Prediction of integral type failures in semiconductor manufacturing through classification methods

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 prediction of ion-source tungsten filament breaks. The PdM has been tested on a real production dataset. © 2013 IEEE.

Insight Extraction for Semiconductor Manufacturing Processes

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.

On Regression Methods for Virtual Metrology in Semiconductor Manufacturing

Virtual metrology (VM) aims to predict metrology values using sensor data from production equipment and physical metrology values of preceding samples. VM is a promising technology for the semiconductor manufacturing industry as it can reduce the frequency of in-line metrology operations and provide supportive information for other operations such as fault detection, predictive maintenance and run-to-run control. The prediction models for VM can be from a large variety of linear and nonlinear regression methods and the selection of a proper regression method for a specific VM problem is not straightforward, especially when the candidate predictor set is of high dimension, correlated and noisy. Using process data from a benchmark semiconductor manufacturing process, this paper evaluates the performance of four typical regression methods for VM: multiple linear regression (MLR), least absolute shrinkage and selection operator (LASSO), neural networks (NN) and Gaussian process regression (GPR). It is observed that GPR performs the best among the four methods and that, remarkably, the performance of linear regression approaches that of GPR as the subset of selected input variables is increased. The observed competitiveness of high-dimensional linear regression models, which does not hold true in general, is explained in the context of extreme learning machines and functional link neural networks.

Microelectronic manufacturing of a microsensor based on polyaniline for ammonia gas sensing

The fabrication and operation of an ammonia chemoresistor is described. The sensor responds to changes in the resistance (impedance) of a thin layer of conductive polymer is due to changes in ammonia concentration. The polyaniline film was deposited by electroless plating (dipping) method on interdigitated array made by photolithographic technique. The PANI film was characterized by UV/VIS and IR Spectroscopy and respectively, Atomic Force Microscopy. Impedance Spectroscopy was used for sensor characterization

Unsteady tip leakage characteristics and heat transfer on turbine blade tip and casing

An unsteady numerical investigation was performed to examine time dependent behaviors of the tip leakage flow structures and heat transfer on the rotor blade tip and casing in a single stage gas turbine engine. A transonic, high-pressure
turbine stage was modeled and simulated using a stage pressure ratio of 3.2. The rotor’s tip clearance was 1.2 mm in height (3% of the rotor span) and its speed was set at 9500 rpm. Periodic flow is observed for each vane passing period. Tip leakage flow as well as heat transfer data showed highly time dependent behaviors. A stator trailing edge shock appears as the turbine stage is operating at transonic conditions. The shock alters the flow condition in the rotor section, namely, the tip leakage flow structures and heat transfer rate distributions. The instantaneous Nusselt number distributions are compared to the time averaged and steady-state results. The same patterns in tip leakage flow
structures and heat transfer rate distributions were observed in both unsteady and steady simulations. However, the unsteady simulation captured the locally time-dependent high heat transfer phenomena caused by the unsteady interaction with the upstream vane trailing-edge shock and the passing wake.

Effect of turbine inlet temperature on blade tip leakage flow and heat transfer

One of the most critical gas turbine engine components, rotor blade tip and casing, are exposed to high thermal load. It becomes a significant design challenge to protect the turbine materials from this severe situation. As a result of geometric complexity and experimental limitations, Computational Fluid Dynamics (CFD) tools have been used to predict blade tip leakage flow aerodynamics and heat transfer at typical engine operating conditions. In this paper, the effect of turbine inlet temperature on the tip leakage flow structure and heat transfer has been studied numerically. Uniform low (LTIT: 444 K) and high (HTIT: 800 K) turbine inlet temperature have been considered. The results showed the higher turbine inlet temperature yields the higher velocity and temperature variations in the leakage flow aerodynamics and heat transfer. For a given turbine geometry and on-design operating conditions, the turbine power output can be increased by 1.48 times, when the turbine inlet temperature increases 1.80 times. Whereas the averaged heat fluxes on the casing and the blade tip become 2.71 and 2.82 times larger, respectively. Therefore, about 2.8 times larger cooling capacity is required to keep the same turbine material temperature. Furthermore, the maximum heat flux on the blade tip of high turbine inlet temperature case reaches up to 3.348 times larger than that of LTIT case. The effect of the interaction of stator and rotor on heat transfer features is also explored using unsteady simulations.

