Mixed product run-to-run process control – An ANOVA model with ARIMA disturbance approach

A novel run-to-run control algorithm based on a dynamic analysis of variance (ANOVA) approach is proposed to deal with run-to-run (RtR) control of a high mixed operation, i.e., many different products are manufactured in many different tools. The conditions of different tools and products are identified based on the ANOVA analysis of the system output. A dynamic term in the form of an autoregressive integrated moving average (ARIMA) disturbance model is included in the process model to characterize the run-to-run disturbances such as drift, shift and/or some other unknown disturbances of different tools. It is shown from the study below that controller performance can be improved by introduction of the dynamic term, especially for products which are produced only occasionally. This makes it highly suitable for mixed product control system. An industrial example is also included to demonstrate superiority of this approach.     

Stability and performance analysis of mixed product run-to-run control

Run-to-run control has been widely used in batch manufacturing processes to reduce variations. However, in batch processes, many different products are fabricated on the same set of process tool with different recipes. Two intuitive ways of defining a control scheme for such a mixed production mode are (i) each run of different products is used to estimate a common tool disturbance parameter, i.e., a “tool-based” approach, (ii) only a single disturbance parameter that describe the combined effect of both tool and product is estimated by results of runs of a particular product on a specific tool, i.e., a “product-based” approach. In this study, a model two-product plant was developed to investigate the “tool-based” and “product-based” approaches. The closed-loop responses are derived analytically and control performances are evaluated. We found that a “tool-based” approach is unstable when the plant is non-stationary and the plant-model mismatches are different for different products. A “product-based” control is stable but its performance will be inferior to single product control when the drift is significant. While the controller for frequent products can be tuned in a similar manner as in single product control, a more active controller should be used for the infrequent products which experience a larger drift between runs. The results were substantiated for a larger system with multiple products, multiple plants and random production schedule.     

基于动态MANCOVA的混合制程虚拟测量算法

在晶圆/液晶面板等批次加工过程中,产品质量的及时估计与品质管制是提高产能和降低成本的有效途径.针对"少量多样"的混合制程,利用逐步回归算法挑选该制程的关毽变量,引入产品的效益因子,建立混合制程的虚拟测量模型;为克服系统扰动对模型精度的影响,以产品效益因子为状态量建立该制程的状态方程,利用Kalman滤波器递归估计模型参数得到动态的MANCOVA模型;最后通过某湿式蚀刻制程的工程应用验证了该算法的有效性.     

Closed-loop tool face control with the bit off-bottom

In this paper, we develop three methods to achieve reliable closed-loop, tool face control for directional drilling operations. This is a necessary step to achieve closed-loop, automated directional guidance. Our algorithms combine existing industry top-drive controllers with new control approaches. The torsional model we use for the drill string has been field validated and takes into account the Coulomb friction between the drill string and the borehole. These distributed friction terms are either assumed known (or measured) or can be estimated using a state-observer. In this work, we improve such a state-observer to obtain an estimation of the tool face orientation in real-time. We then propose different approaches to control the tool face. The first method is based on a feed-forward control law. It uses the flatness of the model and the estimation of the orientation to generate an admissible trajectory which is then tracked. In the second procedure, we require a stable rotation off-bottom before smoothly changing the reference to zero to stop bit rotation. This change of reference induces a change of orientation that can be estimated and finally compensated by repeating the procedure. Finally, the last method uses a series of trapezoidal setpoint inputs – bumps – to calculate the change in downhole tool face per change in surface orientation before arriving at the correct tool face after three iterations. These three algorithms are illustrated in simulations of field scenarios and their effectiveness and limitations, depending on the reliability and availability of downhole orientation data, are discussed.     

Drilling reconfigurable machine tool selection and process parameters optimization as a function of product demand

Special purpose machines (SPMs) are customized machine tools that perform specific machining operations in a variety of production contexts, including drilling-related operations. This research investigates the effect of optimal process parameters and SPM configuration on the machine tool selection problem versus product demand changes. A review of previous studies suggests that the application of optimization in the feasibility analysis stage of machine tool selection has received less attention by researchers. In this study, a simulated model using genetic algorithm is proposed to find the optimal process parameters and machine tool configuration. During the decision-making phase of machine tool selection, unit profit is targeted as high as possible and is given by the value of the following variables: SPM configuration selection, machining unit assignment to each operation group, and feed and cutting speed of all operations. The newly developed model generates any random chromosome characterized by feasible values for process parameters. Having shown how the problem is formulated, the research presents a case study which exemplifies the operation of the proposed model. The results show that the optimization results can provide critical information for making logical, accurate, and reliable decisions when selecting SPMs.     

