Jacques Pinaton

A Literature Review on Sampling Techniques in Semiconductor Manufacturing

This paper reviews sampling techniques for inspection in semiconductor manufacturing. We discuss the strengths and weaknesses of techniques developed in the last last 20 years for excursion monitoring (when a process or machine falls out of specifications) and control. Sampling techniques are classified into three main groups: static, adaptive, and dynamic. For each group, a classification is performed per year, approach, and industrial deployment. A comparison between the groups indicates a complementarity strongly linked to the semiconductor environment. Benefits and drawbacks of each group are discussed, showing significant improvements from static to dynamic through adaptive sampling techniques. Dynamic sampling seems to be more appropriate for modern semiconductor plants.

Impact of control plan design on tool risk management: a simulation study in semiconductor manufacturing

In this paper, we analyze the impact of control plan design of defectivity inspections for tool risk management. Defectivity inspections are performed on products and can reveal the yield loss produced by contaminations or structural flaws. The risk considered in this paper concerns the exposure level of wafers on a tool between two defectivity controls. Our goal is to analyze how control plans can impact the manufacturing robustness from the point of view of wafer at risk on tools. A smart sampling strategy is considered for sampling lots to be measured. Actual data from the Rousset fab of STMicroelectronics are used. The simulation experiments are performed using the S5 Simulator developed by EMSE-CMP. Results show that not only the number and positions of controls operations have an important impact on tool risk management, but also how each control operation covers process operations.

Translation-Invariant Multiscale Energy-Based PCA for Monitoring Batch Processes in Semiconductor Manufacturing

The overwhelming majority of processes taking place in semiconductor manufacturing operate in a batch mode by imposing time-varying conditions to the products in a cyclic and repetitive fashion. These conditions make process monitoring a very challenging task, especially in massive production plants. Among the state-of-the-art approaches proposed to deal with this problem, the so-called multiway methods incorporate the batch dynamic features in a normal operation model at the expense of estimating a large number of parameters. This makes these approaches prone to overfitting and instability. Moreover, batch trajectories are required to be well aligned in order to provide the expected performance. To overcome these issues and other limitations of the conventional methodologies for process monitoring in semiconductor manufacturing, we propose an approach, translation-invariant multiscale energy-based principal component analysis, that requires a much lower number of estimated parameters. It is free of process trajectory alignment requirements and thus easier to implement and maintain, while still rendering useful information for fault detection and root cause analysis. The proposed approach is based on implementing a translation-invariant wavelet decomposition along the time series profile of each variable in one batch. The normal operational signatures in the time-frequency domain are extracted, modeled, and then used for process monitoring, allowing prompt detection of process abnormalities. The proposed procedure was tested with real industrial data and it proved to effectively detect the existing faults as well as to provide reliable indications of their underlying root causes.

Fault prognosis for Discrete Manufacturing Processes

Abstract This paper deals with a fault prognosis method, based on the extraction of a health indicator (HI) from a large amount of raw sensors data, applied to Discrete Manufacturing Processes (DMP). The HI is extracted by locating the significant points of machine which are related to the degradation. The dynamics of HI is then analysed and modelled using an appropriate stochastic process. The adaptive aspect of the prediction model allows the updating of the Remaining Usesul Life (RUL) estimation. The developed approach is applied on a real case provided by ST-Microelectronics, where experimental result shows its efficiency.

Efficient FDC based on hierarchical tool condition monitoring scheme

Tool condition evaluation and prognosis has been an arduous challenge in modern semiconductor manufacturing environment, especially for the foundry and analog companies with high product-mix and complicated technology nodes. More and more embedded and external sensors are installed to capture the genuine tool status for tool fault identification and, thus, tool condition analysis based on real-time equipment data becomes promising but also much more complex with the rapidly-increased number of sensors. In this paper, the feasibility of Generalized Moving Variance (GMV) technique is validated to consolidate the pure variations within tool Fault Detection and Classification (FDC) data into one indicator. Based on GMV, a hierarchical tool condition monitor scheme is developed by analyzing the GMV within functional clusters of sensors. With the introduction of this hierarchy, abnormal tool condition can be diagnosed and drilled down into sensor level for an efficient root cause analysis.

