The 1st Sino-US Research Conference on Quality, Analytics and Innovations

首届中美质量创新高峰论坛

Agenda

日程安排

June 30th , 2016

Time

时间

Event

事件

Venue (Map Illustration)

地点（地图图例标

记）

Others

备注

16:00-20:00 Registration Open

注册签到

First Floor of Antai College of

Economics and Management(A)

安泰经管学院一楼大厅

(A)

Refreshment Provided

茶歇点心供应

July 1st , 2016

Time

时间

Event

事件

Speakers

嘉宾

Chair

主持

Venue (Map Illustration)

地点（地图图例

标记）

7:30-8:30 Registration Opens

注册签到

A507, Antai College of

Economics and Management (A)

9:00-9:20 Opening Speech

开幕、嘉宾致辞

Prof. Yuan Li, Shanghai Jiao Tong University Prof. Kwok-Leung Tsui, City University of Hong Kong Prof. William H. Woodall, VirginiaTech

Wei Jiang

9:20-10:20 Keynote Speech

主题报告 Prof. Jeff Wu, Georgia Institute of Technology Kwok-Leung Tsui

10:20-10:30 Group Photo

合影

First Floor, Front Gate of Antai College of Economics and Management(A)

安泰经管学院一楼大门口（A）

10:30-10:40 Coffee Break

茶歇

Outside of A507 of Antai College of Economics and Management(A)

安泰经管学院 A507 教室门口

10:40-11:00 Keynote Speech 主题报告

Prof. Zuoyi Liu, NSFC Wei Jiang

A507 (A)

11:00-12:00 Keynote Speech 主题报告

Prof. Jay Lee, University of Cincinnati William Woodall

12:00-13:30 Lunch Buffet

午餐

Second Floor of Graduate Dining Hall (I)

研究生食堂二楼自助餐 (I)

13:30-15:30

Invited 1: SPC for Monitoring

Complex Systems

特邀报告 1

[1]. Fugee Tsung, Hong Kong University of Science and

Technology

[2]. William H. Woodall, Virginia Tech

[3]. Changliang Zou, Nankai University

[4]. Xinwei Deng, Virginia Tech

Changliang Zou A507 (A)

15:30-15:50 Coffee Break

茶歇

Outside of A507 of Antai College of Economics and Management(A)

安泰经管学院 A507 教室门口

15:50-17:50

Invited 2: DOE and Computer

Experiments

特邀报告 2

[1]. William Li, University of Minnesota

[2]. Rui Tuo, Chinese Academy of Sciences

[3]. Mei Han, City University of HK

[4]. Matthias Tan, City University of HK

Matthias Tan A507 (A)

July 2nd , 2016

Time

时间

Event

事件

Speakers

嘉宾

Chair

主持

Venue (Map Illustration)

地点（地图图例

标记）

8:00-10:00 Invited 3: Systems Reliability

and Prognostics

[1]. Yimeng Xie, AstraZeneca GMD China

[2]. Lirong Cui，Beijing Institute of Technology David Coit A507 (A)

特邀报告 3 [3]. David Coit, Rutgers University

[4]. Shubin Si, Northwestern Polytechnical University

10:00-10:20 Coffee Break

茶歇

Outside of A507 of Antai College of Economics and Management(A)

安泰经管学院 A507 教室门口

10:20-12:20

Invited 4: Big Data Analytics

and Simulation Analytics

特邀报告 4

[1]. Zhiqiang Zheng, University of Texas, Dallas

[2]. Nan Chen, National University of Singapore

[3]. Zhaolin Hu, Tongji University

[4]. Jie Song, Peking University

Zhaolin Hu A507 (A)

12:20-14:00 Lunch Buffet

午餐

Second Floor of Graduate Dining Hall (I)

研究生食堂二楼自助餐 (I)

14:00-15:40

Parallel 5-A

Reliability

分论坛 5-A

Xuejun Liu, Beihang University

Xue Jing, Beihang University

Juan Wang, Nanjing Polytechnic University

Yanjing Zhang, Nanjing Polytechnic University

Yiwen Zhang, Tianjin University

Shubin Si S301, Xin Shang Yuan 新上院三

楼 （B）

Parallel 5-B

Modeling

分论坛 5- B

Di Wang, Peking University

Lisha Yu, City University of Hong Kong

Jing Li, Zhejiang University

Hongyan Wang, Tufts University

Qiang Zhou, City University of Hong Kong

Qiang Zhou S302, Xin Shang Yuan 新上院三

楼 （B）

Parallel 5-C

SPC

Yanting Li, Shanghai Jiao Tong University

Jian Li, Xi’an Jiao Tong University Tingting Huang S303, Xin Shang

Yuan 新上院三

分论坛 5- C Dong Ding, Xi'an Polytechnic University

Caiwen Zhang, Sun Yat-sen University

Tingting Huang, Beihang University

楼 （B）

15:40-16:00 Coffee Break 茶歇

Outside of S301/S302/S303, Xin Shang Yuan (B)

新上院三楼 S301/S302/S303 教室门口 (B)

16:00-17:40

Parallel 6-A

Reliability

分论坛 6-A

Hong-gen Chen, Zhengzhou University of

Aeronautics

Guodong Wang, Zhengzhou University of

Aeronautics

Xinghong Qin, Tongji University

Jiaheng Shen, Beihang University

Lirong Cui S301, Xin Shang Yuan 新上院三

楼 (B）

Parallel 6-B

DOE

分论坛 6-B

Jai-Hyun BYUN, Gyeongsang National University

Zhaojun Hao, Beihang University

Yanrong Li, Tianjin University

Guilin Li, National University of Singapore

Guilin Li S302, Xin Shang Yuan 新上院三

楼 (B）

Parallel 6-C

SPC

分论坛 6-C

Wenjuan Liang, East China Normal University

Chunjie Wu, Shanghai University of Economics and

Finance

Xue Li, Zhengzhou university of aeronautics

Qijun Zhong, Tianjin University

Chunjie Wu

S303, Xin Shang Yuan 新上院三

楼（B）

 

Conference Co-Chairs

Wei Jiang Shanghai Jiao Tong University

Wei JIANG is the head and DistinguishedProfessor of Operations Management at Antai College of Economics and Management, Shanghai Jiao Tong University. Prior to joining Shanghai Jiao Tong University, he worked in AT&T Labs, Stevens Institute of Technology, and Hong Kong University of Science and Technology for 15 years. His research interests include big data analytics and innovation, Industry 4.0 and quality management, logistics and operations management, information finance and risk management, etc. He was a recipient of NSF CAREER award in 2006 and China National Funds for Distinguished Young Scientists award from NSFC in 2013.He was recognized as an expert of the “1000 Talents Program” in 2012 and “Program of Shanghai Subject Chief Scientist” in 2015 by Shanghai Municipal government.He is currently serving as Associate Editor for Naval Research Logistics, IIE Transactions, Decision Sciences Journal, etc.

 

Kwok Leung TSUI City University of Hong Kong

Professor Kwok L. Tsui is Head and Chair Professor of Industrial Engineering in the Department of Systems Engineering and Engineering Management at the City University of Hong Kong, and the founder and Director of Center for Systems Informatics Engineering. Prior to joining City University of Hong Kong, Professor Tsui was Professor at the School of Industrial and Systems Engineering at the Georgia Institute of Technology. Professor Tsui was a recipient of the National Science Foundation Young Investigator Award. He is Fellow of the American Statistical Association, American Society for Quality, and International Society of Engineering Asset Management; a U.S. representative to the ISO Technical Committee on Statistical Methods. Professor Tsui was Chair of the INFORMS Section on Quality, Statistics, and Reliability and the Founding Chair of the INFORMS Section on Data Mining. Professor Tsui’s current research interests include data mining, surveillance in healthcare and public health, prognostics and systems health management, calibration and validation of computer models, process control and monitoring, and robust design and Taguchi methods.

William H. Woodall, Virginia Polytechnic Institute and State University

William H. Woodall is a Professor of Statistics at Virginia Tech. He is a former editor of the Journal of Quality Technology (2001–2003) and associate editor of Technometrics (1987–1995). He has published over 130 papers, most on various aspects of process monitoring. He is the recipient of the ASQ Shewhart Medal (2002), ENBIS Box Medal (2012), Jack Youden Prize (1995, 2003), ASQ Brumbaugh Award (2000, 2006), Ellis Ott Foundation Award (1987), Soren Bisgaard Award (2012), Lloyd S. Nelson Award (2014), and a best paper award for IIE Transactions on Quality and Reliability Engineering (1997). He is a Fellow of the American Statistical Association, a Fellow of the American Society for Quality, and an elected member of the International Statistical Institute. 

