Rossella Berni

Planning and Optimization of a Numerical Control Machine in a Multiple Response Case

This paper focuses on a specific case of experimental planning and optimization in a multiresponse case. Particularly, our attention is dedicated to a numerical control machine and our final goal is to improve this machines measurement accuracy for a general dental implant. This work substantially aims at addressing two issues: the optimization methods in the presence of more response variables and the related problem of weighting according to the actual importance of these variables. About simultaneous optimization, we suggest an improvement by a new function which takes care of location and dispersion effects. Copyright

Optimization of the Soldering Process With ECAs in Electronic Equipment: Characterization Measurement and Experimental Design

With the introduction of the European Directives restrictions of hazardous substances, electrically conductive adhesives (ECAs) have received great attention in the field of electronics as a possible replacement of the traditional tin-lead soldering technology. So, in this new context, the analysis and the characterization of these alternative materials represent a fundamental topic, above all, from the industrial point of view. Nevertheless, studies on ECAs have rarely been reported in the literature, but more recently research has started to focus on this specific topic. After a comparative assessment concerning soldering materials, this paper focuses attention on their characterization through measurement in order to verify the electrical behavior of the isotropic silver conductive adhesive. In addition, since the soldering process is affected by a large number of variables such as the thickness of the conductive adhesive film, radial velocity, and curing temperature, the optimal selection of the factors is carried out through experimental design theory and the dual-response approach by means of generalized linear models. In this paper, the experimental and comparative studies on soldering made up of epoxy adhesives are carried out, in particular, the adhesives constituted by metallic particles (silver), normally in the form of flakes, in a polymer matrix are considered. The novelty of the kind of adhesive considered is the Ag filler loadings of 50-65% by volume. At these loadings, the materials achieve the percolation threshold and are electrically conductive in all directions after the materials are cured. Two different types of conductive adhesives, characterized by different chemical structures and compositions, are experimented and tested. Then, since the lead-free soldering process is characterized by several critical factors, a statistical approach is used to optimize this process. Experimental results obtained by testing samples with ECA materials prove the validity of this paper. The value of this paper is in the application of a statistical approach to these adhesive materials in order to achieve the optimization of the soldering process with a small number of treatment combinations, satisfying at the same time the stringent requirements and achieving robust electrical interconnections.

Kriging models for payload distribution optimisation of freight trains

Abstract This paper deals with Kriging models applied to optimise braking performances for freight trains. More precisely, it is focused on mass distribution optimisation aimed at reducing the effects of in-train forces among vehicles, e.g. compression and tensile forces, in-train emergency braking. To this end, Kriging models are applied with covariance structure based on the Matérn function, introducing specific input parameters to better outline the payload distribution on the train, also evaluating the shape of the payload distribution. The different shapes, related to the payload distributions, have been implemented into a model through a Python routine, which has been used to ‘assemble’ the simulated trains. The analysed train carries 80% of its maximum payload capacity during an emergency braking from the speed of 30 km/h. Satisfactory results have been obtained considering compression forces, tensile forces and their sum, also considering residuals and diagnositc measures.

Assessment and Optimization for Novel Gas Materials Through the Evaluation of Mixed Response Surface Models

In this paper, an innovative methodology aimed at improving the development of novel gas sensors through a process optimization is carried out by applying mixed response surface (RS) models. High accuracy measurements of new conductometric metal oxide gas sensors, obtained by an efficient control of the working conditions, are gathered. The response of metal-oxide-semiconductor gas sensors changes significantly when the sensors operate at different temperatures and target gas concentrations. To consider all the sources of variability there involved, the RS methodology was applied, including random effects, to improve and optimize the performance of these new gas sensors. More precisely, the optimization is performed exploiting a limited number of observations, systematically collected with an ad hoc measurement system, and it considers external sources of variability, satisfying at the same time stringent requirements. Furthermore, the statistical results and the relative assessment of novel gas materials are obtained by considering fixed as well as random effects, where random variables are considered for better controlling the optimization step.

Process Optimization of a Superfinishing Machine through Experimental Design and Mixed Response Surface Models

ABSTRACT This article deals with process optimization for a centrifugal compressor. More precisely, the technological problem concerns the reduction of the surface roughness of centrifugal compressor impellers through a new technology implemented by GE Oil & Gas called superfinishing. The new technology is studied through statistical methods in order to achieve a minimization of the final roughness according to the best set of levels for the abrasive component mixture and the time process. To this end, an experimental design is planned for three different materials—for example, three types of steel—and mixed response surface models are applied. The application of mixed models allows us to estimate random effects, useful for better controlling the process variance in a robust design approach. Within this framework, a random effect is the initial roughness, measured for each impeller vane before starting the superfinishing process. Furthermore, random effects are included in the final optimization step. The contribution of this article is the study of this new superfinishing process through mixed response surface models and robust design optimization, in order to set the best levels of the abrasive component mixture and time process to minimize the final roughness for a centrifugal compressor impeller.

