Optimal and Robust Designs of Step-stress Accelerated Life Testing Experiments for Proportional Hazards Models

Accelerated life testing (ALT) is widely used to obtain reliability information about a product within a limited time frame. The Cox s proportional hazards (PH) model is often utilized for reliability prediction. My master thesis research focuses on designing accelerated life testing experiments for reliability estimation. We consider multiple step-stress ALT plans with censoring. The optimal stress levels and times of changing the stress levels are investigated. We discuss the optimal designs under three optimality criteria. They are D-, A- and Q-optimal designs. We note that the classical designs are optimal only if the model assumed is correct. Due to the nature of prediction made from ALT experimental data, attained under the stress levels higher than the normal condition, extrapolation is encountered. In such case, the assumed model cannot be tested. Therefore, for possible imprecision in the assumed PH model, the method of construction for robust designs is also explored.

Experimental and CFD study on low temperature drying of loose and compressed bulks

Die Trocknung ist eines der am häufigsten verwendeten Verfahren, um die Haltbarkeit von landwirtschaftlichen Gütern zu verlängern. Von den vielen gebräuchlichen Trocknungsarten ist die Konvektionstrocknung, die Luft als Energie- und Feuchteträger verwendet, weiterhin die wichtigste. Trotz ihrer Nachteile und ihres hohen spezifischen Energieverbrauchs sind Satztrockner für die Getreide-, Kräuter- und Heutrocknung noch weit verbreitet. Ferner werden Konvektionstrockner gemeinhin zur künstlichen Trocknung eingesetzt. Diese Arbeit ist Teil eines Forschungsprojekts, welches eine sorptionsgestützte solare Trocknung von landwirtschaftlichen Gütern untersucht. Bei dieser wird kühle feuchte Luft in Kontakt mit einer konzentrierten hygroskopischen Salzlösung gebracht. Während dieses Prozesses wird die Luft entfeuchtet und mit Hilfe der freigesetzten Verdampfungsenthalpie erwärmt. Die Anwendung dieses Verfahrens zur Trocknung landwirtschaftlicher Güter ist besonders interessant für Produkte, die bei niedrigen Temperaturen verarbeitet werden, um ihre Qualität nicht negativ zu beeinflussen. Bei allen energieintensiven Prozessen und vor allem bei der Verwendung von Solarenergie ist ein effizienter Energieverbrauch äußerst wichtig für den technischen und ökonomischen Erfolg. Eine ungleichmäßige Trocknung ist oftmals bei Satztrocknern zu beobachten, was die Energieeffizienz negativ beeinflusst. Aus diesem Grund lag im agrartechnischen Teil des Projekts der Schwerpunkt auf der Trocknung bei niedrigen Temperaturen und geringer Luftfeuchte, die der Sorptionsprozess liefert, sowie auf der Verbesserung der Luftverteilung und der Gleichmäßigkeit der Trocknung in lockeren und verdichteten Schüttungen. Es wurden Dünnschicht-Trocknungsexperminente für Blattsellerie bei unterschiedlichen Lufttemperaturen und unterschiedlicher relativer Luftfeuchte durchgeführt. Die Ergebnisse zeigen eindeutig, dass nicht nur die Lufttemperatur einen großen Einfluss auf die Trocknungsgeschwindigkeit hat, sondern auch die relative Luftfeuchte. Dies ist vor allem bei niedrigen Temperaturen, wie sie vom Sorptionssystem geliefert werden, der Fall. Um die Luftverteilung und Gleichmäßigkeit der Trocknung lockerer Schüttungen zu untersuchen, wurde ein Kistentrockner experimentell und in Strömungssimulationen getestet. Die Leistung des ursprünglichen Trockners wurde signifikant durch einfache Veränderungen des Designs verbessert. Diese stellten einen gleichmäßigeren Luftstrom durch das Produkt sicher. Die Trocknung von Heu-Rundballen stellt eine Herausforderung dar, da diese sehr stark verdichtet sind und die Dichte innerhalb eines Ballens stark schwankt. Strömungs- und Trocknungssimulationen so wie experimentelle Messungen zeigten, dass das Trocknerdesign einen großen Einfluss auf die Gleichmäßigkeit der Trocknung hat. Bei den einfachsten Trocknervarianten war die Verteilung auch bei optimal gepressten Ballen unzureichend, während komplexere Trockner eine gleichmäßigere Trocknung erzielten. Jedoch werden auch hier die Ergebnisse stark von der Verteilung der Dichte im Ballen beeinflusst, welche in der Praxis weiterhin nicht ideal ist. Abschließend wurde, um den Trocknungsfortschritt zu überwachen, eine Methodik getestet, bei der zur Messung des durchschnittlichen aktuellen Feuchtegehalts des Produktes psychrometrische Messungen und Messungen des Luftstroms in kurzen Abständen durchgeführt wurden. Die Menge des tatsächlich entfernten Wassers stimmte in den meisten Fällen gut mit der geschätzten Menge überein. Jedoch hängt der Erfolg dieser Methode stark von der Genauigkeit der Messfühler ab.