 

Program Co-Chairs

 

Wenhui Zhao Shanghai Jiao Tong University  

Wenhui Zhao is currently an associate professor at Antai College of Economics and Management, Shanghai Jiao Tong University. He received his PhD in Operations Research from The Ohio State University, and a Bachelor's degree from Tsinghua University. His research interests include Supply Chain Management /Finance, Combinatorial Optimization, Game Theory in Economics, etc. His work has been published in Journals such as Management Science, Operations Research, Mathematical Programming, Production and Operations Management, and so on.    

 

Kaibo Wang Tsinghua University  

Kaibo Wang is a Professor in the Department of Industrial Engineering, Tsinghua University, Beijing, China. He received his Ph.D. in IndustrialEngineering and Engineering Management from the Hong Kong University of Science and Technology, Hong Kong. His research focuses on industrial data analytics and statistical quality control, with a special emphasis on the integration of engineering knowledge and statistical theories for solving problems from real industries.

 

 

 

 

 

Keynote Speech 1

Short-Bio  

C. F. Jeff Wu is Professor and Coca Cola Chair in Engineering Statistics at the School of Industrial and Systems Engineering, Georgia Institute of Technology. He was the first academic statistician elected to the National Academy of Engineering (2004); also a Member (Academician) of Academia Sinica (2000). A Fellow of American Society for Quality, Institute of Mathematical Statistics, of INFORMS, and American Statistical Association. He received the COPSS (Committee of Presidents of Statistical Societies) Presidents’ Award in 1987, the COPSS Fisher Lecture Award in 2011, the Deming Lecture Award in 2012, and numerous other awards and honors. He has published more than 165 research articles and supervised 45 Ph.D.'s. He has published two books "Experiments: Planning, Analysis, and Parameter Design Optimization" (with Hamada) and “A Modern Theory of Factorial Designs” (with Mukerjee).  

Speech Title:

Quality improvement: from autos and chips to nano and bio

Abstract: Quality improvement (QI) has a glorious history, starting from Shewhart’s path-breaking work on statistical process control to Deming’s high-impact work on quality management. Statistical concepts and tools played a key role in such work. As the applications became more sophisticated, elaborate statistical methods were required to tackle the problems. In the last three decades, QI has seen more use of experimental design and analysis, particularly the methodology of robust parameter design (RPD). I will first review some major ideas in RPD, focusing on its engineering origin and statistical methodology. I will then discuss more recent work that expands the original approach, including the use of feedback control and operating window. To have an effective solution, the subject matter knowledge often needs to be incorporated. Techniques for fusing data with knowledge will be presented. For advanced manufacturing and high-tech applications, there are new challenges and possible paradigm shift posed by three features: large varieties, small volume and high added value. I will speculate on some new directions and technical development. Throughout the talk, the ideas will be illustrated with real examples, ranging from the traditional (autos and chips) to the modern (nano and bio).

Keynote Speech 2

Short Bio Prof. Zuoyi LIU, Division Head, Division of Management Science and Engineering, Department of Management Sciences, National Natural Science Foundation of China Speech Title:

NSFC’s supporting for the research of big data

Keynote Speech 3

Short Bio Dr. Jay Lee is Ohio Eminent Scholar, L.W. Scott Alter Chair Professor,

and Distinguished Univ. Professor at the Univ. of Cincinnati and is founding director of National Science Foundation (NSF) Industry/University Cooperative Research Center (I/UCRC) on Intelligent Maintenance Systems (www.imscenter.net) He is one of the pioneers in the field of Prognostics and Health Management (PHM) and has mentored his students and won 1st prize of PHM Data Challenges five times since 2008. He also mentored his students and developed a spin-off company Predictronics through NSF ICorps Award in 2012. He serves as Committee member for White House Cyber Physical Systems (CPS) American Challenge Program in Dec. 3013, a member of Technical Executive committee (TEC) of Digital Manufacturing and Design Innovation (DMDI) in Feb. 2014, as well as a member of Leadership Council of MForesight which is a NSF/NIST Newly established manufacturing think tank in Sept. 2015. He also serves as

honorary professor and visiting professor for a number of institutions including Shanghai Jiao Tong Univ., Cranfield Univ. in UK, Lulea Univ. of Technology in Sweden, etc. In addition, he serves as editors and associate editor for a number of journals including IEEE Transaction on Industrial Informatics, Int. Journal on Prognostics & Health Management (IJPHM), Int. Journal on Service Operations and Informatics, etc. He is a Fellow of ASME, SME, as well as a founding fellow of International Society of Engineering Asset Management. He has received a number of awards including the most recent Prognostics Innovation Award at NI Week by National Instruments in 2012 and NSF Alex Schwarzkopf Technological Innovation Prize and MFPT (Machinery Failure Prevention Technology Society) Jack Frarey Award in 2014. In 1994, he received President Clinton’s Appreciation Letter for his participation and contribution to the United States Partnership for Next Generation Vehicle (PNGV) Program. He is also a honorary advisor to the Heifer International-a charity organization working to end hunger and poverty around the world by providing

livestock and training to struggling communities. Speech Title:

Industrial Big Data Analytics for Smart Maintenance & Service Innovation

Abstract: In today’s competitive business environment, companies are facing challenges in dealing with big data issues for rapid decision making for improved productivity. Many manufacturing systems are not ready to manage big data due to the lack of smart analytics tools. Germany is leading a transformation toward 4th Generation Industrial Revolution (Industry 4.0) based on Cyber-Physical System (CPS)-enabled maintenance and service innovation. As more software and embedded intelligence are integrated in industrial products and systems, predictive technologies can further intertwine intelligent algorithms with electronics and tether-free intelligence to predict product performance degradation and autonomously manage and optimize product service needs,

The presentation will address the trends of industrial transformation in big data environment as well as the readiness of smart predictive informatics tools to manage big data to achieve resilient product life cycle management. First, industry transformation including Germany Industry 4.0, Industrial Internet, and Cyber-Physical System (CPS) will be introduced. Second, advanced predictive analytics technologies for smart product manufacturing and service systems with case studies will be presented. In addition, research advances in designing cyber-physical model for smart product maintenance & service systems with many case studies will be discussed. Finally, Dominant Innovation® methodology for service innovation will be discussed.

Invited Sessions

Invited Session 1: SPC for Monitoring Complex Systems  

[1]. Fugee Tsung,

Professor, Department of Industrial Engineering and Logistics Management, Hong Kong University of Science and Technology, season@ust.hk

Title: Quality Engineering faces the challenges of big data and little data Abstract: This talk will present and discuss the challenges and opportunities that quality engineers face in the era of big data. The ability to separate signal and noise in the data-rich-information-poor environment would be the key. The second part of the talk will present and discuss the challenges and opportunities that quality engineers face in the era of additive manufacturing (i.e., 3D printing), where there is little data due to its one-of-a-kind nature. For example, statistical quality control (SPC) originated from mass production cannot be applied directly because such a small or single lot production does not have repeated measures of the same kind.

[2]. William H. Woodall,

Professor, Department of Statistics, Virginia Polytechnic Institute and State University, bwoodall@vt.edu

Title: An Overview and Perspective on Social Network Monitoring Abstract: An Overview and Perspective on Social Network Monitoring In this introductory presentation, I give an overview of some basic social network ideas and analysis with some practical examples. I describe some statistical methods for the modeling and monitoring of social networks. I discuss the advantages and limitations of some methods as well as some relevant sampling and design issues. This is a very promising research area, so some research ideas will be discussed. Reference: Woodall, W. H., Zhao, M., Paynabar, K., Sparks, R., and Wilson, J. D. (2016). “An Overview and Perspective on Social Network Monitoring”, To appear in IIE Transactions.

[3]. Changliang Zou,

Professor, Institute of statistics, Nankai University, nk.chlzou@gmail.com

Title: Statistical Monitoring of High-Dimensional Data streams Abstract: Monitoring high-dimensional data streams has become increasingly important for real-time detection of abnormal activities in many applications. We propose a test statistic which is based on the "divide-and-conquer" strategy, and integrate this statistic into the multivariate EWMA charting scheme for on-line detection. The key idea is to combine many T-square statistics calculated on low-dimensional sub-vectors. The proposed procedure is computation- and storage-efficient. The control limit is obtained through the asymptotic distribution of the test statistic under some mild conditions on the dependence structure. Both (asymptotically) theoretical analysis and numerical results show that the proposed method behaves well in high-dimensional data.