Optimization of Perovskite Gas Sensor Performance: Characterization, Measurement and Experimental Design

Eight different types of nanostructured perovskites based on YCoO3 with different chemical compositions are prepared as gas sensor materials, and they are studied with two target gases NO2 and CO. Moreover, a statistical approach is adopted to optimize their performance. The innovative contribution is carried out through a split-plot design planning and modeling, also involving random effects, for studying Metal Oxide Semiconductors (MOX) sensors in a robust design context. The statistical results prove the validity of the proposed approach; in fact, for each material type, the variation of the electrical resistance achieves a satisfactory optimized value conditional to the working temperature and by controlling for the gas concentration variability. Just to mention some results, the sensing material YCo0.9Pd0.1O3 (Mt1) achieved excellent solutions during the optimization procedure. In particular, Mt1 resulted in being useful and feasible for the detection of both gases, with optimal response equal to +10.23% and working temperature at 312∘C for CO (284 ppm, from design) and response equal to −14.17% at 185∘C for NO2 (16 ppm, from design). Analogously, for NO2 (16 ppm, from design), the material type YCo0.9O2.85+1%Pd (Mt8) allows for optimizing the response value at −15.39% with a working temperature at 181.0∘C, whereas for YCo0.95Pd0.05O3 (Mt3), the best response value is achieved at −15.40% with the temperature equal to 204∘C.

Optimization of ADC Channels of A Smart Energy Meter Including Random Noise Effects

This paper proposes a multiresponse process optimization through mixed response surface models. The robust design approach is used by involving noise effects in the optimization step. In order to illustrate our proposal, a prototype of an energy meter, based on an open source concept, is studied. The proposed device architecture assures easy development of new applications for the imminent migration to smart grid infrastructures and simple adjustments to comply with possible changes in the international power quality standards. The measurement data of the acquisition channels are collected from signals generated using a high-accuracy waveform generation module. Satisfactory results are obtained, and the multiresponse process optimization provides useful information about the smart energy meter firmware in relation to the suitable acquisition strategy in the signal frequency range of interest. Copyright

Choices and conjoint analysis: critical aspects and recent developments

In the literature, a large number of researchers and practitioners are dealing with preference measurements which are considered as one of the most general methods in order to study and improve the consumer’s behaviour intended as the consumer’s decision about improving his/her utility in changing a service or a product. Nevertheless, a wide range of preference measurements’ methods is defined according to the specific aim of the research, or of the application, and the basic theoretical elements involved therein.

Measurement Error Models for Interlaboratory Comparison Measurement Data

Test laboratories with International Organization for Standardization/International Electrotechnical Commission 17025:2005 accreditation are obliged to calculate measurement uncertainty and declare the calculated value. Furthermore, they have to ensure the quality of the test results, and their participation in interlaboratory comparisons is mandatory for the accreditation. To this end, a standard procedure is available, and the laboratorys performance is also assessed by comparing its results with the reference value. While several studies consider the problem of analyzing interlaboratory comparison data, the problem still remains of how to include all the measurements (containing uncertainties and outliers) and all the dispersion effects arising during the test activity, in the analysis. This paper aims to improve the analysis of interlaboratory comparison data by focusing on an error measurement model, which considers the declared measured values and the corresponding uncertainties, and by also accounting for other dispersion effects involved in the interlaboratory activity. The problems of the small sample size and the presence of outliers are taken into account through the calculation of confidence intervals, by also evaluating the contribution of the variances estimated for the uncertainties, namely, by the signal-to-noise and reliability ratios. Moreover, the laboratorys performance is assessed by discriminating for the presence of outliers related to the reference value and/or to the uncertainty. The results are satisfactory in view of the issues addressed in this study, especially if we consider the specific kind of data. Copyright

A Comparison of Alloy-Surface Finish Combinations Considering Different Component Package Types and Their Impact on Soldering Reliability

Reliability specifications for solder joints, as well as for all electronic components, have become a fundamental feature in the qualification of an electronic product. The relevance of these reliability features increases if new components or materials are considered. In this research activity, an accelerated thermal test on customized electronic boards was implemented for an early reliability evaluation; we therefore proposed a study on the reliability behavior of a solder joint by considering different surface finishes, and several component packages. A comparative study was carried out through the application of statistical methods. To this end, Weibull distributed data and non-linear mixed models were evaluated. More precisely, Weibull random-effects models were applied to compare different combinations of surface finishes, (e.g. Hot Air Solder Leveling, Electroless Nickel Immersion Gold, Immersion Tin) and alloys, (e.g. tin-silver-copper, tin-lead), connected to four types of components, also to evaluate how the type of package or the geometry of the joint may affect the reliability of the soldering. Therefore, the aim of this research is a statistical study of the reliability of solder alloys subjected to thermal aging tests by also taking different surface finishes of the printed circuit boards and different types of packaging into account. The study was carried out with two-by-two comparisons of alloy-surface finishes. By evaluating the statistical results, the tin-silver-copper alloy, with the considered finishes, demonstrates a higher reliability with respect to the boards soldered by the traditional combination of soldering alloy and surface finish.