Sequential designs for phase III clinical trials incorporating treatment selection

Most statistical methodology for phase III clinical trials focuses on the comparison of a single experimental treatment with a control. An increasing desire to reduce the time before regulatory approval of a new drug is sought has led to development of two-stage or sequential designs for trials that combine the definitive analysis associated with phase III with the treatment selection element of a phase II study. In this paper we consider a trial in which the most promising of a number of experimental treatments is selected at the first interim analysis. This considerably reduces the computational load associated with the construction of stopping boundaries compared to the approach proposed by Follman, Proschan and Geller (Biometrics 1994; 50: 325-336). The computational requirement does not exceed that for the sequential comparison of a single experimental treatment with a control. Existing methods are extended in two ways. First, the use of the efficient score as a test statistic makes the analysis of binary, normal or failure-time data, as well as adjustment for covariates or stratification straightforward. Second, the question of trial power is also considered, enabling the determination of sample size required to give specified power. Copyright © 2003 John Wiley & Sons, Ltd.

A practical comparison of group-sequential and adaptive designs

Sequential methods provide a formal framework by which clinical trial data can be monitored as they accumulate. The results from interim analyses can be used either to modify the design of the remainder of the trial or to stop the trial as soon as sufficient evidence of either the presence or absence of a treatment effect is available. The circumstances under which the trial will be stopped with a claim of superiority for the experimental treatment, must, however, be determined in advance so as to control the overall type I error rate. One approach to calculating the stopping rule is the group-sequential method. A relatively recent alternative to group-sequential approaches is the adaptive design method. This latter approach provides considerable flexibility in changes to the design of a clinical trial at an interim point. However, a criticism is that the method by which evidence from different parts of the trial is combined means that a final comparison of treatments is not based on a sufficient statistic for the treatment difference, suggesting that the method may lack power. The aim of this paper is to compare two adaptive design approaches with the group-sequential approach. We first compare the form of the stopping boundaries obtained using the different methods. We then focus on a comparison of the power of the different trials when they are designed so as to be as similar as possible. We conclude that all methods acceptably control type I error rate and power when the sample size is modified based on a variance estimate, provided no interim analysis is so small that the asymptotic properties of the test statistic no longer hold. In the latter case, the group-sequential approach is to be preferred. Provided that asymptotic assumptions hold, the adaptive design approaches control the type I error rate even if the sample size is adjusted on the basis of an estimate of the treatment effect, showing that the adaptive designs allow more modifications than the group-sequential method.

Large supersaturated designs

A supersaturated design (SSD) is an experimental plan, useful for evaluating the main effects of m factors with n experimental units when m > n - 1, each factor has two levels and when the first-order effects of only a few factors are expected to have dominant effects on the response. Use of these plans can be extremely cost-effective when it is necessary to screen hundreds or thousands of factors with a limited amount of resources. In this article we describe how to use cyclic balanced incomplete block designs and regular graph designs to construct E (s(2)) optimal and near optimal SSDs when m is a multiple of n - 1. We also provide a table that can be used to construct these designs for screening thousands of factors. We also explain how to obtain SSDs when m is not a multiple of n - 1. Using the table and the approaches given in this paper, SSDs can be developed for designs with up to 24 runs and up to 12,190 factors.