[4]. Xinwei Deng,

Assistant Professor, Department of Statistics, Virginia Polytechnic Institute and State University, xdeng@vt.edu

Title: A Latent Process Approach for Change Point Detection of Mixed-Type Observations Abstract: Mixed-type observations, such as continuous quality measurement, discrete counting, and binary outcome, are commonly present in many applications. The change point detection with mixed-type observations encounters great challenge on how to quantify the hidden association among mixed-type observations. In this work, we proposed a latent process approach to jointly modeling the mixed observation, and effectively detecting the changes. The proposed method combines Discrete Particle Filter (DPF) & Sequential Monte Carlo (SMC) algorithm for parameter estimation and Bayesian inference. The performance of the proposed method is illustrated by several numerical examples. The proposed methodology is also demonstrated using a civil unrest data provided by ICEWS (Integrated Conflict Early Warning System) and GDELT (Global Database of Events, Language, and Tone). This is a joint work with Shuyu Chu at Virginia Tech.

Invited Session 2: DOE and Computer Experiments  

[1]. William Li,

Professor, Department of Supply Chain and Operations, Carlson School of Management, University of Minnesota, Minneapolis, MN 55455, wli@umn.edu

Title: An Easy-To-Implement Variable Selection Method For Models Following Heredity Abstract: In many practical regression problems, it is desirable to select important variables with heredity constraint satisfied. In other words, when an interaction term is selected, it is preferred to select all the corresponding main effects as well. In this paper, we propose a general strategy to maintain heredity in variable selection through a novel heredity-induced data standardization. After the standardization, any variable selection method (including stepwise selection, lasso, SCAD and others) can be applied and the selected model is automatically guaranteed to satisfy the heredity constraint. Furthermore, the same procedure works for all types of regression including linear regression, generalized linear regression and regression with censored outcome. Therefore, our proposed strategy is easy (almost effortless) to implement in practice to maintain the heredity. Simulations and real examples are used to illustrate the merits of the proposed methods.

[2]. Rui Tuo,

Assistant Professor, Department of Statistical Science, Institute of Systems Sciences, Chinese Academy of Sciences, tuorui@amss.ac.cn

Title: Prediction based on the Kennedy-O’Hagan calibration model: asymptotic consistency and other properties Abstract: Kennedy and O’Hagan (2001) proposes a model dedicated to calibrating the unknown parameters in a computer model and estimating the discrepancy between the computer output and physical response. This model is known to have certain identifiability issue. Tuo and Wu (2015) shows that there exist examples that Kennedy-O’Hagan’s method renders unreasonable results in calibration. In spite of its unstable performance on calibration, the Kennedy-O’Hagan’s approach has a more robust behavior in predicting the physical response. In this work, we conduct some theoretical analysis to show the consistency of the predictor based on the Kennedy-O’Hagan’s calibration model under a radial-basis-function context. The results are demonstrated with numerical examples. This is a joint work with C.F. Jeff Wu

[3]. Mei Han,

PhD Students of Dr. TAN Hwai Yong Matthais, City University of HK

Title: Integrated Parameter and Tolerance Design with Computer Experiments Abstract: Robust parameter and tolerance design are effective methods to improve process quality. Li and Wu (1999), demonstrate that the traditional two-stage approach that performs parameter design followed by tolerance design to reduce the sensitivity to variations of input characteristics is suboptimal. To mitigate the problem, they propose an integrated parameter and tolerance design (IPTD) methodology that is suitable for linear models. In this talk, a computer-aided integrated parameter and tolerance design approach for computer experiments is proposed in which the means and tolerances of input characteristics are simultaneously optimized to minimize the total cost. A Gaussian process metamodel is used and a multiobjective optimization approach is proposed to find robust optimal solutions. This is a joint work with Matthias Hwai Yong Tan

[4]. Matthias Tan,

Assistant Professor, Department of Systems Engineering and Engineering Management, City University of HK, matthtan@cityu.edu.hk

Title :Monotonic Metamodels for Deterministic Computer Experiments Abstract: In deterministic computer experiments, the output is often a monotonic function of some of the inputs. In these cases, a monotonic metamodel will tend to give more accurate and interpretable predictions with less prediction uncertainty than a nonmonotonic metamodel. The widely used Gaussian process models are not monotonic. This paper proposes a weighted projection approach to monotonize the GP model together with two computational implementations. The first is isotonic regression on a grid while the second is projection onto a cone of monotone splines. Simulations show that the monotone B-spline metamodel gives particularly good results.

Invited Session 3: Systems Reliability and Prognostics

[1]. David Coit,

Professor, Department of Industrial & Systems Engineering, Rutgers University, Piscataway, NJ, USA, coit@rci.rutgers.edu Title: Advanced Reliability Modeling and Optimization for Systems of Degrading Components Abstract: Advanced Reliability Modeling and Optimization for Systems of Degrading Components New reliability models have been developed for systems subject to competing hard and soft failure processes with shocks that have dependent shock effects or dependent degradation paths. In the new model, hard failure occurs when transmitted system shocks are large enough to cause any component in a series system to fail immediately, and soft failure occurs when any component deteriorates to a certain failure threshold, and system shocks affects both failure processes for all components. Our new research extends previous reliability models that had dependent failure processes, where the dependency was only because of the shared number of shock exposures, and not the shock effects associated with individual system shocks or the degradation path. In practice, the effects of shock damages to the multiple failure processes among components are often dependent, and similarly, degradation paths for components within the same system, can behave similarly even if not functionally dependent. In this paper, we combine both probabilistic and physical degradation modeling concepts to develop the new system reliability model. Four different dependent patterns/scenarios of shock effects on multiple failure processes for all components are considered for series systems. This represents a significant extension from previous research because it is more realistic, yet also more difficult for reliability modeling. This is a joint work with Sanling Song.

[2]. Shubin Si,

Professor, The Department of Industrial Engineering, Northwestern Polytechnical University, Xi’an, China, sisb@nwpu.edu.cn

Title: Integrated Importance Measures: Theories and Applications

Abstract: Importance measures, are useful tools to identify the weakness of systems, are used in various fields to evaluate the relative importance of various objects. The development of complex engineering systems bring great challenges to the reliability optimization. To deal with this kind of problem, we proposed the definition and physical meaning of Integrated Importance Measure (IIM) by introducing sate transition process. Then the IIM is extended to the continuous sate systems and the multi-state systems to identify the key component during whole life cycle of systems. Finally, the corresponding applications of IIM are implemented to estimate the weakness and improve the performance of objective system with limited resources.

[3]. Yimeng Xie,

Statistician, AstraZeneca GMD China. Email: xym@vt.edu

Title: Semi-parametric Models for Accelerated Destructive Degradation Test Data Analysis Abstract: Accelerated destructive degradation tests (ADDT) are widely used in industry to evaluate materials’ long term properties. Even though there has been tremendous statistical research in nonparametric methods, the current industrial practice is still to use application-specific parametric models to describe ADDT data. The challenge of using a nonparametric approach comes from the need to retain the physical meaning of degradation mechanisms and also perform extrapolation for predictions at the use condition. Motivated by this challenge, we propose a semi-parametric model to describe ADDT data. We use monotonic B-splines to model the degradation path, which not only provides flexible models with few assumptions, but also retains the physical meaning of degradation mechanisms (e.g., the degradation path is monotonically decreasing). Parametric models, such as the Arrhenius model, are used for modeling the relationship between the degradation and accelerating variable, allowing for extrapolation to the use conditions. We develop an efficient procedure to estimate model parameters. We also use simulation to validate the developed procedures and demonstrate the robustness of the semi-parametric model under model misspecification. Finally, the proposed method is illustrated by multiple industrial applications.

[4]. Lirong Cui,

Professor, Department of Management Science and Engineering, School of Management & Economics, Beijing Institute of Technology, Beijing, 100081, Lirongcui@bit.edu.cn Title: Extended Hawkes Processes and Their Applications in Reliability Abstract: The self-exciting processes named as Hawkes processes, which constitute a particular class of multivariate point processes, have been widely used in financial area special in empirical high frequency finance. Hawkes processes can easily account for the interaction of various types of events, for the influence of some intensive factors (through marks) or for the existence of non-stationarities. Because of their great simplicity and flexibility, Hawkes processes have also been used in a vast array of domains such as seismology, criminology, neuroscience, psychology, sociology, genome analysis, credit card delinquency, video viewing behavior, reliability and many other fields. In this talk, some extended new Hawkes processes are introduced after reviewing theory of Hawkes processes. Then the properties of these extended new Hawkes processes are discussed, and some application examples of extended new Hawkes processes in reliability field are presented, finally the future researches in this direction are discussed briefly.