A decision support system for automotive product planning and competitive market analysis

A decision support system (DSS) for automotive product marketing, design and manufacturing in China is presented in this paper. The DSS is developed as a tool to support product planning, competitive market analysis, supply chain analysis and subsequent manufacturing systems planning and deployment. The system consists of a number of automotive related databases which provide information about manufacturers' performance in each market segment as well as production information of all existing market players in the Chinese auto industry. Product planning, one of the key modules of the DSS prototype, is highlighted in this paper. It supports decision makers in determining suitable strategies for market entry by analyzing existing competitors' status, growth estimation of each market segment, and competitive market analysis for new vehicle products. A case study for new market entry is included here to demonstrate the feasibility and effectiveness of the proposed methodology.     

Multiparty quantum key agreement protocol based on locally indistinguishable orthogonal product states

Based on locally indistinguishable orthogonal product states, we propose a novel multiparty quantum key agreement (QKA) protocol. In this protocol, the private key information of each party is encoded as some orthogonal product states that cannot be perfectly distinguished by local operations and classical communications. To ensure the security of the protocol with small amount of decoy particles, the different particles of each product state are transmitted separately. This protocol not only can make each participant fairly negotiate a shared key, but also can avoid information leakage in the maximum extent. We give a detailed security proof of this protocol. From comparison result with the existing QKA protocols, we can know that the new protocol is more efficient.     

GAVEL - a new tool for genetic algorithm visualization

This paper surveys the state of the art in evolutionary algorithm visualization and describes a new tool called GAVEL. It provides a means to examine in a genetic algorithm (GA) how crossover and mutation operations assembled the final result, where each of the alleles came from, and a way to trace the history of user-selected sets of alleles. A visualization tool of this kind can be very useful in choosing operators and parameters and in analyzing how and, indeed, whether or not a GA works. We describe the new tool and illustrate some of the benefits that can be gained from using it with reference to three different problems: a timetabling problem, a job-shop scheduling problem, and Goldberg and Horn's long-path problem. We also compare the tool to other available visualization tools, pointing out those features which are novel and identifying complementary features in other tools     

Moving data window gradient‐based iterative algorithm of combined parameter and state estimation for bilinear systems

The combined iterative parameter and state estimation problem is considered for bilinear state‐space systems with moving average noise in this paper. There are the product terms of state variables and control variables in bilinear systems, which makes it difficult for the parameter and state estimation. By designing a bilinear state estimator based on the Kalman filtering, the states are estimated using the input‐output data. Furthermore, a moving data window (MDW) is introduced, which can update the dynamical data by removing the oldest data and adding the newest measurement data. A state estimator‐based MDW gradient‐based iterative (MDW‐GI) algorithm is proposed to estimate the unknown states and parameters jointly. Moreover, given the extended gradient‐based iterative (EGI) algorithm as a comparison, the MDW‐GI algorithm can reduce the impact of noise to parameter estimation and improve the parameter estimation accuracy. The numerical simulation examples validate the effectiveness of the proposed algorithm.     

Designing a product family of meshing tools

Applying software engineering concepts can improve the quality of any software development, and this is even more dramatic for complex, large and sophisticated software, such as meshing tools. Software product families are series of related products that make intensive reuse of already developed components. Object-oriented design promotes reusability, so it is specially suited for designing the structure of product families. In this paper we present an object-oriented design of a product family of meshing tools, where all family members share the software structure. By instantiating the structure with particular algorithms and parameters, we can easily produce different tools of the family. A good family design allows us not only to combine existing algorithms but also to easily incorporate new ones, improving software family evolution. We show how the family design is used for the generation of finite element and finite volume meshing tools, as well as a new tool for image processing.     

