abstract: Statistical process control (SPC) and predictive analytics have been used in industrial manufacturing and design, but up until now have not been applied to threshold data of vital sign monitoring in remote care settings. In this study of 20 elders with COPD and/or CHF, extended months of peak flow monitoring (FEV1) using telemedicine are examined to determine when an earlier or later clinical intervention may have been advised. This study demonstrated that SPC may bring less than a 2.0% increase in clinician workload while providing more robust statistically-derived thresholds than clinician-derived thresholds. Using a random K-fold model, FEV1 output was predictably validated to .80 Generalized R-square, demonstrating the adequate learning of a threshold classifier. Disease severity also impacted the model. Forecasting future FEV1 data points is possible with a complex ARIMA (45, 0, 49), but variation and sources of error require tight control. Validation was above average and encouraging for clinician acceptance. These statistical algorithms provide for the patient's own data to drive reduction in variability and, potentially increase clinician efficiency, improve patient outcome, and cost burden to the health care ecosystem. / Dissertation/Thesis / Ph.D. Engineering 2013

Índices de capacidade mal empregados geram conclusões errôneas, comprometendo o estudo e análise do processo, prejudicando o atendimento de exigências gerenciais ou de clientes externos. Assim, o objetivo do presente trabalho é propor índices de capacidade para processos monitorados com gráficos de controle baseado em modelos de regressão conteplando limites simétricos e assimétricos. Realizou-se um mapeamento dos artigos publicados em nove periódicos no período de 2000 a 2009, visando identificar as abordagens pesquisadas sobre capacidade de processos. Um fluxograma orientativo é apresentado, com a finalidade de direcionar a escolha do tipo de gráfico de controle e índices de capacidade para processos com variáveis de resposta: (i) não correlacionadas, (ii) correlacionadas dependentes da variável de controle e (iii) autocorrelacionadas. O uso do fluxograma foi ilustrado através da aplicação do mesmo em um processo de torneamento de uma empresa de máquinas agrícolas. / Capability indices wrongly appointed generate erroneous conclusions, compromising the study and analysis of the process, jeopardizing the fulfillment of the requirements management or external customers. Therefore, the objective of this study is to propose capability indices for processes monitored with control charts based on regression models considering symmetric and asymmetric limits. A mapping of the articles published in nine journals, during the period of 2000 to 2009, was conducted to identify the different approaches surveyed on process capability. A flowchart is presented to direct the choice of control chart and capability indices for processes with response variables: (i) noncorrelated, (ii) correlated dependent control variables and (iii) autocorrelated. The use of the flowchart was illustrated by applying it to the hard turning process of an agricultural machinery company.

O presente trabalho propõe um framework que inclui a organização de procedimentos e técnicas estatísticas para a verificação da premissa e suposições dos gráficos de controle. Ao final do framework o usuário tem a indicação de qual gráfico é mais propício a condição dos dados em relação as suposições verificadas. O método é dividido em 4 fases que engloba a verificação da premissa de estacionariedade e das suposições de normalidade, independência e homocedasticidade. Procedimentos com o objetivo de adequar os dados as suposições são apresentados. Esta dissertação apresenta sugestões para solução dos problemas relacionados a violação da suposição de homocedasticidade. Descreve os principais modelos de obtenção de resíduos independentes e normal e identicamente distribuídos como solução para a violação de independência. São efetuados dois estudos de simulação Monte Carlo onde, como principais resultados, obteve-se: (i) um procedimento eficiente para verificação da premissa de que o processo encontra-se sob controle antes da implantação dos gráficos de controle e; (ii) o efeito da não normalidade na probabilidade de erros do tipo I nos gráficos X e S de Shewhart. Além disso, apresenta a relação entre tamanho de amostra e não normalidade como aspecto importante na construção de gráficos de controle do tipo X e S de Shewhart em relação ao erro do tipo I. / This paper proposes a framework that includes the organization of procedures and statistical techniques for the verification of the control chart's premise and assumptions. At the end of the framework is an indication of which chart has more favorable data condition on assumptions noted. The method is divided into 4 phases which includes verification of the stationarity premise and assumptions of normality, independence and homoscedasticity. Procedures with the goal of matching the data were been presented. This Master's work presents suggestions for solving problems related to violation of the homoscedasticity assumption. Describes the main types of models to intend get normal independent and identically distributed residuals as a solution to the violation of assumptions in the original data. Two studies are performed in Monte Carlo simulation and the main results obtained is: (i) an efficient procedure for verifying the premise that the process is under control before the implantation of control charts, (ii) the effect of non-normality in the probability of Type I error in and S Shewhart's control charts. In addition, shows the relationship between sample size and non-normality as important factor in building and S Shewhart's control charts on the error of Type I.

