Improving forecasting performance by employing the Taguchi method

To satisfy the volatile nature of today’s markets, businesses require a significant reduction in product development lead times. Consequently, the ability to develop precise product sales forecasts is of fundamental importance to decision-makers. Over the years, many forecasting techniques of varying capabilities have been introduced. The precise extent of their influences, and the interactions between them, has never been fully clarified, although various forecasting factors have been explored in previous studies. Accordingly, this study adopts the Taguchi method to calibrate the controllable factors of a forecasting model. An L₉(3⁴) inner orthogonal array is constructed for the controllable factors of data period, horizon length, and number of observations required. An experimental design is then performed to establish the appropriate levels for each factor. At the same time, an L₄(2³) outer orthogonal array is used to consider the inherited parameters of forecasting method as the noise factors of Taguchi method simultaneously. An illustrated example, employing data from a power company, serves to demonstrate the thesis. The results show that the proposed model permits the construction of a highly efficient forecasting model through the suggested data collection method.     

Chemical diffusion of oxygen in tin dioxide: Effects of dopants and oxygen partial pressure

Tin dioxide SnO_2−δ is a pronounced n-type electron conductor due to its oxygen deficiency. This study investigates the rate of chemical diffusion of oxygen in SnO_2−δ single crystals, which is a crucial step in the overall stoichiometry change of the material. The chemical diffusion coefficient D^δ was determined from conductivity- and EPR-relaxation methods. The temperature dependence was found to be . The dependence on crystal orientation, dopant content and oxygen partial pressure was below experimental error. The latter observation leads to the conclusion that the chemical diffusion coefficient is close to the diffusion coefficient of oxygen vacancies. Along with the relaxation process resulting from the chemical diffusion of oxygen, additional processes were observed. One of these was attributed to complications in the defect chemistry of the material. The relevance of the results for the kinetics of drift processes of Taguchi sensors is discussed.     

Optimization of alkaline protease production by Bacillus sp. using taguchi methodology

Optimization of alkaline protease production parameters by Bacillus sp. was investigated using Taguchi methodology. The pH of the medium was observed to be the most significant factor among all selected optimization parameters at an individual level. The combinatorial influence of least significant factors, inoculum level and salt solution concentration (at the individual level), resulted in an interacting severity index of 76%, suggesting their interactive role in the regulation of protease production in this microbial species. Protease production could be improved more than 100% with Taguchi’s optimized conditions of the medium composition by this microorganism.     

Fabrication and Optimization of Poly(vinyl alcohol)/Zirconium Acetate Electrospun Nanofibers Using Taguchi Experimental Design

This paper presents an investigation regarding poly(vinyl alcohol)/zirconium acetate (organic–inorganic) (PVA/Zrace) nanofibers prepared by electrospinning which could be used as a precursor for fabricating ceramic metal oxide nanofibers. The effect of some processing variables, including polymer solution concentration, tip to collector distance and applied voltage of electrospinning, and the amount of Zrace and their interactions, on the diameter of the nanofibers were studied. Taguchi experimental design and a statistical analysis (ANOVA) were employed and the relationship between experimental conditions and yield levels determined. It was concluded that to obtain a narrow diameter distribution as well as maximum fiber fineness, a polymer concentration of 10 wt%, tip to collector distance of 18 cm and applied voltage of 20 kV variables were the optimum. Furthermore, it was also concluded that the ratio of Zrace (6 g) to PVA solution (10% wt) played an important role for achieving the minimum fiber diameter. Under these optimum conditions, the diameters of the electrospun composite fibers ranged from 86 nm to 381 nm with a diameter average of 193 nm. The experiments were done with Qualitek-4 software with “smaller is better” as the quality characteristics. The optimized conditions showed an improvement in the fibers diameter distribution and the average fibers diameter showed good resemblance with the result predicted using the Taguchi method and the Qualitek-4 software. The ANOVA results showed that all factors had significant effects on the fibers diameter and distribution, but the effect of PVA concentration and zirconium acetate were more significant than the other factors.     

Experimental Design Approach to Investigate the Effects of Operating Factors on the Surface Tension,Viscosity, and Stability of Heavy Crude Oil-in-Water Emulsions

In this article, the effects of various operating factors on the surface tension, viscosity, and stability of two heavy oil types in water emulsions for pipeline transportation are studied using the Taguchi experimental design approach. The surface tension of heavy crude oil-in-water emulsion is decreased by increasing the emulsifier concentration while the stability of emulsions is increased. The viscosity and stability are increased by an increase in oil content. An increase in the salinity and mixing speed leads to an increase in the stability of emulsion.     

Effect of hybrid fillers on the mechanical behavior of polypropylene based hybrid composites

The present study investigates the effect of hybrid fillers such as graphene nanoplatelets (GnPs) and Titanium di-oxide (TiO₂) in polypropylene (PP) composites on the mechanical properties. The compatibilizing agent of Maleic anhydride grafted polypropylene (MAPP) are used in the polypropylene based composites to increase the interfacial adhesion between matrix and fillers. The experiments are designed according to L₁₆ orthogonal array (OA) based design of experiments (DOE). The parameters selected for this study are GnPs, TiO₂ and MAPP with four different levels are used.By using Orthogonal array and Taguchi based experimental design, the performance characteristics of tensile modulus, tensile strength, elongation at break and toughness can be analyzed with more objective through a small set of experiments.Taguchi based analysis are used to find out the optimal parameters to maximize the tensile properties for the GnPs and TiO₂ reinforced PP hybrid composites. Further, analysis of variance (ANOVA) is investigated to identify the most significant parameters which influence the mechanical properties.From the analysis it was found that the optimal parameters of 3 ?wt% GnPs, 2 ?wt% TiO₂ and 6 ?wt% MAPP for maximum tensile modulus and tensile strength. The most significant parameter for tensile modulus and tensile strength is GnPs followed by TiO₂ and MAPP according to ANOVA analysis.     