Invited Session 4: Big Data Analytics and Simulation Analytics

[1]. Zhiqiang Zheng,

Professor, Department of Information Systems and Operations Management, Jindal School of Management, University of Texas, Dallas, ericz@utdallas.edu

Title: Competitive Analytics in Multi-channel Cross-Competitor Advertising Attribution Abstract: Digital innovations in marketing over the last decade have enabled customers to reach sellers

through various digital channels. How to attribute customer purchases across multi-channel advertising become an immediate challenge. Existing attribution models predominantly focus on analyzing customers' converting path with respect to one focal firm while largely overlooking the impact of competitors. We address this problem by developing a competitive analytics model that considers three distinctive stages of a consumer’s purchase funnel – awareness, alternative evaluation and purchase - across all competitors within one industry.

[2]. Chen Nan, Assistant Professor, Department of Industrial & Systems Engineering, National University of Singapore, isecn@nus.edu.sg Title: Statistical Analysis of Simulation Output from Parallel Computing Abstract: This talk studies the statistical output analysis of parallel transient simulations. It focuses on a common type of simulations, where the simulation time correlates with the output value. In this case, when the transient simulations are deployed to parallel computing architecture to increase the simulation replications in a short time, the output sequence from the parallelized computation units no longer comprise i.i.d observations. Neglecting this fact can lead to significant bias when analyzing the simulation output. In this talk, we analyze and discuss the fundamental reasons behind this phenomenon, and propose a statistical approach to provide more accurate estimations. The proposed algorithm is applicable to general terminating simulations, where the terminating event is defined by the hitting of a set of states over the course of state transitions. We have conducted thorough numerical studies to illustrate the performance of the algorithm, and applied it to analyzing the output sequence from queueing system simulation and control chart evaluation.

[3]. Zhaolin Hu,

Associate Professor, Department of Management Science and Engineering, Tongji University, russell@tongji.edu.cn

Title: Convex Risk Measures: Efficient Computations via Monte Carlo Abstract: With the development of financial risk management, the notion of convex risk measures has been proposed and has gained more and more attentions. Utility-based shortfall risk (SR), as a specific and important class of convex risk measures, has become popular in recent years. In this talk we focus on computational aspects for SR, which are significantly understudied but fundamental for risk assessment and management. We discuss efficient estimation of SR, sensitivity analysis for SR, as well as optimization of SR, based on Monte Carlo techniques and stochastic optimization methods. We also conduct extensive numerical study on the proposed approaches. The numerical results further demonstrate the effectiveness of these approaches. This is a joint work with Dali Zhang of Shanghai Jiao Tong University.

[4]. Jie Song,

Assistant Professor, Department of Industrial & Management Engineering, Peking University, Beijing 100871, China, songjie@coe.pku.edu.cn Title: A Simulation Optimization on the Hierarchical Healthcare Delivery System Patient Flow Based on Multi-fidelity Models Abstract: The mismatching patient flow distribution in the healthcare system in urban China is a great social issue that attracts lots of public attention. In this research, we propose a simulation-based optimization method using the Multi-fidelity Optimization with Ordinal Transformation and Optimal Sampling (MO2TOS) algorithm to evaluate the patient flow distribution, so as to continuously improve the hierarchical healthcare service system. The low-fidelity model applying the queueing network theory is constructed for the Ordinal Transformation part of the MO2TOS, followed by a high-fidelity but

time-consuming Discrete Event Simulation model for the Optimal Sampling part. An empirical study on the background of the hierarchical healthcare delivery system in China is presented, where the proposed MO2TOS method is implemented to optimize the system profit by guiding the patient flow distribution. A comparison with other widely used simulation optimization methods sustains the efficacy of the MO2TOS with the evidence that acquiring effective information from the low-fidelity model indeed retrenches the computing budget used to explore the feasible domain.

Parallel Sessions

Parallel Session 5-A: Reliability

 

[1]. Xuejun Liu,

School of Reliability and Systems Engineering, Beihang university, liuxuejun1007@163.com

Title: Reliability-Based Design Optimization for a Centrifugal Compressor

Abstract: A multi-objective reliability-based design optimization of a centrifugal compressor is performed considering manufacturing uncertainty. The purpose of our study is to acquire a high-performance and reliable centrifugal compressor and reduce the sensitivity of the structural parameters to the final performance. In other words, the centrifugal compressors produced have a very large probability to satisfy the performance requirements. The NUMECA® software is utilized to carry out the numerical simulation experiments. To reduce the computational cost, three approximation models between the key design variables and response are constructed by response surface model. Monte Carlo simulation is adopted to calculate the failure probability of atomizing angle and perform reliability analysis. Genetic algorithm is selected to accomplish the optimization of atomizing angle and the determination of manufacturing tolerances. Eventually, the results show that a reliable and robust design which contains the median and tolerance of design variables is obtained.

[2]. Xue Jing,

School of Reliability and Systems Engineering, Beihang University, xjing@buaa.edu.cn

Title: Optimization of Plasma Spraying Process Based on Reliability

Abstract: Plasma spraying is a kind of thermal spraying technology widely used in aero-engine parts manufacturing. Despite it has achieved considerable development over the past several decades, how to control various defects in the plasma spraying coating is still the main problem in the current technology. In the paper, the surface spraying in the aero-engine compressor casing inner ring was selected as the research target. This research focuses on optimizing plasma spraying process from the perspective of process reliability, which performing a closed loop, namely definition, measurement, analysis, optimization, and verify, to plasma spraying process in order to improve the process reliability, in turn elevate the quality level of coating and the performance of aero-engine.

Firstly, the qualification rate of coating is defined as the evaluation index of plasma spraying process reliability by identifying the important requirement of coating according to the application. Secondly, the process failure modes and effects analysis (FMEA) is made for the plasma spray process. Thirdly, the process parameter settings of plasma spray have been optimized by means of integrating the Orthogonal test design method, Back-Propagation Neural Networks (BPNN), and Particle Swarm Optimization (PSO) algorithm.

It has studied one indicator of coating quality (micro-hardness) together with nine process parameters (argon pressure, hydrogen pressure, argon gas flow, send powder gas pressure, send powder gas flow

and amount of powder, stand-off distance, electric current and voltage). The Orthogonal test design was used to arrange the typical experiments in order to reduce the number of experiments and obtain the significant impact factors along with a group of preliminary parameters level combination. Data from Orthogonal test was utilized for training and verifying the BPNN whose structure had been optimized by particle swarm algorithm. Then, PSO method was employed again to optimize process parameters combined with the relationship model between the process parameters and indicators established by the optimized BPNN. Finally, a confirmation experiment has been performed to illustrate the effectiveness of the optimal parameter settings.

[3]. Juan Wang,

School of Economics and Management, Nanjing Polytech University, 961055477@qq.com

Title: Reliability Sensitivity Analysis based on Kriging-model and Importance Sampling

Abstract: Due to the lack of the gradient information, the reliability sensitivity analysis can’t be implemented in an analytical manner. Therefore the paper proposed a simulation-based sensitivity method. It is suggested to compute the failure probability by the combination of Kriging model and importance sampling at first, and then the estimator of failure probability is differentiated through the score function approach. The combination of Kriging and importance sampling resort to the construction of an accurate Kriging surrogate of the limit state function through stepwise uncertainty reduction(SUR) criterion which is commonly used in Inversion problems, and score function approach enables the estimation of the gradient of the failure probability without any additional evaluation to the limit state function. The results illustrate that the proposed method is efficient and precise, especially when the performance function involves the output of an expensive-to-evaluate computational model or the reliability sensitivity of a system is considered.

[4]. Yanjing Zhang,

Department of management science, Nanjing University of Science and Technology, zhangyanjing_1211@163.com

Title: Series System Reliability Modeling based on Competing Failure Processes and Maintenance Strategy

Abstract: Aiming to resolve the problem of reliability assessment of a series system which is affected by natural degradation and random shocks, a reliability model based on two competing failure processes is developed. The model takes into account the impact of shocks to degradation, which is the increase of degradation resulted from random shocks. Then a condition-based maintenance strategy is proposed and a cost rate function is constructed. Finally a series system containing four components is applied to estimate the reliability and the condition-based maintenance strategy is used to analyze it. The system’s reliability function and optimal prevention threshold are determined. The results not only validate the rationality and effectiveness of the model, but also reflect the feasibility and economy of the condition-based maintenance strategy.

[5]. Yiwen Zhang,

School of Economics and Management, Tianjin University, 857881577@qq.com

Title: To Be Updated.

