Design of a product quality control system based on the use of data mining techniques

In recent years industries have increasingly used customer relationship labor in management as a tool to improve their position in the marketplace. This research involves using a data warehouse, decision-tree-based data mining, and neural network pattern classification analysis to isolate the causes of non-conformity in IC packaging. The correctness of the classification analysis produced using the two methods is compared. Our objective is to establish an information analysis system, that is able to quickly identify the causes of problems thereby reducing the time taken to solve quality-related problems. It is shown that predictions made about the target group using decision tree analysis are more accurate than those made by neural network classification, indicating that decision tree analysis is an effective means of classification analysis of a company's quality problems.     

Dynamic properties of carbon black filled chlorinated polyethylene/hindered phenol blends

Aggregates of carbon black (CB) in a polymer matrix have a tendency to form a CB network. The dynamic mechanical properties of binary systems of chlorinated polyethylene (CPE) and CB or 3,9‐bis{1,1‐dimethyl‐2[β‐(3‐tert‐butyl‐4‐hydroxy‐5‐methylphenyl)propionyloxy]ethyl}‐2,4,8,10‐tetraoxaspiro[5,5]‐undecane (AO‐80) and their ternary systems were investigated. It was found that the dynamic mechanical properties of those systems depend on the colloidal properties, surface oxides, and surface modification of CB. For binary systems of CPE and CB, oxidized CB gives a high modulus at low strain amplitude and a large Payne effect compared with untreated CB. In contrast, the reverse effect was observed for their ternary systems. Consequently, a good micro‐dispersion is obtainable by surface modification due to physical adsorption of AO‐80 on oxidized CB particles via hydrogen bonds. © 2003 Society of Chemical Industry     

走向IMS的PS域核心网规划建设方案

在3G牌照正式发放之前，我们仍然需要为新一轮大规模网络建设继续做准备。3G网络和现有2G网络如何互连互通，数据用户如何在3G网络和2G网络之间进行无缝连接，如何使系统网络方便地演进到IMS构架，这些问题都需要大家思考。一个经济有效且易于演进的网络结构，将对运营商顺利开展新一代移动业务，快速抢占发展的制高点，起到相当关键的作用。     

Effects of α‐form or β‐form nuclei on polymorphic crystalline morphology of poly(butylene adipate)

The effects of α‐form and β‐form nuclei on polymorphic morphology of poly(butylene adipate) (PBA) upon recrystallization from the molten state up to various T_max values were examined by differential scanning calorimetry (DSC), wide‐angle X‐ray diffraction (WAXD) and polarized light microscopy (PLM). In this study, PBA with complex melting and polymorphism behaviour was used as a model for examining different types and extents of residual nuclei. As the PBA initially containing the sole α‐crystal was brought to a molten state of various T_max, the extents of trace α‐form crystal nuclei varied and were dependent on T_max. Furthermore, it did not matter whether, initially, the PBA contained α‐ or β‐form crystals (or both) because only a single type of α‐nuclei could be left upon treatment to the molten liquid state at T_max. Therefore, only the α‐crystal in PBA had ‘memory capacity’ in the molten liquid state while the β‐crystal did not. This was so because the latter had been completely transformed into the solid state prior to being heated into a liquid. PBA crystallized before α‐nuclei could be packed into α‐crystal, regardless of the crystallization temperature (T_c). For recrystallization from molten PBA without any nuclei, the crystalline polymorphism was correspondingly influenced by T_c. Copyright © 2005 Society of Chemical Industry     

MLR rule for selecting the most reliable design

One of the most interesting decision problems is how to select the most reliable design from among k competing designs. Under a Type II censoring plan, this paper constructs an MLR (modified likelihood ratio) rule associated with a simple algorithm to compute the sample size, number of failures and critical value which are called for by this rule. Besides, the performance of this selection rule was subject to several criteria to compare with the intuitive selection rule. It is seen that the MLR selection rule is better than the intuitive selection rule.     

