Abstracts

  • This talk will introduce a new method for constructing supersaturated designs (SSDs). The method leads to a partitioning of the columns of the design so that the columns within a group are correlated to the others within the same group, but are orthogonal to any factor in any other group. These new designs are called group-orthogonal supersaturated designs (GO-SSDs). Using this group structure it is possible to find an unbiased estimate of the error variance and develop an effective, design-based model selection procedure. Simulation results show that the use of these designs, in conjunction with our model selection procedure, enables the identification of larger numbers of active main effects than have previously been reported for supersaturated designs. These designs and their automated analysis are new in JMP 15. This talk will provide an example of this new design approach applied to the time it takes the custom design tool in JMP to create a design as a function of 12 factors. With only 12 runs, the design correctly identified the effects of the five most important factors.

  • Enhanced Data Analysis for Reliability Evaluation

    Zhang Yi, Senior Engineer of QA Center, BOE DAS BG

    Improvement of shadow defect for a certain model: The first batch of samples encountered frequent defects in the form of shadows. A possible cause is that the new material becomes transparent too quickly when affected by internal temperature and input current, resulting in black shadow appearing on the edge. The model fails to meet the required service life. The service life is predicted through data analysis and modeling of improved products for a quantitative observation of the improvement.

  • Simulated Application of Statistics in Chemical Engineering

    Lee Li, Continuous Improvement Manager, Perfetti Van Melle Confectionery (China) Co., LTD

    The typical output from an Ethylene Amin (EA) product line is a slate consisting of different homologues (EDA, DETA, PIP, AEEA, BA-20, etc.). For inability to influence the composition of the slate in line with market demand. One of the products named PIP was oversupplied in 2016. On the other hand, another product DETA was undersupplied. If the PIP selectivity can be reduced by 0.5% and at the same time the DETA selectivity can be increased by 1%, there will be € xx million cost avoidance (cost of incinerating hazardous goods), and a generation of € xx million net profits. The first goal of this project is to set up a statistical model to see what factors are affecting slate selectivity, and thus control the parameters to increase the flexibility to match business demand.

  • JMP® Clinical to Generate Patient Narratives

    Wang Yiqun, Statistical Analysis Programmer, CStone Pharmaceuticals

    How can you view CDISC standard variables necessary for generating a JMP Clinical-based analysis report? As for non-CDSIC data, how can you quickly establish their corresponding relations with CDISC standard variables and apply them to JMP Clinical?

  • The predictive value, practical value and relevant parameters of width spread for 1,580 hot and finish rolling are used for training, verification and fitting through smart analysis and modeling of JMP Pro. The two resultant smart control models of Neural Network and Bootstrap Forest are on par with Germany’s neural network in terms of theoretical prediction and verification accuracy. With JMP Pro’s model launch feature, we have acquired its C language code, which is compiled and integrated into the online control setting model for the production site. Data of previous months are used for simulation on the standby machine. The existing model, Germany model and SAS smart model are evaluated with the same standard deviation formula. As seen from the results, the model generated by JMP Pro forest fitting is neck and neck with Baosteel’s existing model in prediction accuracy. A solid foundation has been laid for further testing and verification on the production line. The new prediction formula function released by JMP Pro is key to bring the smart model of offline fitting to field application.

  • How JMP® Improves OTA Services

    Li Mengying, Lean Six Sigma Master Black Belt, Tongcheng-Elong Holdings Co., Ltd.

    With the vision of becoming the most trustworthy online travel platform, Tongcheng-Elong works to deliver top-notch travel solutions. In 2016, Tongcheng-Elong introduced Lean Six Sigma (LSS) methodology and JMP. Li Mengying will share a case study of LSS in the internet industry – how an online travel agency incorporates JMP’s analysis and modeling tools, e.g. two-sample test, division, fit model, and multivariate fitting, into an LSS program and address pain points and challenges in the provision of online travel products and services.

  • New Opportunities in Steel Industry - Smart Manufacturing Practice and Reflection at Baosteel

    Zou Yuxian, Smart Manufacturing Deputy Director, China Baowu Steel Group Corp., Ltd.

    The steel industry currently faces many external constraints, and the traditional model of development is being challenged. As market competition intensifies and industry players give full play to their competitive advantages, the gap between industry frontrunner Baosteel and other contenders continues to narrow. How to break through the bottleneck with an "upgraded" mindset? Why is intelligent manufacturing the defining trend? How about the practices and prospects of intelligent manufacturing at Baosteel? How can Baosteel and its partners across the supply chain achieve operational efficiency while exploring the possibilities of intelligent manufacturing?

  • Quality Improvement With Statistical Analysis

    Xinming Wang, Director of Quality Management Department, North China Pharmaceutical Co., Ltd.

    Xinming Wang, Director of Quality Management Department of North China Pharmaceutical Co., Ltd., has 20 years of experience in pharmaceutical production and quality management. The senior engineer is both a licensed pharmacist and QA engineer. He will talk about the role of data analysis in quality review. By reviewing and analyzing quality indicators, product stability, public support data and flawed data, enterprises can determine whether processes are stable and reliable and whether raw materials and final products meet applicable quality standards. In this way, they can quickly identify negative trends and the necessary improvements in products and processes, so as to meet the rigorous GMP requirements on quality review.

  • Realities and Mistakes About Application of Ratio Test in Industrial Engineering

    Kang Sheng, Data Scientist, Chief Engineer of Statistics and Data Analysis, Corning (Shanghai) Management Co., Ltd.

    Kang Sheng has a master’s degree in statistics from East China Normal University, and is an IEEE member, statistics expert and data scientist. Kang has published several papers in periodicals at home and abroad, focusing on quality management, statistical analysis, data mining and reliability technology.

  • Reliability Allocation Based in JMP®

    Peng Liu, JMP Reliability R&D Expert, SAS

    Reliability allocation focuses on how to ensure system reliability by determining the reliability of individual components.  The decision-making process also takes into account other factors, for example, the cost and feasibility for higher component reliability. Mathematically, it is about restrained optimization and it aims to realize standard system reliability at minimal cost. Existing software and current literature generally assume that the cost formula is a smooth and continuous function of component reliability. The reason behind the assumption is that the optimization algorithm adopted needs the formula to be smooth and derivable which is, however, meaningless or hard to obtain in practice. Blindly using the method means that the wrong question is being solved. A more acceptable cost formula that is commonly seen is discussed here to demonstrate how JMP helps analyze the reliability allocation of two simple systems. Though the example is simple, the steps included are suitable for a complex system or a different yet practical cost formula.

  • Data Analysis in Production Recovery

    Alex Wu, FATP Product Engineering, Wistron InfoComm (Zhongshan) Corporation

    Learn ways to make the analysis much faster than normal, improve the accuracy of the judgment from experienced engineer, shorten the yield loss for production, save UPH loss, cut down on the analysis cost and more.

  • JMP® Clinical for Medical Monitoring in Clinical Trials

    Icy Du, Director of Statistical Programming, Zai Lab

    Learn ways to make the analysis much faster than normal, improve the accuracy of the judgment from experienced engineer, shorten the yield loss for production, save UPH loss, cut down on the analysis cost and more.

Back to Top