Laser welding of thin foil nickel-titanium shape memory alloy

In this study, two L27 Taguchi experiments were carried out to study the effect of fibre laser welding parameters and their interactions upon the weld bead aspect ratio of nickel-titanium thin foil. The optimum parameters to produce full penetrated weld with the largest aspect ratio and desirable microstructure were successfully obtained by the Taguchi experimental design. The corrosion property of the optimized NiTi weld in Hank’s solution at 37.5 °C was studied and compared with the as-received NiTi. To improve the corrosion properties of the weld, the effect of post-weld-heat-treatments ranging from 573 to 1173 K was investigated. The corrosion properties, surface morphology, microstructure and Ti/Ni ratio of the heat-treated NiTi weld were analysed. It was found that a post-weld heat treatment at 573 K for 1 h provided the best pitting corrosion resistance at the weld zone.     

Prediction models for the key mechanical properties of EPDM/PP blends as affected by processing parameters and their correlation with stress relaxation and phase morphologies

The optimum condition of processing parameters (mixing temperature, rotor speed, fill factor, and blend ratio) and prediction models for the best key mechanical properties of ethylene propylene diene terpolymer/polypropylene thermoplastic vulcanizates (EPDM/PP TPVs) was investigated by using the Taguchi's optimization technique and data analysis. The results reveal that all of the processing parameters affected significantly the mechanical properties of the EPDM/PP TPVs, but specifically the blend ratio contributed more than 90% in effect size on tensile strength and tension set. There were three main factors, the mixing temperature, the fill factor, and the blend ratio, influencing the elongation at break. Furthermore, the mathematic models were effective and reliable in predicting the properties of TPVs. The correlation of mechanical properties, stress relaxation, and phase morphologies of the TPVs prepared from the predicted models was also investigated. It can be summarized that the morphological structure and stress relaxation of the TPVs were strongly governed by the EPDM content in the blend ratio. That is, the higher the EPDM content, the better phase morphology having smaller size of the vulcanized EPDM particles distributed in the PP matrix and the higher rate of stress relaxation. Moreover, these two properties were then principally pushing the mechanical characteristics of the EPDM/PP TPVs. Copyright © 2015 John Wiley & Sons, Ltd.     

Prediction and simulation of wear response of Linz–Donawitz (LD) slag filled glass–epoxy composites using neural computation

This article reports on the implementation of a soft computing technique based on artificial neural networks (ANNs) in analyzing the wear performance of a new class of hybrid composites filled with Linz–Donawitz slag (LDS). LDS is a major solid waste generated in huge quantities during steel making. It comes from slag formers such as burned lime/dolomite and from oxidizing of silica, iron etc. while refining the iron into steel in the LD furnace. In this work, hybrid composites consisting of short glass fiber (SGF) reinforced epoxy filled with different LDS content (0, 7.5, 15 and 22.5 wt%) are prepared by simple hand lay‐up technique. Solid particle erosion trials, as per ASTM G 76 test standards, are conducted on the composite samples following a well‐planned experimental schedule based on Taguchi design of experiments. Significant process parameters predominantly influencing the rate of erosion are identified. The study reveals that the LDS content is the most significant among various factors influencing the wear rate of these composites. Further, a model based on ANN for the prediction of erosion performance of these composites is implemented. The ANN prediction profiles for the characteristic wear properties exhibit very good agreement with the measured results demonstrating that a well‐trained network has been created. The simulated results explaining the effect of significant process variables on the wear rate indicate that the trained neural network possesses enough generalization capability of predicting wear rate even beyond the experimental range. Copyright © 2014 John Wiley & Sons, Ltd.     

X波段过模弯曲圆波导TM01-HE11模式变换器研究

在波束波导和反射面天线的馈源应用中, 为了产生低副瓣且方向图等化的高斯波束, 需要将高功率微波转换为准高斯模HE₁₁模辐射. 本文利用弯曲圆波导可同时从TM₀₁模产生TE₁₁模和TM₁₁模的原理, 提出了采用双弯曲过模圆波导结构直接将TM₀₁转换为HE₁₁的模式变换器, 避免了常规微波领域中首先将TM₀₁转换为TE₁₁再用波纹式或半径渐变式TE₁₁-HE₁₁转换器转换为准高斯波束功率容量不足或尺寸过长的不足. 基于模式耦合理论和Taguchi优化算法对模式变换器的弯曲半径、相移直端长度及引入位置进行了优化, 使输出的TE₁₁和TM₁₁成一定比率, 以组成HE₁₁模式, 并对设计的模式变换器进行了全电磁波仿真分析, 结果表明输出波束的标量高斯含量在9.05–9.8 GHz范围内均高于99%, 理论功率容量可达4.5 GW. 关键词： 高功率微波 模式耦合理论 Taguchi优化算法 模式变换器