Some theory for constructing minimum aberration fractional factorial designs

Minimum aberration is the most established criterion for selecting a regular fractional factorial design of maximum resolution. Minimum aberration designs for n runs and n/2 less than or equal to m < n factors have previously been constructed using the novel idea of complementary designs. In this paper, an alternative method of construction is developed by relating the wordlength pattern of designs to the so-called 'confounding between experimental runs'. This allows minimum aberration designs to be constructed for n runs and 5n/16 less than or equal to m less than or equal to n/2 factors as well as for n/2 less than or equal to m < n.

Designs for an adaptive tuned vibration absorber with variable shape stiffness element

An adaptive tuned vibration absorber (ATVA) with a smart variable stiffness element is capable of retuning itself in response to a time-varying excitation frequency., enabling effective vibration control over a range of frequencies. This paper discusses novel methods of achieving variable stiffness in an ATVA by changing shape, as inspired by biological paradigms. It is shown that considerable variation in the tuned frequency can be achieved by actuating a shape change, provided that this is within the limits of the actuator. A feasible design for such an ATVA is one in which the device offers low resistance to the required shape change actuation while not being restricted to low values of the effective stiffness of the vibration absorber. Three such original designs are identified: (i) A pinned-pinned arch beam with fixed profile of slight curvature and variable preload through an adjustable natural curvature; (ii) a vibration absorber with a stiffness element formed from parallel curved beams of adjustable curvature vibrating longitudinally; (iii) a vibration absorber with a variable geometry linkage as stiffness element. The experimental results from demonstrators based on two of these designs show good correlation with the theory.

Sociality and external validity in experimental economics

It is sometimes argued that experimental economists do not have to worry about external validity so long as the design sticks closely to a theoretical model. This position mistakes the model for the theory. As a result, applied economics designs often study phenomena distinct from their stated objects of inquiry. Because the implemented models are abstract, they may provide improbable analogues to their stated subject matter. This problem is exacerbated by the relational character of the social world, which also sets epistemic limits for the social science laboratory more generally.

Does high-frequency repetitive transcranial magnetic stimulation produce residual and/or cumulative effects within an experimental session?

A common procedure for studying the effects on cognition of repetitive transcranial magnetic stimulation (rTMS) is to deliver rTMS concurrent with task performance, and to compare task performance on these trials versus on trials without rTMS. Recent evidence that TMS can have effects on neural activity that persist longer than the experimental session itself, however, raise questions about the assumption of the transient nature of rTMS that underlies many concurrent (or "online") rTMS designs. To our knowledge, there have been no studies in the cognitive domain examining whether the application of brief trains of rTMS during specific epochs of a complex task may have effects that spill over into subsequent task epochs, and perhaps into subsequent trials. We looked for possible immediate spill-over and longer-term cumulative effects of rTMS in data from two studies of visual short-term delayed recognition. In 54 subjects, 10-Hz rTMS trains were applied to five different brain regions during the 3-s delay period of a spatial task, and in a second group of 15 subjects, electroencephalography (EEG) was recorded while 10-Hz rTMS was applied to two brain areas during the 3-s delay period of both spatial and object tasks. No evidence for immediate effects was found in the comparison of the memory probe-evoked response on trials that were vs. were not preceded by delay-period rTMS. No evidence for cumulative effects was found in analyses of behavioral performance, and of EEG signal, as a function of task block. The implications of these findings, and their relation to the broader literature on acute vs. long-lasting effects of rTMS, are considered.

Experimental economics and the artificiality of alteration

A neglected critique of social science laboratories alleges that they implement phenomena different to those supposedly under investigation. The critique purports to be conceptual and so invulnerable to a technical solution. I argue that it undermines some economics designs seeking to implement features of real societies, and counsels more modesty in experimental write‐ups. It also constitutes a plausible argument that laboratory economics experiments are necessarily less demonstrative than natural scientific ones. More radical sceptical conclusions are unwarranted.