Application of the blind source separation method to feature extraction of machine sound signals

As the result of vibration emission in air, a machine sound signal carries important information about the working condition of machinery. But in practice, the sound signal is typically received with a very low signal-to-noise ratio. To obtain features of the original sound signal, uncorrelated sound signals must be removed and the wavelet coefficients related to fault condition must be retrieved. In this paper, the blind source separation technique is used to recover the wavelet coefficients of a monitored source from complex observed signals. Since in the proposed blind source separation (BSS) algorithms it is generally assumed that the number of sources is known, the Gerschgorin disk estimator method is introduced to determine the number of sound sources before applying the BSS method. This method can estimate the number of sound sources under non-Gaussian and non-white noise conditions. Then, the partial singular value analysis method is used to select these significant observations for BSS analysis. This method ensures that signals are separated with the smallest distortion. Afterwards, the time-frequency separation algorithm, converted to a suitable BSS algorithm for the separation of a non-stationary signal, is introduced. The transfer channel between observations and sources and the wavelet coefficients of the source signals can be blindly identified via this algorithm. The reconstructed wavelet coefficients can be used for diagnosis. Finally, the separation results obtained from the observed signals recorded in a semi-anechoic chamber demonstrate the effectiveness of the presented methods .     

Nucleotide sequence and genome structure of mink enteritis virus

Diabetes mellitus is a common disease. It affects multiple organ systems. Adverse effects of hyperglycemia on infection, fracture healing, and bone remodeling have been recently reported. This study was conducted to evaluate the clinical and radiographic results of 93 total hip arthroplasties in 78 consecutive patients with diabetes. All femoral components were cemented using contemporary cementing techniques. Prophylactic antibiotics were given in each case. The mean follow-up period was 4.1 years (range, 2-6.5 years). Ninety-six percent of the hips were rated excellent or good. Radiolucencies were observed in only 3.7% of the stems, while 22.2% of the cups showed radiolucencies. There was a 4% revision incidence. There was no postoperative infection in this series--a distinct improvement from previously reported series. However, complications remained high at 24.3%. The most frequent complication was urinary tract infection (14.2%). The most serious complication was myocardial infarction. The authors believe total hip arthroplasty can be safely performed in patients with diabetes, provided that adequate medical and follow-up evaluations are performed. The medium-term clinical and radiographic follow-up evaluations are not inferior to reported series in patients without diabetes.     

An efficient approach for calculating the dispersions of photonic-crystal fibers: design of the nearly zero ultra-flattened dispersion

A novel and efficient approach is proposed to calculate the dispersions of the guided modes of the photonic-crystal fibers (PCFs). Based on the vector boundary-element method (VBEM), the surface integral equations for the first and second derivatives of the propagation constants with respect to the wavelength are explicitly derived. Compared with the three-point finite-difference approach, which needs to solve and search three effective indexes near the interested wavelength, this approach can determine the dispersions of the PCFs by only solving one effective index at this wavelength based on the derived formulations. This novel approach saves over 60% computing time without losing the accuracy. Based on this approach, a novel four-ring PCF is designed by optimizing only three geometrical parameters to achieve the nearly zero ultra-flattened dispersion property. Compared with previously presented dispersion-flattened PCFs, the design procedure for the four-ring structure could be more efficient and easier because relatively lesser parameters need to be optimized.     

Application of artificial neural networks to the analysis of two‐dimensional fluorescence spectra in recombinant E coli fermentation processes

A two‐dimensional (2D) spectrofluorometer was used to monitor various fermentation processes with recombinant E coli for the production of 5‐aminolevulinic acid (ALA). The whole fluorescence spectral data obtained during a process were analyzed using artificial neural networks, ie self‐organizing map (SOM) and feedforward backpropagation neural network (BPNN). The SOM‐based classification of the whole spectral data has made it possible to qualitatively associate some process parameters with the normalized weights and variances, and to select some useful combinations of excitation and emission wavelengths. Based on the classified fluorescence spectra a supervised BPNN algorithm was used to predict some of the process parameters. It was also shown that the BPNN models could elucidate some sections of the process's performance, eg forecasting the process's performance. Copyright © 2005 Society of Chemical Industry