A knowledge-based tool for designing cyber physical production systems

Changing production systems and product requirements can trace their origin in volatile customer behaviour and evolving product requirements. This dynamic nature of customer requirements has been described as a constantly moving target, thus presenting a significant challenge for several aspects of product development. To deal with this constant and sometimes unpredictable product evolution, cyber physical production systems (CPPS) that employ condition monitoring, self-awareness and reconfigurability principles, have to be designed and implemented. This research contributes a CPPS design approach that proactively provides the required CPPS design knowledge. This approach aims to minimise or avoids future consequences and disruptions on the CPPS. This knowledge needs to be provided at the right time whilst not being intrusive to the production system designer’s cognitive activity. To effectively deal with the complexity of the cyber physical production system design activity with a manual method would lead to a time consuming, and complex support tool which is hard to implement, and difficult to use. The CPPS design approach has therefore been implemented in a prototype digital factory tool. This paper describes in detail the system requirements and system architecture for this tool. In order to establish the effectiveness of the proposed approach for designing cyber physical production systems, the prototype digital factory tool has been evaluated with a case study and a number of semi-structured interviews with both industrial and scientific stakeholders. The encouraging results obtained from this research evaluation have shown that such an approach for supporting the CPPS design activity makes stakeholders aware of their decision consequences and is useful in practice. This result can lead the way for the development and integration of such knowledge-based decision-making approaches within state-of-the-art digital factory and Computer Aided Engineering Design (CAED) tools.     

State estimation in high-mix semiconductor manufacturing

The traditional way of state estimation in semiconductor manufacturing, known as “threaded” state estimation, segregates the process data into different bins and uses the ones that match the current event of the specific context information (such as tools, layers, products) to update the process state. The limitation of threaded state estimation is that a narrowly defined process stream can result in too many different threads and insufficient data for each thread. This limitation becomes more severe in high-mix manufacturing, where there can be many products and many tools. Hence there is great interest in estimation methods that utilize all available data in the analysis. In this work, the characteristics inherent in state estimation of high-mix semiconductor manufacturing processes are analyzed, and a general framework is introduced for the non-threaded state estimation methods, i.e., state estimation without segregating the process data into different bins. The framework is based on the best linear unbiased estimate (BLUE) of a simplified stationary singular Gauss–Markov process, and non-threaded state estimation methods based on the Kalman filter, least squares and recursive least squares (RLS) are analyzed using the general framework. Simulation examples are presented to illustrate the equivalence between different algorithms. As real processes are rarely stationary, modifications to the Kalman filter and RLS are discussed. We show that in non-threaded state estimation, how to regulate the estimate covariance plays a significant role in estimation performance. To handle nonstationary disturbances that often occur in semiconductor processes, Bayesian-enhanced adaptive versions of the Kalman filter and RLS are proposed. Both simulated and industrial nonstationary processes are used to demonstrate the effectiveness of the proposed adaptive methods.     

融合贝叶斯网络与前景理论的产品工业设计多阶段决策方法#br#

针对产品工业设计决策中的不确定性与单一设计决策阶段难以准确描述全局决策结果的问题,引入三参数区间灰数对决策者的意见进行描述,构建贝叶斯网络(BN)模型学习用户群体对市场上现有成熟产品的决策信息,获得目标产品工业设计方案在各决策属性上的状态分布概率。为反映决策者对设计方案感知相对收益和损失的心理行为,融合前景理论(PT)与 BN 构建不同决策阶段产品工业设计方案的前景价值函数,以认知递进假设建立优化模型计算产品工业设计决策多阶段权重,通过综合前景价值计算判断设计方案优劣。以数控磨床工业设计方案决策的多阶段融合为例验证了方法的有效性,结果表明该方法能够引入用户群体的多阶段意见偏好估计设计决策属性的概率分布,以前景价值实现产品工业设计多阶段决策信息的有效集结,提高设计决策的全局性和科学性。     

Statistical process control to improve coding and code review

Software process comprises activities such as estimation, planning, requirements analysis, design, coding, reviews, and testing, undertaken when creating a software product. Effective software process management involves proactively managing each of these activities. Statistical process control tools enable proactive software process management. One such tool, the control chart, can be used for managing, controlling, and improving the code review process.     