Abstract: After sales service means offering maintaining, repair or replace service ensures customers interest and avoids customers right from being hurt. We research on the supply chain with one

manufacturer and one retailer in a single period. We study the supply chain coordination mechanism and warranty period length setting strategy considering the influence on product demand, mainly for the influence of service level and warranty period length. Nowadays the after-sales market is a hot research and economic area, and what we concentrate on in this paper will also attract a lot of attention.\

 

Parallel Session 5-B: Modeling  

[1]. Di Wang,

Department of Industrial Engineering and Management, Peking University, di.wang@pku.edu.cn

Title: An Approximation Method for Grain Temperature Field via Spatiotemporal Dynamic Model

Abstract: Wireless sensor network has been widely used to monitor the grain storage condition, and the temperature profile monitoring in the grain warehouse is one of the critical issues for the quality assurance of the grains in storage depot. However, due to the limited resources for sensor deployment in a single warehouse, only sparse sensors are engaged to collect data, which could not sufficiently characterize the temperature dynamics in the grainhouse, lacking an appropriate decision for grain maintenance. This article presents a full-scale approximation approach to model spatio-temporal dynamics of the interior temperature of warehouse through integrating thermodynamics model and spatiotemporal stochastic processes. Specifically, we integrate a 3-D unsteady heat transfer model into a Gaussian Markov random field to achieve a parsimonious representation of spatial patterns. To characterize spatiotemporal dependence structures, we consider the temporal and spatial data simultaneously by using a particle filter to sequentially update the parameters of spatial model with the recursive Bayesian estimation, in which spatiotemporal interaction is taken into account to model the physical process adequately. To validate our approach, both of the simulation and real case are conducted to demonstrate the effectiveness of the developed method. The results of the simulation and real case study bring a deep insight of spatial-temporal dynamics of the interior temperature, and meanwhile provide a guidance for future sensor placement design and parameter calibration

[2]. Lisha Yu, Department of System Engineering and Engineering Management City University of Hong Kong, yls33@hotmail.com

Title: Joint modeling of user's activity and interest for travel prediction and analysis

Abstract: The ability to create geotagged photos enables users to share their personal experiences during travel at specific locations and times. Using location history data, we aim to understand what basic laws govern users’ travel and dynamics. The previous work considers only latent geographical factors in a multiplicative manner by assuming users are clustered into distinct styles of travel. Here we propose a probabilistic behavior model that jointly incorporate both user interest and spatial features of locations by combing Markov topic model and latent factor model. By taking into account a user’s travel history, the latent personal interest and propensity on location can be inferred effectively. It better explains the user behavior from interpretable properties: spatial influence and personal influence. The method is demonstrated on two sources of travel data: geotags from Japan travel group and Flickr images.

[3]. Jing Li, State Key Laboratory of Fluid Power and Mechatronic Systems, Zhejiang University.

lijing33@zju.edu.cn

Title: Bus idle time distribution analysis for controller area network containing errors

Abstract: Controller Area Network (CAN) is a widely used Field Bus protocol in various industrial applications. In order to understand the network behaviors under errors for optimal design of the networked control systems, the bus idle time of the CAN nodes need to be analyzed. This paper presents a novel analysis method to estimate the bus idle time distribution for CAN networks operating on polling communication mode. Firstly, the bus idle time distribution of single slave node is analyzed based on given error distribution, then the preliminary analysis result on the multiple nodes scenario is introduced by merging multiple nodes. Experiments are conducted to illustrate the proposed the method, and the results show that the bus idle time distributions obtained by proposed method agree well with the actual observations.

[4]. Hongyan Wang, Department of Mathematics of Tufts University, hongyan0627@gmail.com

Title: To Be Updated.

Abstract: The Extended Yard-Sale model of asset exchange is an agent-based economic model with binary transactions, and simple models of redistribution and Wealth-Attained Advantage. The evolution of its wealth distribution is described by a nonlinear, nonlocal Fokker-Plank equation. The solutions to this equation fit remarkably well to the actual wealth distributions of the U.S. in 2013. We can get both steady state solutions and time-dependent solutions of this equation by Monte Carlo method, Shooting method and Finite-Element method.

[5]. Qiang Zhou, Assistant Professor, Department of SEEM, City University of Hong Kong, q.zhou@cityu.edu.hk

Title: Multivariate Profile Modeling Using Multivariate Gaussian Processes

Abstract: Although existing profile monitoring methods almost exclusively deal with univariate profiles, observations from a set of correlated profiles (multivariate profile data) are not uncommon. These data are seldom analyzed due to lack of effective modeling tools. Here we propose to analyze them using the multivariate Gaussian process model. To mitigate the prohibitively high computation in building such models, a pairwise estimation strategy is adopted. Asymptotic normality of the parameter estimates from this approach has been established. Simulation and real data studies are conducted.

 

Parallel Session 5-C: SPC  

[1]. Yanting Li,

Department of Industrial Engineering and Management, Shanghai Jiao Tong University, ytli@sjtu.edu.cn

Title:  A False Discovery Approach for Scanning Spatial Disease Clusters with Arbitrary Shapes

Abstract: The spatial scan statistic is one of the main tools for testing the presence of clusters in a geographical region. The fast subset scan (FSS) method recently proposed by Neill (2012) represents an important extension as it is computationally efficient and enables detection of clusters with arbitrary shapes. Aimed at automatically and simultaneously detecting multiple clusters of any shapes, this paper explores the false discovery (FD) approach originated from multiple hypothesis testing. We show that the FD approach can provide higher detection power and better identification capability than the

standard scan and FSS methods on average.

[2]. Jian Li,

School of Management of Xi’an Jiao Tong University, jianli@mail.xjtu.edu.cn

Title: Multivariate Ordinal Categorical Process Control based on Log-linear Modeling

Abstract: In many applications, the quality of products or services tends to be measured by multiple categorical characteristics, each of which is classified into attribute levels such as good, marginal, and bad. Here there is usually natural order among these attribute levels. However, traditional monitoring techniques ignore such order among them. By assuming that each ordinal categorical quality characteristic is determined by a latent continuous variable, this paper incorporates the ordinal information into an extended log-linear model and proposes a multivariate ordinal categorical control chart based on generalized likelihood ratio test. The proposed chart is efficient in detecting location shifts and dependence shifts in the corresponding latent continuous variables of ordinal categorical characteristics based on merely the attribute level counts of the ordinal characteristics.

[3]. Dong Ding,

School of Management of Xi'an Polytechnic University, laceyding@foxmail.com

Title: Rank-based Process Control for Mixed-type Data

Abstract: Conventional statistical process control tools target either continuous or categorical data but seldom both at the same time. However, mixed-type data consisting of both continuous and categorical observations are becoming more common in modern manufacturing processes and service management, which cannot be tackled by traditional methods. By assuming that there is a latent continuous variable that determines the attribute levels of a categorical variable, the ordinal information among the attribute levels can be exploited. This enables us to describe and monitor continuous and categorical data simultaneously in a unified framework of standardized ranks, based on which a multivariate exponentially weighted moving average control chart is proposed. This control chart specializes in detecting location shifts in continuous data and in latent continuous distributions of categorical data. Numerical simulations show that our proposed chart can efficiently detect location shifts and is robust to various distributions.

[4]. Caiwen Zhang,

Business School, Sun Yat-sen University, zhangcw@mail.sysu.edu.cn

Title: Monitoring the Shape Parameter of the Weibull Renewal Process

Abstract: This research has arisen from a challenge faced in real practice—monitoring changes to the Weibull shape parameter. From first-hand experience we understand that a mechanism for such a purpose is very useful. This study is primarily focused on monitoring the shape parameter of a Weibull renewal process. We derive a novel statistic on the Weibull shape parameter making use of the maximum likelihood theory, which is demonstrated to follow an approximately normal distribution. This desirable normality property makes the statistic well suited for monitoring the Weibull shape parameter. It also allows for a simple approach to constructing a Shewhart-type control chart, named Beta chart. The parameter values necessary for designing a Beta chart are provided. A procedure based on sequential sampling is also proposed for establishing a Phase I Beta chart. The average run length (ARL) performance of the Beta chart is evaluated through Monte Carlo simulation. A comparison with the moving range EWMA chart proposed in Akhundjanov and Pascual (2015) shows that the Beta chart has much better ARL performance when properly designed. Application examples using simulated and real data demonstrate that the Beta chart is effective and makes good sense in real practice.

