This thesis presents a system to display Structural Health Monitoring (SHM) data interactively at multiple scales that range from milliseconds to years. Typically, visualizing large SHM datasets produce static plots that take significant time to render. Our system improves upon standard tools by providing an interactive interface and a speed-optimized binning algorithm. Using the interface, a user is able to view data collected from a bridge's sensors at multiple scales in a web browser. This allows a user to visually inspect the entire range of their data to see both short and long-term trends. To render the data, the system uses a binning algorithm to calculate a five-number summary of a range of data. Those bins are combined to generate increasingly high levels of bins, which are then rendered as a binned line chart. The chart is rendered using a standard web browser on both desktop and mobile devices.

Bachelor-Arbeit Univ. St. Gallen, 2005.

abstract: There has been much research involving simultaneous monitoring of several correlated quality characteristics that rely on the assumptions of multivariate normality and independence. In real world applications, these assumptions are not always met, particularly when small counts are of interest. In general, the use of normal approximation to the Poisson distribution seems to be justified when the Poisson means are large enough. A new two-sided Multivariate Poisson Exponentially Weighted Moving Average (MPEWMA) control chart is proposed, and the control limits are directly derived from the multivariate Poisson distribution. The MPEWMA and the conventional Multivariate Exponentially Weighted Moving Average (MEWMA) charts are evaluated by using the multivariate Poisson framework. The MPEWMA chart outperforms the MEWMA with the normal-theory limits in terms of the in-control average run lengths. An extension study of the two-sided MPEWMA to a one-sided version is performed; this is useful for detecting an increase in the count means. The results of comparison with the one-sided MEWMA chart are quite similar to the two-sided case. The implementation of the MPEWMA scheme for multiple count data is illustrated, with step by step guidelines and several examples. In addition, the method is compared to other model-based control charts that are used to monitor the residual values such as the regression adjustment. The MPEWMA scheme shows better performance on detecting the mean shift in count data when positive correlation exists among all variables. / Dissertation/Thesis / Ph.D. Industrial Engineering 2010

Fast and accurate recognition of the Statistical Control Chart Patterns (SPCCP) is significant for supervising manufacturing processes to accomplish better control and to make high value products. SPCCP can display eight kinds of patterns: normal, stratification, systematic, increasing trend, decreasing trend, up shift, down shift and cyclic. With the exception of the natural pattern, all other patterns indicate that the supervised manufacturing process is not performing properly and actions need to be taken to correct the problems. This research proposes new approaches, neural networks and neural-fuzzy systems, to the (SPCCP) recognition. This dissertation also investigates the use of features extracted from statistical analysis for simple patterns, and wavelet analysis for concurrent patterns as the components of the input vectors. Results based on simulated data show that the proposed approaches perform better than conventional approaches. Our work concluded that the extracted features improve the performance of the proposed recognizer systems.

Rapporten täcker vidareutvecklingen av en grafkomponent som från början kunde visa linjediagram med datapunkter bestående av reella tal. Grafkomponenten var en del av programvaran Tunnel Manager som är utvecklad av Sogeti till Atlas Copco. Tunnel Manager används i kombination med Atlas Copcos borrigg Boomer. Grafkomponenten utvidgades med stapeldiagram med flera serier, stapeldiagram med adderade serier och med cirkeldiagram. Den utvidgades även med nya datapunktstyper i form av datum och tid. Även gruppering av data för stapeldiagram lades till. Utökad information visades också när muspekaren hölls över ett diagram, ett så kallat tooltip. Zoom och panorering i diagrammen implementerades så användaren kunde granska vissa områden i detalj.  Rapporten omfattar även en utredning där det undersöktes vilken information och vilka diagram Atlas Copco hade behov av i framtiden i Tunnel Manager. Det visades sig att det fanns stort behov av att visa diverse information i diagram för att få ett bra underlag till beslutsfattning. Dels för planering av användandet av borriggen. Dels för underhåll av borriggen.  När stora mängder information samlas in behövs bra sätt att sammanfatta den på. Diagram är ett mycket bra sätt för detta ändamål. Diagrammen behöver dock följa vissa grundläggande regler för att de ska vara tillförlitliga. Bland annat att diagram som jämförs ska ha samma skala för att underlätta jämförelsen. Vidareutvecklingen av grafkomponenten tog hänsyn till dessa regler, det bidrog till att den lämpar sig att använda i produktion. / This report covers the further development of a chart component. The component could display a linechart with real number datapoints at the beginning. The chart component was part of as software called Tunnel Manager, developed by Sogeti for Atlas Copco. Tunnel Manager is used in combination with Atlas Copcos drilling rig Boomer. The charts added were barchart with support for several dataseries, stacked barchart with support for stacked dataseries and piechart. A new datapoint type for date and time was added. Grouping of data for the barcharts was also added.  Extended information was shown when the mouse pointer was held over a diagram, a tooltip. Zoom and panning in the charts was implemented, allowing the user to view some parts in detail.  The report also covers an investigation. The investigation finds out what kind of information, and what kinds of charts Atlas Copco had need of in the future in Tunnel Manager. There was a great need for displaying information in charts to get a good base for decision making. The information was needed for planning and maintenance of the drilling rigs.  When a lot of information is gathered from different sources a good way is needed for compilation and displaying of the information. Charts are a very good way of doing this. The carts need to follow a set of basic rules to be trustworthy. For example if several charts is to be compared, they need to have the same scale, to make it easier to compare. The further development of the chart component took these rules into account and it made it suitable for usage in production.

