multiple classifier machine learning methodology for Predictive Maintenance (PdM) is presented. PdM is a prominent strategy for dealing with maintenance issues given the increasing need to minimize downtime and associated costs. One of the challenges with PdM is generating so called ’health factors’ or quantitative indicators of the status of a system associated with a given maintenance issue, and determining their relationship to operating costs and failure risk. The proposed PdM methodology allows dynamical decision rules to be adopted for maintenance management and can be used with high-dimensionaland censored data problems. This is achieved by training multiple classification modules with different prediction horizons to provide different performance trade-offs in terms of frequency of unexpected breaks and unexploited lifetime and then employing this information in an operating cost based maintenance decision system to minimise expected costs. The effectiveness of the methodology is demonstrated using a simulated example and a benchmark semiconductor manufacturing maintenance problem.

In semiconductor manufacturing, wafer quality control strongly relies on product monitoring and physical metrology. However, the involved metrology operations, generally performed by means of scanning electron microscopes, are particularly cost-intensive and time-consuming. For this reason, in common practice a small subset of only a productive lot is measured at the metrology stations and it is devoted to represent the entire lot. Virtual Metrology (VM) methodologies are able to obtain reliable predictions of metrology results at process time, without actually performing physical measurements; this goal is usually achieved by means of statistical models and by linking process data and context information to target measurements. Since semiconductor manufacturing processes involve a high number of sequential operations, it is reasonable to assume that the quality features of a given wafer (such as layer thickness and critical dimensions) depend on the whole processing and not on the last step before measurement only. In this paper, we investigate the possibilities to enhance VM prediction accuracy by exploiting the knowledge collected from previous process steps. We present two different schemes of multi-step VM, along with dataset preparation indications; special consideration will be reserved to regression techniques capable of handling high-dimensional input spaces. The proposed multi-step approaches are tested on production data provided by a partner semiconductor manufacturing industry. 

Slow release drugs must be manufactured to meettarget speci?cations with respect to dissolution curve pro?les.In this paper we consider the problem of identifying thedrivers of dissolution curve variability of a drug from historicalmanufacturing data. Several data sources are considered: rawmaterial parameters, coating data, loss on drying and pellet sizestatistics. The methodology employed is to develop predictivemodels using LASSO, a powerful machine learning algorithmfor regression with high-dimensional datasets. LASSO providessparse solutions facilitating the identi?cation of the most importantcauses of variability in the drug fabrication process.The proposed methodology is illustrated using manufacturingdata for a slow release drug.

Faulty operations of Heating, Ventilation and Air Conditioning (HVAC) chiller systems can lead to discomfort for the occupants, energy wastage, unreliability and shorter equipment life. Such faults need to be detected early to prevent further escalation and energy losses. Commonly, data regarding unforeseen phenomena and abnormalities are rare or are not available at the moment of HVAC systems installation: for this reason in this paper an unsupervised One-Class SVM classifier employed as a novelty detection system to identify unknown status and possible faults is presented. The approach, that exploits Principal Component Analysis to accent novelties w.r.t. normal operations variability, has been tested on a HVAC literature dataset.

Process monitoring and Predictive Maintenance (PdM) are gaining increasing attention in most manufacturing environments as a means of reducing maintenance related costs and downtime. This is especially true in industries that are data intensive such as semiconductor manufacturing. In this paper an adaptive PdM based flexible maintenance scheduling decision support system, which pays particular attention to associated opportunity and risk costs, is presented. The proposed system, which employs Machine Learning and regularized regression methods, exploits new information as it becomes available from newly processed components to refine remaining useful life estimates and associated costs and risks. The system has been validated on a real industrial dataset related to an Ion Beam Etching process for semiconductor manufacturing.