[5]. Tingting Huang ,

Assistant Professor, The Department of Systems Security and Reliability Engineering, Beihang University, Beijing, China, htt@buaa.edu.cn

Title: Quality Control for Fused Deposition Modeling Based Additive Manufacturing: Current Research and Future Trends Abstract: Additive manufacturing (AM) is a novel way of manufacturing in the recent decade, and it has complexity free and customization advantages over subtractive manufacturing. Fused deposition modeling (FDM) is one way of AM, and it fabricates products using liquefied thermoplastic material. Products are manufactured layer-by-layer based on 3D computer aided designs (CAD). Research on quality control of FDM is urgently needed to help improving the manufacturing process, and guarantee the quality of outgoing products. This talk discusses the current research in quality control in FDM, including process and quality parameters, measuring instruments, EPC and SPC methods. Some tentative topics of future research in this area are proposed finally: (1) EPC of multi-parameter based on fused information; (2) SPC based on profile data of machine and operation specific parameters; (3) SPC based on regular image data of quality characteristics; (4) SPC based on infrared image data of operation specific parameters; (5) Run-to-run process control.

 

Parallel Session 6-A: Reliability

[1]. Hong-gen Chen,

School of Management Engineering, Zhengzhou University of Aeronautics, microwr@163.com

Title: Extended Hawkes Processes and Their Applications in Reliability

Abstract: As two key tools for process management, Statistical Process Control and Maintenance Management can create remarkable economic benefits, particularly when they are coordinated. Distinguish from the integrated models that mostly built based on traditional static control charts, the objectives of this research are to develop an integrated model between adaptive control charts and Planned Maintenance. To do this, a joint economic design model of variable sample size control chart (VSS control chart) and Planned Maintenance is presented. In addition, the model is extended from four policies used in the consideration (Zhou & Zhu, 2008) to five policies as the result that a dormant fault will deteriorate into functional failure with Delay-time system running if the dormant fault is not identified and corrected. A mathematical model is given to found the optimal values of sample size n1, sample size n2, sampling interval h, and control chart limit w, and Planned Maintenance period T that minimize the expected cost per unit time. Finally, the sensitivity analyses are employed to demonstrate the effect of model parameters by using multiple regressions.

[2]. Guodong Wang,

Department of Management Engineering, Zhengzhou University of Aeronautics, gdwang@tju.edu.cn

Title: Improving Reliability Using Experimental Design with a Non-constant Shape Parameter

Abstract: Reliability has always been considered to be one of the most important characteristics of a product. There is a strong need for design of experiments (DOE) that allow practitioners to identify the factors which affect lifetime significantly. In this article, we consider how to improve reliability through the use of designed experiments based on Weibull distribution with a non-constant shape parameter. We

illustrate our proposed method through one real experiment contained censoring. Analysis results show that the misspecification of a constant shape parameter can lead to misidentify spurious effects.

[3]. Xinghong Qin,

Department of management and economy, Tongji university, qinxinghong4515@sina.com

Title: Extended Warranty Strategies for Competing Online Shopping Supply Chain with Different Component Reliabilities

Abstract: This article presents the issue of extended warranty and management strategies in a three-echelon competing online shopping supply chain (OSSC) comprised of two competing component suppliers, a manufacturer and an online store, with price- and base warranty period-dependent demand. We employ the game theory to develop decision models to explore interactions between the component suppliers and manufacturer as well as competition between the two suppliers. For this purpose, two scenarios are considered: either the manufacturer offers a paid extended warranty to customers or doses not. In each scenario, base warranties are assumed to be bundled with products. Our results show that when manufacturer’s repair costs change in a proper range, providing extended warranty can benefit both the manufacturer and online store; otherwise, the manufacturer has no incentive to offer the extended warranty. Reducing repair costs or improving reliability or shortening the base warranty encourages the manufacturer to achieve significantly better value of the extended warranty, vice versa. High reliability of components benefits both the manufacturer and the online store; however, higher reliability of components is more conducive to the manufacturer. Extending the length of the base warranty adversely affects profit of the manufacturer and the value of the extended warranty.

[4]. Jiaheng Shen,

School of Reliability and System Engineering, Beihang University, shenjh0606@163.com

Title: To Be Updated.

Abstract: Swirler is an important component of combustor in aero-engine and the performance of swirler would have a direct effect on the operation stability of the whole combustor and then affect the performance of aero-engine. Thus, a swirler with high performance is desirable. Based on the current engineering experience, five potential structural parameters of the swirler are identified as performance sensitive parameters. The value ranges of the five parameters are also determined simultaneously. Then, two sets of experiments including 17 tests are designed utilizing the Sequential Latin Hypercube design. The FLUENT software is used to carry out the numerical simulation experiments. Eventually a set of optimized parameters approaching theoretical optimal parameter combination is found.

Parallel Session 6-B: DOE  

[1]. Jai-Hyun Byun,

Department of Industrial and Systems Engineering of Gyeongsang National University, jbyun@gnu.ac.kr

Title: A Screening Design Approach for Multiple Responses with a Case Study

Abstract: For product or process design and development, it is common to optimize multiple responses (characteristics) based on experimental data. To determine optimal factor conditions, we need to design the experiment, obtain proper model for each response, and optimize the multiple responses simultaneously. There are several techniques and many research papers on optimizing multiple

responses simultaneously, when the experimental data are available. However, the experimental design issue for optimizing multiple responses is not fully discussed yet. This paper proposes some idea on how to plan screening design when we consider multiple responses to be optimized. A case study is also presented to demonstrate the validity of the approach

[2]. Hao Zhaojun,

School of Reliability and Systems Engineering Beihang University, gouliu0324@buaa.edu.cn

Title: The Applications of DOEs and Computational Emulation Technology to Process Optimizations of the Turbine Blades of Aircraft Engines

Abstract: Aero engine turbine blade is the key part of the aero-engine, whose ability to withstand temperature is the critical factor to evaluate engine performance and determine the life of the aero & engine. Therefore, manufacturing quality and accuracy of the blade also have a direct impact on overall performance and service life of aero engine.

It is revestment precision casting technology that is widely used in the manufacturing process of turbine blade. With the development of high thrust, high efficiency, long service life of turbine engine, the engine needs to be continuously improved on the inlet temperature of gas turbine so that turbine blade design structure become more and more complex, which brings greater difficulties in the process, and ultimately directly affects the production quality of the blade. In fact, most of the product defect is the loose mainly located in the blade basin.

Experimental design is a technique used for arranging the tests economically and scientifically, so as to improve the output of the product and reduce the fluctuation of the quality. In this paper, the method of experiment design has been applied to analyze several process parameters (mold temperature, filling temperature, and holding time, etc.) related to the leaf quality, and provide a specific experiment scheme; subsequently, the ProCAST software was used to carry out machine process simulation tests according to the experiment scheme; finally, specific defect location and size were analyzed based on the simulation experiment results. In addition, both the optimal process parameters combination and operation scheme have acquired together. Actually, the final results of the process parameters were put into practical application, not only the pass rate has been greatly increased, but also the engine life and performance have been correspondingly improved.

[3]. Yanrong Li,

College of Management and Economics, Tianjin University, rongrong_bonita@163.com

Title: To Be Updated.

Abstract: With the rapid development of technology, many products improve their features with a frequent upgrading process, especially for electronic products. When the produce process is finished and sales volume comes down in the end of selling process, for old generation products, contiguous generations have an obvious substitution effect, we can provide warranty by new generation and abandon the infrastructure for old generation repairmen to avoid a huge fixed cost, so we consider an optimal moment for manufacturers to transform their after-sale warranty service from repair and spares part change to replace with new generation products taking inventory management and impact of learning efficiency on alternative cost in to account. Previous research focuses on inventory decision, reduction and extend production as warranty policy in the end of life for products, however we in allusion to the concrete replace process for new generation and verify cost for warranty supplier. Furthermore, numerical analysis and sensitive analysis can explain how different parameters influence the total warranty cost for old generation products and search an equilibrium point in time for strategy alternative.

[4]. Guilin Li,

Department of Industrial & Systems Engineering, National University of Singapore, guilin_li@u.nus.edu

Title: Bayesian Experimental Design for the Estimation of Optimum Point in Generalized Linear Models

Abstract: As Integrated Circuits are rapidly migrated to smaller feature size, metallic shorts become a severe reliability issue. A preliminary experiment has been conducted to explore the functional dependency of the shorts count on the controlled variables such as metal width and metal length. A generalized linear regression model is adopted to model the preliminary experiment result. An important quantity of interest to the industry is the optimum metal width and length setting that would give the minimum number of shorts. To better estimate this optimum setting, an additional follow up batch experiments can be conducted. In this work, a novel framework is developed to design the desired experiment. This framework uses a Bayesian treatment of parameter uncertainty, employs a Shannon information utility measure, and accommodates high order polynomial structure in the covariates. Efficient approximation strategies are explored to calculate the expected utility function, followed by a heuristic optimization method to find the optimal experiment design.