Health Index Extraction Methods for Batch Processes in Semiconductor Manufacturing

This paper deals with a study of three methods for health index (HI) extraction in semiconductor manufacturing equipments. The first method uses degradation reconstruction-based identification with basic principal component analysis (PCA), the second one uses multiway PCA and the last one extracts HI from the significant points related to degradation. A comparison of these methods are made discussing about their efficiency and shortcoming for the implementation. The studied methods are applied on two data sets: 1) a simulation case and 2) a real case provided by ST-Microelectronics, where experimental results highlight the advantages and limits of each one.

A semantic support to improve the collaborative control of manufacturing processes in industries

This paper put forwards how to improve the collaborative work of company engineers for the control of a manufacturing process. Sharing and using same manufacturing information in industries -such as control indicators- is necessary to efficiently ensure the control task. However, the generalization of Commercial Off-The-Shelf (COTS) systems in manufacturing companies to get such indicators has rapidly entailed the increase of the quantity of the digital resources, where a resource represents manufacturing data presented in specific formats. Retrieving such resources may become impossible without a semantic support. Our solution provides a semantic framework to enhance the retrieval of manufacturing information through the use of a business ontology and a domain-specific dictionary. The specificity of this business semantics is that it brings the resources closer to the business needs on one hand, and it unifies the business vocabulary in the company on other hand.

A Survey of Health Indicators and Data-Driven Prognosis in Semiconductor Manufacturing Process

Abstract Semiconductor device fabrication is considered today as one of the most complicated manufacturing process, characterized by an important complexity of production context, in an uncertain environment. To improve process efficiency and productivity, it is of prime importance for engineers to dispose of reliable indicators to drive decision-making on maintenance operations. To this end, terabytes of different data are collected during the manufacturing process, to feed statistical analysis tools and production management systems. Prognostics and health management (PHM) is defined as the discipline that links studies of failure mechanisms to system lifecycle management. Among the different approaches existing for prognosis, data-driven techniques learn models directly from monitored operational data related to system health. There is therefore a great interest in applying data-driven PHM methodologies to address semiconductor manufacturing issues. This paper surveys works on data-driven approaches for two issues of PHM methodologies with applications focused on semi-conductor manufacturing process: the development of indicators for health assessment, and prognostic methods.

A survey of semantic web standards to representing knowledge in problem solving situations

With the increase of information resources and software assets in industries, many companies seek today to improve knowledge sharing and the access to information. In fact, information retrieval has become a daily challenge for many actors of these companies where the quantity of information is huge and is from different sources. The semantic web field tries to address this problem by offering many techniques to attach semantic annotations to resources using the ontology of a given domain. This paper presents a survey of semantic web standards which aim at representing knowledge for many contexts of application. Our goal is to choose the most appropriate standard for representing knowledge in the context of problem solving.

A smart sampling scheduling and skipping simulator and its evaluation on real data sets

As modern manufacturing technology progresses, measurement tools become scarce resources since more and longer control operations are required. It thus becomes critical to decide whether a lot should be measured or not in order to get as much information as possible on production tools or processes, and to avoid ineffective measurements. To minimize risks and optimize measurement capacity, a smart sampling algorithm has been proposed to efficiently select and schedule production lots on metrology tools. This algorithm and others have been embedded in a simulator called “Smart Sampling Scheduling and Skipping Simulator” (S5). The characteristics of the simulator will be presented. Simulations performed on several sets of instances from three different semiconductor manufacturing facilities (or fabs) will be presented and discussed. The results show that, by using smart sampling, it is possible to drastically improve various factory performance indicators when compared to current fab sampling.