In the semiconductor manufacturing environment it is very important to understand which factors have the most impact on process outcomes and to control them accordingly. This is usually achieved through design of experiments at process start-up and long term observation of production. As such it relies heavily on the expertise of the process engineer. In this work, we present an automatic approach to extracting useful insights about production processes and equipment based on state-of-the-art Machine Learning techniques. The main goal of this activity is to provide tools to process engineers to accelerate the learning-by-observation phase of process analysis. Using a Metal Deposition process as an example, we highlight various ways in which the extracted information can be employed.

Classification methods with embedded feature selection capability are very appealing for the analysis of complex processes since they allow the analysis of root causes even when the number of input variables is high. In this work, we investigate the performance of three techniques for classification within a Monte Carlo strategy with the aim of root cause analysis. We consider the naive Bayes classifier and the logistic regression model with two different implementations for controlling model complexity, namely, a LASSO-like implementation with a l1 norm regularization and a fully Bayesian implementation of the logistic model, the so called relevance vector machine. Several challenges can arise when estimating such models mainly linked to the characteristics of the data: a large number of input variables, high correlation among subsets of variables, the situation where the number of variables is higher than the number of available data points and the case of unbalanced datasets. Using an ecological and a semiconductor manufacturing dataset, we show advantages and drawbacks of each method, highlighting the superior performance in term of classification accuracy for the relevance vector machine with respect to the other classifiers. Moreover, we show how the combination of the proposed techniques and the Monte Carlo approach can be used to get more robust insights into the problem under analysis when faced with challenging modelling conditions.

In semiconductor fabrication processes, effectivemanagement of maintenance operations is fundamental todecrease costs associated with failures and downtime. PredictiveMaintenance (PdM) approaches, based on statistical methodsand historical data, are becoming popular for their predictivecapabilities and low (potentially zero) added costs. We presenthere a PdM module based on Support Vector Machines forprediction of integral type faults, that is, the kind of failuresthat happen due to machine usage and stress of equipmentparts. The proposed module may also be employed as a healthfactor indicator. The module has been applied to a frequentmaintenance problem in semiconductor manufacturing industry,namely the breaking of the filament in the ion-source ofion-implantation tools. The PdM has been tested on a realproduction dataset.

In semiconductor manufacturing plants, monitoring physical properties of all wafers is crucial to maintain good yield and high quality standards. However, such an approach is too costly, and in practice, only few wafers in a lot are actually monitored. Virtual metrology (VM) systems allow to partly overcome the lack of physical metrology. In a VM scheme, tool data are used to predict, for every wafer, metrology measurements. In this paper, we present a VM system for a chemical vapor deposition (CVD) process. On the basis of the available metrology results and of the knowledge, for every wafer, of equipment variables, it is possible to predict CVD thickness. In this work, we propose a VM module based on least angle regression to overcome the problem of high dimensionality and model interpretability. We also present a statistical distance-based clustering approach for the modeling of the whole tool production. The proposed VM models have been tested on industrial production data sets.

Smart management of maintenances has become fundamental in manufacturing environments in order to decrease downtime and costs associated with failures. Predictive Maintenance (PdM) systems based on Machine Learning (ML) techniques have the possibility with low added costs of drastically decrease failures-related expenses; given the increase of availability of data and capabilities of ML tools, PdM systems are becoming really popular, especially in semiconductor manufacturing. A PdM module based on Classification methods is presented here for the prediction of integral type faults that are related to machine usage and stress of equipment parts. The module has been applied to an important class of semiconductor processes, ion-implantation, for the predictionof ion-source tungsten filament breaks. The PdM has been tested on a real production dataset.

Semiconductor fabrication involves several sequentialprocessing steps with the result that critical productionvariables are often affected by a superposition of affects overmultiple steps. In this paper a Virtual Metrology (VM) systemfor early stage measurement of such variables is presented;the VM system seeks to express the contribution to theoutput variability that is due to a defined observable partof the production line. The outputs of the processed systemmay be used for process monitoring and control purposes. Asecond contribution of this work is the introduction of ElasticNets, a regularization and variable selection technique for themodelling of highly-correlated datasets, as a technique for thedevelopment of VM models. Elastic Nets and the proposed VMsystem are illustrated using real data from a multi-stage etchprocess used in the fabrication of disk drive read/write heads.