Durante os anos 70, os sistemas de produção evoluíram de planos de produção em massa para planos flexíveis, capazes de prover para cada consumidor produtos ou serviços diferenciados através de um processo ágil, flexível e integrado com baixo custo. O aumento de opções do menu de escolha por parte dos clientes resulta na produção de pequenos lotes de produtos e, consequentemente, dados em volume insuficiente para estimar os parâmetros do processo necessários para o monitoramento da característica de qualidade. A literatura reconhece que não existem métodos capazes de tratar o pro- blema do monitoramento de ferramentas de qualidade em sistemas de produção flexíveis e customizados. O objetivo deste trabalho é propor ferramentas de controle estatístico do processo para este tipo de ambiente. Verifica-se que cenários de produção customi- zados sujeitos à produção em pequenos lotes são suscetíveis a violações de suposições e estimativas imprecisas dos parâmetros do processo. Além disso, as ações adotadas pelos engenheiros de qualidade na fase I da implantação de cartas de controle nas empresas visitadas é conflitante com o que é sugerido na literatura. Verificação dos pressupostos de normalidade e independência não são executados, mesmo quando cartas de controle são implantadas. Os principais procedimentos de controle de qualidade do atual estado da arte são apresentados para verificar alternativas para a implantação de ferramentas de controle estatístico do processo em ambientes onde as estratégias atuais de produção causaram o aumento da necessidade de flexibilidade. Nestes casos, a característica da qualidade costuma ser a mesma, mas para diferentes produtos, o que significa que se tem uma única observação para cada produto em cada momento, no mesmo processo. Discussões sobre a viabilidade em implantar as principais metodologias neste contexto são apresentadas. Entre os métodos levantados no estado da arte, uma única abordagem ca- paz é a carta que utiliza o desvio do alvo no instante t, apresentada por Del Castillo et. al. (1996), porém não é aplicável ao problema de pesquisa apresentado nesta tese, pois exige uma fase retrospectiva de análise. Além disso, a inclusão de um novo produto ou uma mudança significativa no processo antigo em um ambiente flexível não pode ser tratada pelos métodos apresentados. Uma carta de controle multivariada de auto-inicialização baseada no filtro de Kalman para o sistema de múltiplos setups é proposta além de adap- tações da carta de auto-inicialização de Shewhart e da carta de controle de Quesenberry (1991) com média e variância desconhecidos. O modelo baseado no filtro de Kalman foi escolhido depois de um estudo de simulação que comparou o desempenho preditivo dos modelos ARIMA, Regressão PLS e Estrutural Básico. O desempenho das abordagens de controle estatístico do processo propostas foi comparado através de medidas de ARL e a análise considerou a implantação das ferramentas em uma série de dados real e em séries simuladas de ambientes sujeitos à violação das suposições básicas das cartas tradicionais. Como benchmark, foi utilizada a análise retrospectiva da carta de resíduos apresentada por Del Castillo et. al. (1996). Os resultados mostram que a hipótese de violação da nor- malidade é a que mais deteriora o desempenho das cartas de controle. Efeitos de violação da independência e da suposição de homocedasticidade não são significativos ao nível de 5%, de acordo com a análise de variância realizada. Conclui-se que, dos cinco procedimentos propostos, os adaptados obtiveram resultados similares ao benchmark, podendo ser tratados como alternativas para lidar com o problema de monitoramento da qualidade em ambientes organizados com sistemas de produção customizados. / During the 70s, production systems have evolved from mass production plans to flexible plans, able to provide for each constumer differentiated products or services with low cost through an agile, flexible and integrated process. The number increase of choi- ces results in production of small batches of products, and therefore insufficient volume data to estimate the required process parameters for the monitoring of quality characte- ristics. The literature shows that there are no methods to treat the problem of quality monitoring in flexible and customized production systems. The objective of this thesis is to propose tools for this type of environment. It is verified that customized scenarios are subject to small-batch production and are also susceptible to violation of assumpti- ons and inaccurate estimate of process parameters. In addition, actions taken by quality engineers in Phase I implementation of control charts in visited companies is conflicting with what is suggested in the literature. Verification of assumptions such as normality or independence are not performed, even when control charts are implemented. The state of art quality control procedures are presented in order to verify how to implement quality control tools in environments where the current strategies of production has increasing flexibility. In these cases, the quality characteristic is usually the same, but for different products, which means that it has an observation for each product in each time frame for the same process. Discussion on the feasibility of implementing the main methodologies are presented in this context. Among the methods considered the state of the art, the chart which uses the deviation from the target at time t, presented by Del Castillo et al. (1996) is the only that could be able to work in this problem, but shall be not applied to the research problem presented in this thesis, since it requires a Phase I of retrospective analysis. Moreover, the inclusion of a new product or significant change in old process in a flexible environment can not be treated by the methods presented. A self-start multivariate control chart based on Kalman filter for multiple setups is proposed, as well as the adaptations of the self-start Shewhart control chart and the self-start Quesenberry’s control chart (1991). The model based on the Kalman filter was chosen after a simulation study that compared the predictive performance of ARIMA, PLS Regression and Struc- tural Basic models. The performance of the proposed quality control approaches was compared with measures of ARL and the analysis considered the implementation of tools in series of real data and simulated data subject to violation of the traditional charts’ basic assumptions. As a benchmark, we used a retrospective analysis of the chart presented by Del Castillo et al. (1996). The results show that the assumption of normality violation is the one that most deteriorates the performance of control charts. Effects of independence and homoscedasticity violation of assumptions are not significant at 5%, according to the analysis of variance performed. We conclude that, among the five proposed procedures, the adapted ones obtained similar results to the benchmark and can be treated as alter- natives to deal with the problem of quality monitoring in environments with customized production systems.