Slow Release Drug Dissolution Profile Prediction in Pharmaceutical Manufacturing: a Multivariate and Machine Learning Approach

Slow release drugs must be manufactured to meet target specifications with respect to dissolution curve profiles. In this paper we consider the problem of identifying the drivers of dissolution curve variability of a drug from historical manufacturing data. Several data sources are considered: raw material parameters, coating data, loss on drying and pellet size statistics. The methodology employed is to develop predictive models using LASSO, a powerful machine learning algorithm for regression with high-dimensional datasets. LASSO provides sparse solutions facilitating the identification of the most important causes of variability in the drug fabrication process. The proposed methodology is illustrated using manufacturing data for a slow release drug.

A novel scalable manufacturing process for the production of hydrogel-forming microneedle arrays

A novel manufacturing process for fabricating microneedle arrays (MN) has been designed and evaluated. The prototype is able to successfully produce 14 × 14 MN arrays and is easily capable of scale-up, enabling the transition from laboratory to industry and subsequent commercialisation. The method requires the custom design of metal MN master templates to produce silicone MN moulds using an injection moulding process. The MN arrays produced using this novel method was compared with centrifugation, the traditional method of producing aqueous hydrogel-forming MN arrays. The results proved that there was negligible difference between either methods, with each producing MN arrays with comparable quality. Both types of MN arrays can be successfully inserted in a skin simulant. In both cases the insertion depth was approximately 60% of the needle length and the height reduction after insertion was in both cases approximately 3%.

A Eulogy to Seed Sharing: Manufacturing Genuine Scarcity

Seeds are traditionally considered as common or even public goods, their traits as ‘products of nature’. They are also essential to biodiversity, food security and food sovereignty. However, a suite of techno-legal interventions has legislated the enclosure of seeds: seed patents, plant variety protections, and stewardship agreements. These instruments create and protect private proprietary interests over plant material and point to the interface between seeds, capitalism, and law. In the following article, we consider the latest innovations, the bulk of which have been directed toward genetically disabling the reproductive capacities of seeds (terminator technology) or tying these capacities to outputs (‘round-up necessary’). In both instances, scarcity moves from artificial to real.
For the agro-industrial complex, the innovations are perfectly rational as they can simultaneously control supply and demand. For those outside the complex, however, the consequences are potentially ruinous. The practices of seed-saving and exchange no longer are feasible, even covertly. Contemporary genetic controls have upped the ante, by either disabling the reproductive capacity of seeds or, through cross-pollination and outcrossing, facilitating the autonomous spread of the genetic modifications that are importantly still traceable, identifiable and therefore capable of legal protection. In both instances, genuine scarcity becomes the new standard as private interests dominate what was a public sphere.

Manufacturing at double the speed

The speed of manufacturing processes today depends on a trade-off between the physical processes of production, the wider system that allows these processes to operate and the co-ordination of a supply chain in the pursuit of meeting customer needs. Could the speed of this activity be doubled? This paper explores this hypothetical question, starting with examination of a diverse set of case studies spanning the activities of manufacturing. This reveals that the constraints on increasing manufacturing speed have some common themes, and several of these are examined in more detail, to identify absolute limits to performance. The physical processes of production are constrained by factors such as machine stiffness, actuator acceleration, heat transfer and the delivery of fluids, and for each of these, a simplified model is used to analyse the gap between current and limiting performance. The wider systems of production require the co-ordination of resources and push at the limits of human biophysical and cognitive limits. Evidence about these is explored and related to current practice. Out of this discussion, five promising innovations are explored to show examples of how manufacturing speed is increasing—with line arrays of point actuators, parallel tools, tailored application of precision, hybridisation and task taxonomies. The paper addresses a broad question which could be pursued by a wider community and in greater depth, but even this first examination suggests the possibility of unanticipated innovations in current manufacturing practices.

DEVELOPING AN INTELLIGENT DIGITAL MOCK UP TO INTEGRATE DESIGN AND MANUFACTURING DISCIPLINES

This paper presents an approach to develop an intelligent digital mock-up (DMU) through integration of design and manufacturing disciplines to enable a better understanding of assembly related issues during design evolution. The intelligent DMU will contain tolerance information related to manufacturing capabilities so it can be used as a source for assembly simulations of realistic models to support the manufacturing decision making process within the design domain related to tolerance build ups. A literature review of the contributing research areas is presented, from which identification of the need for an intelligent DMU has been developed. The proposed methodology including the applications of cellular modelling and potential features of the intelligent DMU are presented and explained. Finally a conclusion examines the work to date and the future work to achieve an intelligent DMU.