Ion Implantation is one of the most sensitiveprocesses in Semiconductor Manufacturing. It consists inimpacting accelerated ions with a material substrate and isperformed by an Implanter tool. The major maintenanceissue of such tool concerns the breaking of the tungstenfilament contained within the ion source of the tool. Thiskind of fault can happen on a weekly basis, and theassociated maintenance operations can last up to 3 hours.It is important to optimize the maintenance activities bysynchronizing the Filament change operations with otherminor maintenance interventions. In this paper, a PredictiveMaintenance (PdM) system is proposed to tackle such issue;the filament lifetime is estimated on a statistical basisexploiting the knowledge of physical variables acting onthe process. Given the high-dimensionality of the data,the statistical modeling has been based on RegularizationMethods: Lasso, Ridge Regression and Elastic Nets. Thepredictive performances of the aforementioned regularizationmethods and of the proposed PdM module have beentested on actual productive semiconductor data.

Silicon Epitaxial Deposition is a process strongly influenced by wafer temperature behaviour, that has to be constantly monitored to avoid the production of defective wafers. However, temperature measurements are not reliable and the sensors have to be appropriately calibrated with some dedicated procedure. A Predictive Maintenance (PdM) System is proposed here with the aim of predicting process behaviour and scheduling control actions on the sensors in advance. Two different prediction techniques have been employed and compared: the Kalman predictor and the Particle Filter with Gaussian Kernel Density Estimator. The accuracy of the PdM module has been tested on real industrial production datasets.

Soft Sensors (SSs) are on-line estimators of “hardly to be measured” quantities of a process. The difficultyin measuring can be related to economic or temporal costs that cannot be afforded in a high-intensivemanufacturing production. In semiconductor manufacturing this technology goes with the name of Virtual Metrology(VM) systems. While a lot of efforts in research have been produced in the past years to identify the bestregression algorithms for these statistical modules, small amount of work has been done to develop algorithms fordata clustering of the entire production. This paper contains a new Information Theory-based approach to dataclustering for Virtual Metrology and Soft Sensors; the proposed algorithm allows to automatically split the datasetinto groups to be equally modeled. The proposed approach has been tested on real industrial dataset.

Virtual Metrology (VM) module have become popular in the past years and they are now widely adopted in the semiconductor plants. However, nowadays, still few works have been presented to deal with the interaction between VM and Run-to-Run (R2R), the most common control approach in the fabs. We present in this paper a new strategy to integrate VM with R2R based on Information Theory measure. The proposed control method penalizes statistical measure based on their statistical distance from the physical measure. This new approach also cope with the virtual loop control, where the R2R runs for several process iterations without in-situ measures, but based only on VM predictions. The results are compared with the actual state-of-the-art. 

Semiconductor manufacturing is one of the most technologically advanced industrial sectors. Process quality and control are critical for decreasing costs and increasing yield. The contribution of automatic control and statistical modeling in this area can drastically impact production performance. For this reason in the past decade major collaborative research projects have been undertaken between fab industries and academia in the areas of Virtual Metrology, Predictive Maintenance, Fault Detection, Run-to-Run control and modeling. In this paper we review some this research, discuss its impact on production and highlight current challenges. 

Many modeling problems require to estimate ascalar output from one or more time series. Such problemsare usually tackled by extracting a fixed number of featuresfrom the time series (like their statistical moments), with aconsequent loss in information that leads to suboptimal predictivemodels. Moreover, feature extraction techniques usuallymake assumptions that are not met by real world settings (e.g.uniformly sampled time series of constant length), and failto deliver a thorough methodology to deal with noisy data.In this paper a methodology based on functional learningis proposed to overcome the aforementioned problems; theproposed Supervised Aggregative Feature Extraction (SAFE)approach allows to derive continuous, smooth estimates oftime series data (yielding aggregate local information), whilesimultaneously estimating a continuous shape function yieldingoptimal predictions. The SAFE paradigm enjoys severalproperties like closed form solution, incorporation of first andsecond order derivative information into the regressor matrix,interpretability of the generated functional predictor and thepossibility to exploit Reproducing Kernel Hilbert Spaces settingto yield nonlinear predictive models. Simulation studies areprovided to highlight the strengths of the new methodology withrespect to standard unsupervised feature selection approaches.