Parallel Session 6-C: SPC  

[1]. Wenjuan Liang,

Dr., School of Statistic, East China Normal University，wjliang0627@163.com

Title：Monitoring Large Sparse Contingency Table in Multivariate Categorical Process

Abstract: In modern statistical process control (SPC), more and more real applications involve multiple categorical quality characteristics, whose distribution can be displayed by contingency table. Traditional methods to monitor them are usually developed for the “small cell number and large sample size”. When the number of categorical variables increases, the number of cells in contingency table grows extremely fast, so that most of the cell entries are very small or zeros counts, this is so-called sparse contingency table. In this situation, the traditional methodologies are inadequate to use because of the non-existence of the maximum likelihood estimate (MLE) of the expected cell counts in contingency table. In this paper, we are devoted to developing a monitoring method for sparse contingency table. Firstly, a two-stage group lasso method is developed to perform models selection and parameter estimation in high-dimension log-linear models. Secondly, based on the modified Pearson χ2 statistic, a new EWMA control scheme for monitoring the probability distribution of sparse contingency table is proposed. Compared with the control charts based on power-divergence test family, such as Pearson χ2 test statistic, likelihood ratio test statistic and Freeman-Tukey test statistic, our proposed control chart has a better overall performance. Finally, a real data example is used to demonstrate the effectiveness of the proposed control chart.

[2]. Chunjie Wu，

Associate Professor, School of Statistics and Management ， Shanghai University of Economics and

Finance，wumaths@mail.shufe.edu.cn

Title: A Robust Latent CUSUM Chart for Monitoring Customer Attrition

Abstract: The customer attrition rate is one of the key analytic metrics for customer-based businesses. Shifts occurring in the customer attrition rate of a business are usually small and persistent, and so CUSUM charting methodology would seem ideal for their detection. However, customer summaries are available only on an uneven time scale, and this prevents the use of traditional CUSUM methods. This article develops a latent CUSUM chart which is based on an oversimplified exponential model for

customer departures which are then summarized as monthly departures. Estimation of parameters via both maximum likelihood and least squares is considered. While both estimation methods perform well, rapidly detecting both small and large shifts, least squares methods are advantageous when attempting to detect a very small shift. The robustness of the chart to departures from the (too simple) latent model is studied, and the chart is found to perform well. The charts are applied to customer departure data from a large e-commerce company.

[3]. Xue Li, Prof., School of management engineering , Zhengzhou university of aeronautics, xueli_fys@sina.com Title: To Be Updated. Abstract: Statistical process control (SPC) is usually used to promote a product quality in the production processes. Control charts, which control the production process condition and guarantee the process quality, are the important statistics process control tools. Their design methods directly affect the efficiency and cost of quality control in manufacturing process. The exponential weighted moving average (EWMA) control charts which accumulate the small shift of the manufacturing process by exponential form, have the better monitoring efficiency than traditional shewhart control chart. Saccucci etal first proposed variable sampling interval (VSI) EWMA control charts, their results showed that the monitoring efficiency of VSI EWMA control charts excel to the fixed sampling intervals (FSI) EWMA control chart. Economic designs of VSI EWMA control charts have been widely investigated and insure that the economic design of control chart actually has a lower cost. A preventive maintenance can reduce the failure rate to an out of control state by an amount proportional to the preventive maintenance level. This paper presents an integrated model for combining the preventive maintenance and the economic design of VSI EWMA control charts using the Taguchi loss function. The maintenance activities are coordinated with the statistical characteristics of the sampling results. Finally, a numerical experiment is conducted to investigate the model’s working underlying the effect of preventive maintenance on the quality control costs.

[4]. Qijun Zhong, College of Management and Economics, Tianjin University, 277270555@qq.com Title: To Be Updated.

Abstract: Lean Six Sigma (LSS) has been widely used and proven to be an effective framework for performance improvement in a wide variety of industries for 30 years. Though many companies have generated significant business results through deploying LSS, quite a few quit the journey of LSS after several years implementation because they lost their ways and did not know how to consistently push Six Sigma forward. To enhance the future deployment of LSS, It is quite important for a company to locate where it stands and clearly understand the strengths and opportunities for improvement. Unfortunately in academia research on how to measure the maturity of LSS is quite limited. We developed a maturity assessment model based on theoretical development and implementation practice of LSS in China. The model includes leadership, LSS infrastructure, LSS linking to strategy, customer satisfaction, project management, motivation and incentives, and business results. The model has been used as a measurement tool for LSS excellence businesses evaluation since 2007. Through this model and data from more than 50 companies evaluated with it, some common problems of LSS implementation in China are found. This model can also be adopted to other countries in terms of LSS implementation.

Name University EmailXiaoli Bao Shanghai Jiao Tong University baoxiaoli_jane@163.comJai-Hyun Byun Gyeongsang National University jbyun@gnu.ac.kr

Hong-gen Chen Zhengzhou University of Aeronautics microwr@163.com

Nan Chen National University of Singapore isecn@nus.edu.sg

Zhuo CHEN Shanghai Jiao Tong University chenzhuo95@sjtu.edu.cn

David Coit Rutgers University davecoit@gmail.com

Lirong Cui Beijing Institute of Technology Lirongcui@bit.edu.cn

Anshu Dai Tianjin university anshudaitju@163.com

Xinwei Deng Virginia Tech xdeng@vt.edu

Dong Ding Xi'an Polytechnic University laceyding@foxmail.com

Juntao Fang Tianjin University fangjuntao@tju.edu.cn

Weidong Guo Shanghai Jiao Tong University weidongguo@yeah.netMei Han City University of Hong Kong

Zhaojun Hao Beihang University gouliu0324@buaa.edu.cn

Zhen He Tianjin University zhhe@tju.edu.cn

Yili Hong Virginia Tech yilihong@vt.edu

Zhaolin Hu Tongji University russell@tongji.edu.cn

Min Huang Beihang University buaahuang@163.com

Tingting Huang Beihang University htt@buaa.edu.cn

Wenpo Huang Shanghai Jiao Tong University bobhuang09@gmail.com

Wei Jiang Shanghai Jiao Tong University jiangwei@sjtu.edu.cn

Jay Lee University of Cincinnati Jay.lee@uc.edu

Guilin Li National University of Singapore guilin_li@u.nus.edu

Jian Li Xi'an Jiaotong University jianli@mail.xjtu.edu.cn

Jing Li Zhejiang University lijing33@zju.edu.cn

William Li University of Minnesota wli@umn.edu

Yanting Li Shanghai Jiao Tong University ytli@sjtu.edu.cn

Yanrong Li Tianjin University rongrong_bonita@163.com

Yuan LI Shanghai Jiao Tong University llliyuan@sjtu.edu.cn

Xue Li Zhengzhou university of aeronauticss xueli_fys@sina.com

Shujun LIU Shanghai Jiao Tong University liushujun_uestc@163.com

Xiao Liang Tsinghua University liangq12@mails.tsinghua.edu.cn

Wenjuan Liang East China Normal University wjliang0627@163.com

Xuejun Liu Beihang university liuxuejun1007@163.com

Zuoyi Liu National Science Foundation of China

Yaqi Lou Tianjin University louyaqi@tju.edu.cn

Jun Luo Shanghai Jiao Tong University jluo_ms@sjtu.edu.cn

Xinggang Luo Northeastern University

Yizhong Ma Nanjing University of Science and Technology

Yi Man Shanghai Jiao Tong University alexmanyi@foxmail.com

Sen Niu Shanghai Jiao Tong University nionsion@163.comShizhe Peng Shanghai Jiao Tong University pengshiz@126.com

Jiaheng Shen Beihang University shenjh0606@163.com

Shubin Si Northwestern Polytechnical University sisb@nwpu.edu.cn

Jie Song Peking University songjie@coe.pku.edu.cn

Qiang Su Tongji University suq@tongji.edu.cn

Matthias Tan City University of Hong Kong matthtan@cityu.edu.hk

Pariticipants List

Jiafu Tang Northeastern University jftang@mail.neu.edu.cn

Kwok Leung Tsui City University of Hong Kong kltsui@cityu.edu.hk

Fugee Tsung Hong Kong University of Science and Technology season@ust.hk

Rui Tuo Chinese Academy of Sciences tuorui@amss.ac.cn

Azmat Ullah Shanghai Jiao Tong University azmat_aries897@yahoo.comGuohuan WAN Shanghai Jiao Tong University ghwan@sjtu.edu.cn