含随机测量时延的EWMA批间控制器设计

批间控制(RtR)是半导体晶圆生产过程控制的有效算法. 然而, 受测量手段与测量成本的限制, 难以实时检 测晶圆的品质数据, 即: 存在一定的测量时延, 通常该测量时延是随机, 时变的, 且直接影响批间控制器的性能. 为 此, 本文基于指数加权移动平均(EWMA)算法, 提出一种含随机测量时延的扰动估计方法. 在分析测量概率的基础 上, 建立包含测量时延概率的扰动估计表达式; 并采用期望最大化(EM)算法估计该测量时延的概率; 然后分析系统 可能存在的静差项, 给出相应的补偿算法; 最后讨论系统的稳定性. 仿真实例验证所提算法的有效性.     

Domain modeling as a basis for building a meshing tool software product line

Meshing tools are highly complex software for generating and managing geometrical discretizations. Due to their complexity, they have generally been developed by end users – physicists, forest engineers, mechanical engineers – with ad hoc methodologies and not by applying well established software engineering practices. Different meshing tools have been developed over the years, making them a good application domain for Software Product Lines (SPLs). This paper proposes building a domain model that captures the different domain characteristics such as features, goals, scenarios and a lexicon, and the relationships among them. The model is partly specified using a formal language. The domain model captures product commonalities and variabilities as well as the particular characteristics of different SPL products. The paper presents a rigorous process for building the domain model, where specific roles, activities and artifacts are identified. This process also clearly establishes consistency and completeness conditions. The usefulness of the model and the process are validated by using them to generate a software product line of Tree Stem Deformation (TSD) meshing tools. We also present Meshing Tool Generator, a software that follows the SPL approach for generating meshing tools belonging to the TSD SPL. We show how an end user can easily generate three different TSD meshing tools using Meshing Tool Generator.     

Product identification in industrial batch fermentation using a variable forgetting factor

For reliable operation and the optimization of production, industrial fermentation processes require appropriate tools for monitoring the process in real time. This work presents the structure and operation of a soft sensor for the on-line monitoring of biomass and product concentration during salinomycin and bacitracin fermentation in an industrial, 80-m³ batch reactor; moreover it provides a tool for evaluation of batch production verified in industrial application. The process estimation algorithm consists of decoupled growth and product models, which ensures an unbiased convergence of the estimator and the robustness of the model. The production of secondary metabolites is described with a non-structured model upgraded with a variable forgetting factor that demonstrated a successful estimation of the non-measured parameters and states of this highly interactive and interlinked system with complex dynamics. The possibility of using various input signals in product identification yields independent soft sensors. This serves to improve the reliability of the predictions, mutual sensor control and enables the detection of irregularities in the fermentation process before the broth becomes useless.     

A systematic knowledge-based method for design of transformable product

A transformable product can perform different functions or change functionality by changing its physical structure. It is formed by integration of different components whose states can be transformed each other. However, there is a lack of systematic methods to guide design of the transformable product. In order to improve the design efficiency of transformable products, a large number of products are studied in this research to build a case base of transformation parameters and transformation principles for the design knowledge. A systematic design process is proposed to apply the design knowledge. The transformation design problem is first mapped from the problem domain to the knowledge domain expressed by transformation parameters. A general solution is then obtained in the knowledge domain. A multi-classification support vector machine is used to train a model of the transformation recommendation based on transformation parameters. Finally, the general solution is mapped into the problem domain for the specific solution using the analogy design. The effectiveness of the method is verified in the design of a self-propelled boom sprayer.     

Data management of green product development with generic modularized product architecture

Many manufacturers are facing a complex situation in the mixed production environment, in which green and non-green products are fabricated simultaneously. They are losing competitiveness as a downstream supplier due to lacking of a cost-effective approach to managing product variations compliant with different green directives. This paper presents a methodology based on generic modularized product architecture that facilitates data management of green product development. The four-level architecture allows one unified representation for multiple product models. An option control mechanism enables a quick generation of their BOMs (bills of material). A procedure consisting of seven steps is proposed to accomplish this. PDM functions are implemented to demonstrate the effectiveness of the methodology using a real LCD TV family as an example. This work complements the past studies on green product development, which mainly tackled the problem from design, process, and supply chain improvement. In contrast, from a management perspective, the proposed methodology provides a simple but useful tool for small-to-medium-sized enterprises (SMEs) to perform green product development in an economical manner.