Influence of microgap location and configuration on the periimplant bone morphology in submerged implants. An experimental study in dogs

Objectives: The vertical location of the implant-abutment connection influences the periimplant bone morphology. It is unknown, however, whether different microgap configurations cause different bone reactions. Therefore, in this study the bone morphologies of two different implant systems were compared.Material and methods: Three months after tooth extraction in eight mongrel dogs, two grit-blasted screw implants with internal Morse taper connection (ANK group) were placed on one side whereas the contralateral side received two oxidized screw implants with external hex (TIU group). One implant on each side was placed level with the bone (equicrestal), the second implant was inserted 1.5mm below bone level (subcrestal). After 3 months the implants were uncovered. Three months after stage two surgery, histometrical evaluations were performed in order to assess the periimplant bone levels (PBL), the first bone-to-implant contact points (BICP), the width (HBD) and the steepness (SLO) of the bone defect.Results: All implants osseointegrated clinically and histologically. Bone overgrowth of the microgap was seen in ANK implants only. No significant differences between ANK and TIU could be detected in neither vertical position for PBL and BICP. However, a tendency in favor of ANK was visible when the implants were placed subcrestally. In the parameters HBD (ANK equicrestal -0.23mm; TIU equicrestal -0.51mm; ANK subcrestal +0.19mm; TIU subcrestal -0.57mm) and SLO (ANK equicrestal 35.36 degrees; TIU equicrestal 63.22 degrees; ANK subcrestal 20.40 degrees; TIU subcrestal 44.43 degrees) more pronounced and significant differences were noted.Conclusions: Within the limits of this study, it is concluded that different microgap designs cause different shapes and sizes of the periimplant ('dish-shaped') bone defect in submerged implants both in equicrestal and subcrestal positions.

Multistratum response surface designs

Response surface designs are usually described as if the treatments have been completely randomized to the experimental units. However, in practice there is often a structure to the units, implying the need for blocking. If, in addition, some factors are more difficult to vary between units than others, a multistratum structure arises naturally. We present a general strategy for constructing response surface designs in multistratum unit structures. Designs are constructed stratum by stratum, starting in the highest stratum. In each stratum a prespecified treatment set for the factors applied in that stratum is arranged to be nearly orthogonal to the units in the higher strata, allowing-for all the effects that have to be estimated. Three examples are given to show the applicability of the method and are also used to check the relationship of the final design to the choice of treatment set. Finally, some practical considerations in randomization are discussed.

Experimental RTM manufacturing analysis of carbon/epoxy composites for aerospace application

The success of manufacturing composite parts by liquid composite molding processes with RTM depends on tool designs, efficient heat system, a controlled injection pressure, a stabilized vacuum system, besides of a suitable study of the preform lay-up and the resin system choice. This paper reports how to assemble a RTM system in a laboratory scale by specifying heat, injection and vacuum system. The design and mold material were outlined by pointing out its advantages and disadvantages. Four different carbon fiber fabrics were used for testing the RTM system. The injection pressure was analyzed regarding fiber volume content, preform compression and permeability, showing how these factors can affect the process parameters. The glass transition temperature (Tg) around 203 ºC matched with the aimed temperature of the mold which ensured good distribution of the heat throughout the upper and lower mold length. The void volume fraction in a range of 2% confirmed the appropriate RTM system and parameters choice.

Use of uniform designs in combination with neural networks for viral infection process development

This work aimed to compare the predictive capacity of empirical models, based on the uniform design utilization combined to artificial neural networks with respect to classical factorial designs in bioprocess, using as example the rabies virus replication in BHK-21 cells. The viral infection process parameters under study were temperature (34°C, 37°C), multiplicity of infection (0.04, 0.07, 0.1), times of infection, and harvest (24, 48, 72 hours) and the monitored output parameter was viral production. A multilevel factorial experimental design was performed for the study of this system. Fractions of this experimental approach (18, 24, 30, 36 and 42 runs), defined according uniform designs, were used as alternative for modelling through artificial neural network and thereafter an output variable optimization was carried out by means of genetic algorithm methodology. Model prediction capacities for all uniform design approaches under study were better than that found for classical factorial design approach. It was demonstrated that uniform design in combination with artificial neural network could be an efficient experimental approach for modelling complex bioprocess like viral production. For the present study case, 67% of experimental resources were saved when compared to a classical factorial design approach. In the near future, this strategy could replace the established factorial designs used in the bioprocess development activities performed within biopharmaceutical organizations because of the improvements gained in the economics of experimentation that do not sacrifice the quality of decisions.

Caracterização experimental de sistemas mecânicos com comportamento não-linear

Pós-graduação em Engenharia Mecânica - FEIS