The purpose of this research is to develop a statistical method to monitor the fairness of network resource distribution. The newly developed fairness score function allows users to have the same or different priority levels. Especially, this function possesses all the necessary properties required as a quality characteristic for the purpose of statistical process control. The main objective is to find the critical values for the statistical test. Monte Carlo simulation is used to find the critical values. When the users have the same priority level, a table of the critical values is given for different sample sizes and different significance levels. When the users have different priority levels, it is difficult to generate a similar table since the users’ priority levels vary. Therefore, the critical values are computed for given priority levels. In both cases, an example is given to demonstrate the approach developed in this study.

Nos últimos anos diversas áreas do setor produtivo vêm experimentando acelerados avanços tecnológicos e, desta forma, exigindo ferramentas cada vez mais específicas para monitorar e avaliar estes processos. Melhorias significativas nos processos de medições das características de qualidade têm sido observadas. Em relação ao controle de processo, em 1924, Walter Shewhart desenvolveu o conceito estatístico das cartas de controle para processos cujos dados sejam independentes e normalmente distribuídos, suposição esta que deve ser atendida para a construção das cartas de controle. Este artigo tem como objetivo mostrar a aplicação das cartas de controle estatístico de processo, no setor de pintura da Empresa Metasa, Rio Grande do Sul, Brasil, cuja característica do processo de pintura industrial analisada foi à espessura da camada de tinta. Os dados foram coletados levando em consideração, as seqüências que constituem o processo da aplicação da tinta (demãos de pintura), procurando-se oferecer uma melhoria sensível nos níveis de qualidade desse setor, com o objetivo de reduzir custos de produção. Os resultados encontrados são de grande importância para a empresa, pois foram estabelecidos os limites de controle que permitirão monitorar os processos. Para os casos onde a análise, entretanto, diagnosticou a permanência do processo fora de controle, se fez necessário o estudo das causas da variabilidade do mesmo. / In the last years many areas in the productive section have experienced fast technologic progress and, this way, demanding even more specific tools to control and assess these processes. Significant improvement in the processes of quality characteristics measurements has been noted. Regarding the process control, in 1924, Walter Shewhart (1931) developed the statistic concept of control letters to processes which data are independent and normally distributed, supposition that must be understood to control letters construction. This article has the goal to show the process statistic control letters, in the painting sector of the Company Metasa, Rio Grande do Sul, Brazil, which process characteristic of industrial painting analyzed was the paint coat thickness. The data were collected taking into consideration, the sequences that form the paint application process (painting coating), searching to offer a sensible improvement in the quality levels of this sector, with the objective of reducing production costs. The results found are of great importance to the company, since limits of control were established that will allow monitor the processes. For the cases where the analysis, otherwise, diagnosed the process permanence out of control, the study of their variability causes was necessary.