Di Wang Peking University di.wang@pku.edu.cn

Dongfan Wang Tianjin university wangdongfan123@yeah.net

Guodong Wang Zhengzhou University of Aeronautics gdwang@tju.edu.cn

Hao Wang Tsinghua University w-h14@mails.tsinghua.edu.cn

Hongyan Wang Tufts University hongyan0627@gmail.com

Kaibo Wang Tsinghua University kbwang@tsinghua.edu.cn

Juan Wang Nanjing University of Science and Technology 961055477@qq.com

Lian WANG Shanghai Jiao Tong University wang_lian@sjtu.edu.cn

Zhiyuan WANG Shanghai Jiao Tong University bigbigwzy@sjtu.edu.cn

Zihui Wang Beihang University wangzihui1994@163.com

Qingming WEI Shanghai Jiao Tong University weiqm1999@163.com

William H. Woodall Virginia Tech bwoodall@vt.edu

Jeff Wu Georgia Institute of Technology jeff.wu@isye.gatech.edu

Yimeng Xie AstraZeneca GMD China xym@vt.edu

Jing Xue Beihang University xjing@buaa.edu.cn

Jinyu Yang Nankai University jyyang@nankai.edu.cn

Shuang YANG Shanghai Jiao Tong University 277531971@qq.com

Kai Yu Shanghai Jiao Tong University yukai2015@163.com

Lisha Yu City University of Hongkong yls33@hotmail.com

Caiwen Zhang Sun Yat-sen University zhangcw@mail.sysu.edu.cn

Chongqi Zhang Guangzhou University cqzhang@gzhu.edu.cn

Yanjing Zhang Nanjing University of Science and Technology zhangyanjing_1211@163.com

Yiding Zhang Shanghai Jiao Tong University zydzyd@sjtu.edu.cn

Yiwen Zhang Tianjin University 857881577@qq.com

Wenhui Zhao Shanghai Jiao Tong University zhaowenhui@sjtu.edu.cn

Yu Zheng Beihang University 296180982@qq.com

Zhiqiang Zheng University of Texas at Dallas ericz@utdallas.edu

Qijun Zhong Tianjin University 277270555@qq.com

Qiang Zhou City University of Hong Kong q.zhou@cityu.edu.hk

Zhizhong Zhou Shanghai Jiao Tong University zhouzhzh@outlook.com

Changliang Zou Nankai University nk.chlzou@gmail.com

Others

Transportation

From Shanghai Pudong International Airport (PVG):

Take metro line 2 and transfer to line 10 at East Nanjing Road, and then get off at Shanghai Jiao Tong

University.

About 1 hour’s taxi ride at the cost of 200-250 RMB.

From Shanghai Hongqiao International Airport (SHA):

Take metro line 10 and get off at Shanghai Jiao Tong University

About 30 minutes’ taxi ride at the cost of 50-60 RMB.

From Shanghai Hongqiao Railway Station:

Take metro line 10 and get off at Shanghai Jiao Tong University

About 30 minutes’ taxi ride at the cost of 50-60 RMB.

Recommended Hotel For the Conference

Please reserve the hotel on your own, the recommended hotel is list below: Hotel

Name

Crowne Plaza Hotel

银星皇冠酒店

The Suites – SHANGHAI Hotel

西华酒店

Hanting Hotel 汉庭酒店

Jinjiang Inn

锦江之星

No. 1 Dormitory Building (EMBA dormitory) 交大研一楼

Address No. 400, Pan Yu Road

番禺路 400 号

No. 1, Huai Hai W. Road

淮海西路 1 号

No. 955, Pan Yu Road

番禺路 955 号

No. 319, Guang Yuan W.Road

广元西路 319 号

Shanghai Jiaotong University (SJTU) No.655 Panyu Rd

交大校内番禺路 655 号

Tel 021-61458888 021-51801133 021-64480808 021-64471000 021-62932690 Website www.ihg.com/cro

wneplaza/hotels/us/en/shanghai/shgch/hoteldetail

http://www.xihuahotels.com/english/index1.html

http://hotels.huazhu.com/hotel/detail?HotelId=2003001

http://www.jinjianginns.com/HotelDetail?hotelId=12078

Please contact Mr. Chaoyang Liu liuchaoyang@sjtu.edu.cn for dorm reservation(About 300 per night)

Trip In Shanghai

Taxi

Shanghai's 45,000 taxis are easy to flag down in leisure hours, but not in rush hours. The taxi drivers are bound to offer you good service. If you do not speak Chinese, it is recommended that you have your destination written down in characters.

The taxi companies listed below are most strongly recommended.

Main Taxi Company For Booking For Complaint Taxi Color

Dazhong (+86 21)96822 (+86 21)6258 0780 Sky blue

Qiangsheng/Bashi (+86 21)62580000 (+86 21)6258 1234 Orange

Haibo (+86 21)96933 (+86 21)6213 0011 Sapphire blue

Jinjiang (+86 21)96961 (+86 21)6416 9292 White

The service time and the corresponding charge standard of the above taxis are as follows.

0~3 KM 3~10 KM Above 10 KM

Daytime: (5:00~23:00) RMB 14 RMB 2.4 / KM RMB 3.6 / KM

Night: (23:00~5:00) RMB 18 RMB 3.1 / KM RMB 4.7 / KM

Notices & Tips:

Please request an invoice when you get to your destination. If you experience anything unreasonable, you may file a complaint by calling

(+86)-21-6323-2150 or the complaint telephone number in the chart. If there is a waiting time, the driver will charge you as one kilometer’s cost for each five

minutes Bridges and tunnels in the downtown area are free of charge. Check the charge standard marked on the windows of a taxi before taking it. Pay taxi fare as

displayed on the price screen. You can refuse to pay if the driver does not start the meter or doesn't give you a valid

receipt.

Train

Bullet Train is most welcomed to travel within Chinese main cities, its highest speed could climb to 300 km per hour, so it will be faster than plane if the destination is less than 1,500 KM away. So we highly suggest you to choose bullet train if your want to visit a city not so far from Shanghai.

Shanghai Railway Reservation Telephone Numbers

Shanghai Railway Administration ticket office 9510 5105

Enquiries to Shanghai Railway Station, Shanghai South Railway Station or Hongqiao Railway Station

(+86)-21-6317-9090 (+86)-21-5436-9511 (+86)-21-5124-5555

Baggage service for Shanghai Railway Station, or Shanghai South Railway Station

(+86)-21-6317-4168

Public Services Contacts

Emergency

Police - 110 Fire - 119 Ambulance - 120

Though Shanghai is a safe city, you may occasionally need help from local police. In case of any burglary or petty theft, dial 110 for police.

Fire is a nightmare in big cities. The situation goes from bad to worse if a skyscraper turns into a fiery building. If you run into such an unfortunate event, do not forget to dial 119.

Roads in Shanghai often get congested and are prone to accidents, so always remember 120. This number may very well determine life or death.

Money Issues

RMB/Yuan, the official currency in China is a decimalized currency. For current conversion rates you can check online at http://www.xe.net/ucc/convert.cgi or

http://www.oanda.com/converter/classic, the estimated exchange rate are: 1 USD: 6.3 RMB, 1 EUR: 8.2 RMB.

One Yuan is worth 100 cents. Coins are issued in denominations of 10 cents, 50 cents and 100 cents. Currency notes are available in 1, 5, 10, 20, 50, 100 denominations. Foreign currencies can be conveniently changed at banks with your passport as ID.

Most internationally recognized currencies (like USD, EUR, GBP, SGD) and tourist cheques can be exchanged at commercial banks, top-end hotels and the international airports in Shanghai. All branches of Bank of China and other large banks offer currency exchange services.

Credit cards (Visa, Mastercard, American Express, etc.) can be used at most hotels, restaurants and major shops. Some credit cards can be used at Bank of China ATM machines.

Other Practicalities

The Standard Time in China is Beijing Time, which is 8 hours ahead of GMT.

Telephone Codes

Calling to:

Int’l Access Code Country Code City Code Phone No.

Shanghai 00 86 21 1234-5678

Beijing 00 86 10 1234-5678

When dialling within China you need to add '0' before the city code.

Climate

Shanghai usually has a relatively short spring and autumn and a long winter and summer. The weather is usually warm and humid. The middle of October in Shanghai is usually cool and dry, with an average temperature of 15-20℃.

Plugs and Electricity

There are two official standards for plugs and sockets in the People's Republic of China. The first is the grounded, three-blade CPCS-CCC which is practically interchangeable with the type of socket found in Australia. The second is Chinese non-grounded two-blade plug, very much like a two-blade North American/Japanese plug. No matter what the plug type is, electrical sockets (outlets) in the People's Republic of China usually supply electricity at between 220 and 240 volts AC. If you're plugging in an appliance built for 220-240 volt electrical input, or an appliance that is compatible with multiple voltages, then an adapter is all you need.

For further information about outlets and power, please consult with hotel receptionist