In semiconductor manufacturing plants, monitoringphysical properties of all wafers is fundamental in order tomaintain good yield and high quality standards. However, suchan approach is too costly and in practice only few wafers in a lotare actually monitored. Virtual Metrology (VM) systems allowto partly overcome the lack of physical metrology. In a VMscheme, tool data are used to predict, for every wafer, metrologymeasurements. In this paper, we present a VM system for aChemical Vapor Deposition (CVD) process. On the basis ofthe available metrology results and of the knowledge, for everywafer, of equipment variables, it is possible to predict CVDthickness. In this work we propose a VM module based onLARS to overcome the problem of high dimensionality andmodel interpretability. The proposed VM models have beentested on industrial production data sets.

In semiconductor manufacturing, state of the art for wafer quality control relies on product monitoring and feedback control loops; the involved metrology operations, performed by means of scanning electron microscopes, are particularly cost-intensive and time-consuming. For this reason, it is not possible to evaluate every wafer: in common practice, a small subset of a productive lot is measured at the metrology station and devoted to represent the whole lot. Virtual Metrology (VM) methodologies are able to obtain reliable predictions of metrology results at process time, without actually performing physical measurements; this goal is usually achieved by means of statistical models, linking process data and context information to target measurements. Since semiconductor manufacturing processes involve a high number of sequential operations, it is reasonable to assume that the quality features of a certain wafer (such as layer thickness, critical dimensions, electrical test results) depend on the whole processing and not only on the last step before measurement. In this paper, we investigate the possibilities to improve the Virtual Metrology quality relying on knowledge collected from previous process steps. We will present two different scheme of multistep VM, along with dataset preparation indications; special consideration will be reserved to regression techniques capable of handling high dimensional input spaces. The proposed multistep approaches will be tested against actual data from semiconductor manufacturing industry. 

Epitaxy is a process strongly dependent onwafer temperature. Unfortunately, the performance ofthe pyrometers in charge of sensing wafer temperaturedeteriorate with the usage. This represents the majormaintenance issue for epitaxy process engineers who haveto frequently calibrate pyrometers emissivity coefficient. Atthe present state the change of the emissivity coefficient isheuristically based on fab tradition and process engineersexperience. We present a statistical tool to map therelationship between change in the temperature readingsand emissivity adjustments. The module has been testedon real industrial dataset.

Silicon Epitaxial Deposition is a process strongly influenced by wafer temperature behavior, that has to be constantly monitored to avoid the production of defective wafers. A Predictive Maintenance (PdM) System is here proposed with the aim of predicting process behavior and scheduling control actions in advance. Two different prediction techniques have been employed and compared: the Kalman predictor and the Particle Filter with Gaussian Kernel Density Estimator. The accuracy of the PdM module has been tested on real fab data. The proposed approach is flexible and can handle the presence of different recipes on the same equipment.

In semiconductor manufacturing plants, monitoring of all wafers is fundamental in order to maintain good yield and high quality standards. However, this is a costly approach and in practice only few wafers in a lot are actually monitored. With a Virtual Metrology (VM) system it is possible to partly overcome the lack of physical metrology. In a VM scheme, tool data are used to predict, for every wafer, metrology measurements. In this paper, we present a VM system for a Chemical Vapor Deposition (CVD) process. Various data mining techniques are proposed. Due to the huge fragmentation of data derived from CVD's mixed production, several kind of data clustering have been adopted. The proposed models have been tested on real productive industrial data sets.