The master’s thesis is focused on the reduction of rejection rate in production of plastic bowls with the use of statistic methods. The theoretical part of this thesis describes among others Statistical Process Control, Process capability and types regulatory diagrams. The basis for the practical part is data collection over a certain period of production, from which the percentage of the rejection rate is evaluated. Experiments and corrective actions are designed to improve the process and thereby reduce scrap.

Master’s thesis describes statistical process control as a support for company management. Statistical process control methods and tools are introduced. Requirements for automotive industry are stated specifically for IATF 16949 standard and VDA 6.3 process audit method. Thesis presents current state in company Sanmina-SCI Czech Republic s.r.o regarding statistical regulation. Applies statistical process controls into the new process considering automotive requirements. It is all happening together with a new business introduction of automotive customer.

The master’s thesis is focused on process monitoring using statistical process control SPC tools. It is designed as a guide for the implementation and use of control charts in the manufacturing process of the base plates, which would ultimately be conducive to improving and stabilizing the process. The theoretical parts are the basic attitudes and practices methodology of SPC, especially classical Shewhart control charts. The next section describes the current status in Strojírny Třinec, a.s. and the manufacturing process of the base plates. Followed by application of control charts for a given quality features and analysis using statistical software. Based on the evaluation are findings and recommendations for long-term monitoring of the defined quality features.

With the ongoing transformation of modern manufacturing systems in an industry 4.0 environment, industrial actors may see great improvements with respect to quality towards a state of near zero defects. For the aerospace industry, where increased quality and reduced risk is strongly related, new technologies may be used in manufacturing to see to the increasing demands on products. The safety, as well as the manufacturing complexity of products and processes, make the collected measurement data an integral asset for enterprises within the aerospace industry. Collected data may be analysed using statistical tools and methods to improve process capability and, in extension, product quality. Communicating the need for zero defects, original equipment manufacturers demand increased capability from product and component manufacturers. Hence, zero defects are typically operationalised through exhibiting a process capability of Cpk= 2.0. In response to the challenge, GKN Aerospace need to raise the traditional process capability targets of Cpk=1.33. By employing an exploratory research strategy with a deductive approach, the thesis combines theoretical knowledge from the literature with empirical findings in a thematic analysis. The thematic analysis was conducted by employing six phases as suggested by Braun and Clarke (2006) and resulted in the identification of 50 codes from a total of 459 data extracts. Based on the empirical interview data, a framework for how zero defects is interpreted at GKN Aerospace was developed, which describes zero defects as a cycle. Taking into account that zero defects is operationalised through Cpk= 2.0, the cycle consists of six phases that start with a vision and is completed by delivering a true and reliable Cpk of 2.0. In addition, the codes from the thematic analysis were collated into a thematic mind map, focusing on key aspects of working with statistical process control (SPC) to support zero defects. Two main themes are presented in the mind map, statistical approach to improvement work; highlighting necessary aspects of statistical process control and measurability, and removing barriers for improvement; highlighting fundamental organisational barriers that impede proactive quality improvement. To support the findings and give a practical example of how process data may be presented and analysed using tools and methods within statistical process control, an SPC study was conducted on a set of data. In the SPC study, the construction and analysis of individuals Shewhart control charts and moving range charts were described in detail. These procedures provide better insights about process behaviour through statistical thinking and thus better knowledge on how to approach more proactive process improvements.