The diploma thesis is pursuing the analysis of principles on which the technical analysis is based, with focus on the relevancy. The attention will be paid to the fact whether technical analysis is able to generate successful trade signals and thus be recognized as scientific discipline. We will try to remove the subjectivity (for which is often criticized) by means of statistical methods. The diploma also includes basics of psychological analysis (especially new stances). In work, I evaluate whether technical analysis can be informative and add value to investment process. I also compare the returns based on technical analysis with Buy and Hold strategy.

Master of Science

Quando há interesse em monitorar a variância de uma característica da qualidade de interesse através de gráfico de controle por variáveis, o gráfico S2 é a alternativa mais usual. Entretanto, há situações onde mensurar a característica da qualidade é caro, consome mais tempo por unidade de inspeção, requer maior esforço dos operadores quanto à obtenção dos dados ou envolve ensaios destrutivos. Nestes casos, a classificação da variável contínua em categorias através de um dispositivo torna-se uma alternativa interessante. A avaliação pode ser mais rápida, a análise e o equipamento utilizado podem ser mais simples, de modo que o custo final da inspeção seja menor. O objetivo do trabalho é propor um gráfico de controle por atributos para monitoramento da variabilidade. Para tanto a estatística GS2 é calculada e gráfico sinaliza se GS2 > LC, LC limite de controle determinado de modo que minimize o ARL1, fixado um valor de ARL0. Como resultado a performance do gráfico GS2 é comparada ao gráfico S2 em termos de ARL1. / In cases aiming at monitoring the variance of a products quality characteristics using a variable control chart, chart S2 is the most used alternative. However, in some situations, this solution can be expensive, demand more time per individual inspected unit, demand greater efforts from operators to acquire data or involve destructive tests. In such cases, the use of a gauge measurement tool to classify the continuous variable into categories, becomes an interesting alternative. The assessment can be faster, the analysis and the tool used can be simple, resulting in less costly final inspections. This work proposes the use of an attribute control chart to monitor variability. Statistics GS2 is calculated and control chart signalize if GS2 > CL, whereas CL is the determined control limit, minimizing ARL1 for a fixed value of ARL0. GS2 control chart performance is compared to S2 chart based on ARL1.

Quando há interesse em monitorar a variância de uma característica da qualidade de interesse através de gráfico de controle por variáveis, o gráfico S2 é a alternativa mais usual. Entretanto, há situações onde mensurar a característica da qualidade é caro, consome mais tempo por unidade de inspeção, requer maior esforço dos operadores quanto à obtenção dos dados ou envolve ensaios destrutivos. Nestes casos, a classificação da variável contínua em categorias através de um dispositivo torna-se uma alternativa interessante. A avaliação pode ser mais rápida, a análise e o equipamento utilizado podem ser mais simples, de modo que o custo final da inspeção seja menor. O objetivo do trabalho é propor um gráfico de controle por atributos para monitoramento da variabilidade. Para tanto a estatística GS2 é calculada e gráfico sinaliza se GS2 > LC, LC limite de controle determinado de modo que minimize o ARL1, fixado um valor de ARL0. Como resultado a performance do gráfico GS2 é comparada ao gráfico S2 em termos de ARL1. / In cases aiming at monitoring the variance of a products quality characteristics using a variable control chart, chart S2 is the most used alternative. However, in some situations, this solution can be expensive, demand more time per individual inspected unit, demand greater efforts from operators to acquire data or involve destructive tests. In such cases, the use of a gauge measurement tool to classify the continuous variable into categories, becomes an interesting alternative. The assessment can be faster, the analysis and the tool used can be simple, resulting in less costly final inspections. This work proposes the use of an attribute control chart to monitor variability. Statistics GS2 is calculated and control chart signalize if GS2 > CL, whereas CL is the determined control limit, minimizing ARL1 for a fixed value of ARL0. GS2 control chart performance is compared to S2 chart based on ARL1.

The determination of economic control charts and the determination of specification limits with minimum cost are two different research topics. In this study, we first combine the design of economic control charts and the determination of specification limits to maximize the expected profit per unit time for the smaller the better quality variable following the gamma distribution. Because of the asymmetric distribution, we design the EWMA control chart with asymmetric control limits. We simultaneously determine the economic EWMA control chart and upper specification limit with maximum expected profit per unit time. Then, extend the approach to determine the economic variable sampling interval EWMA control chart and upper specification limit with maximum expected profit per unit time. In all our numerical examples of the two profit models, the optimum expected profit per unit time under inspection is higher than that of no inspection. The detection ability of the EWMA chart with an appropriate weight is always better than the X-bar probability chart. The detection ability of the VSI EWMA chart is also superior to that of the fixed sampling interval EWMA chart. Sensitivity analyses are provided to determine the significant parameters for the optimal design parameters and the optimal expected profit per unit time.