Using BPNN and DEMATEL to modify importance–performance analysis model – A study of the computer industry

The importance–performance analysis (IPA) model has been widely used as the primary tool for customer satisfaction management. IPA is a 2-D matrix analysis based on the importance and performance of the organization from the customer perception of quality. The firm’s customer satisfaction management strategy is formulated according to the IPA analysis results. However, both conventional and modified IPA models have important hidden assumptions: (1) assumptions regarding the importance of quality characteristics and performance; (2) the assumption that performance and satisfaction have a linear relationship; (3) that quality characteristics are mutually independent variables, with no causal relationship. Under these assumptions, if the quality characteristics cannot meet the above-mentioned assumptions, the conventional and modified IPA models will not accurately analyze the importance and priority ranking for improvement, leading to wrongful decision making. This study puts forth a new decision making and analysis methodology that will, on one hand, exploit the back-propagation neural network (BPNN) to establish quality characteristics and the hidden important integral satisfaction assumptions. The decision making trial and evaluation laboratory (DEMATEL) is used to calculate the causal relationship and extent of mutual influence among the qualities to adjust the importance of the quality characteristics and identify the core Order-Winners and Qualifiers problems. The proposed method modifies the quality importance and improves the IPA model ranking and also resolves difficult practical problems with fewer resources. This study illustrates using Taiwan industrial computer, working in conjunction with IPA models established with BPNN and DEMATEL to observe its application and effect.     

多源信息云服务质量的犹豫模糊优劣IPA评价方法

为有效评价多源信息云服务质量水平,在IPA框架下提出了一种以指标优劣水平为性能度量的犹豫模糊方法。首先,从用户满意角度构建了多源信息云服务质量评价的IPA层次结构框架;接着,以犹豫模糊集表述评价信息,基于参考解依赖评价规则,融合TOPSIS方法的最优与最劣参考解和EDAS方法的平均参考解,提出了一种以指标优劣水平为性能度量的犹豫模糊IPA分析方法,用于多源信息云服务质量评估。此外,考虑到现有的犹豫模糊距离测度无法度量不对称信息问题,构建了一种基于Squared-χ2的犹豫模糊测度。最后,以东莞A企业多源信息云服务质量评价为实例进行分析,结果表明了该方法的可行性,并通过与近似方法的对比分析可知,该方法能够有效评价多源信息云服务质量。     

Empirical comparisons of feed-forward connectionist and conventional regression models for prediction of first lactation 305-day milk yield in Karan Fries dairy cows

In this paper, two connectionist models are proposed based on different learning paradigms, viz., back propagation neural networks (BPNN) and radial basis function neural networks (RBFNN) to predict the first lactation 305-day milk yield (FLMY305) in Karan Fries (KF) dairy cattle. Also, a conventional multiple linear regression (MLR) model is developed for the prediction. In this study, all the models have been developed using a scientifically determined optimum dataset of representative breeding traits of the cattle. The prediction performances of the connectionist models are compared with that of the conventional model. This study shows that the RBFNN model performs relatively better than the MLR model. However, the BPNN model performs more or less in the close vicinity of the conventional MLR model. Hence, it is inferred that the connectionist models have potential as an alternative to the conventional models for predicting FLMY305 in KF cattle.     

An automatically constructed thesaurus for neural network based document categorization

This paper presents a method for computing a thesaurus from a text corpus, and combined with a revised back-propagation neural network (BPNN) learning algorithm for document categorization. Automatically constructed thesaurus is a data structure that accomplished by extracting the relatedness between words. Neural network is one of the efficient approaches for document categorization. However the conventional BPNN has the problems of slow learning and easy to involve into the local minimum. We use a revised algorithm to improve the conventional BPNN that can overcome these problems. A well constructed thesaurus has been recognized as valuable tool in the effective operation of document categorization, it overcome some problem for the document categorization based on bag of words which ignored the relationship between words. To investigate the effectiveness of our method, we conducted the experiments on the standard Reuter-21578. The experimental results show that the proposed model was able to achieve higher categorization effectiveness as measured by the precision, recall and F-measure.     

Stochastic model predictive control for remanufacturing system management

Uncertainties in the quality, quantity, and operational time of used products pose a challenge to the management of remanufacturing systems. In addition, it becomes a necessity to optimize the operation of the remanufacturing system to balance the quality of products, remanufacturing efficiency, and service level. In this study, a stochastic discrete-time dynamical model is proposed to represent a remanufacturing system, where the relationship between the market satisfaction, inventory status, and operational actions is explicitly modeled. This includes production and inventory planning, resource allocation and acquisition. To handle uncertainties, a stochastic model predictive control approach is proposed to plan the actions that optimize the remanufacturing efficiency. Our results in the simulation examples show that: (a) without supplies, the remanufacturing system has better stability and robustness than a conventional manufacturing system with the same initial stocks; and (b) with insufficient initial stocks, the remanufacturing system demands fewer and more gradual supplies, thereby keeping the system stable. Finally, a sensitivity analysis is conducted for testing the performance of the remanufacturing system. By changing the operational action capacity, different state equilibria are discovered, which correspond to distinct system response characteristics. The study reveals notable managerial insights and effects of product commonality, demand patterns, and operational actions scheduling on the efficiency of the remanufacturing system.     

Investigating the penalty reward calculus of software users and its impact on requirements prioritization

ContextThe current requirements engineering techniques for prioritization of software requirements implicitly assume that each user requirement will have an independent and symmetric impact on user satisfaction. For example, it is assumed that implementing a high priority user requirement will positively impact his satisfaction and not implementing a high priority user requirement will negatively impact his satisfaction. Further, the impacts of implementing multiple user requirements on his satisfaction are expected to be additive. But is this always the case?ObjectiveThis paper empirically examines whether the assumptions of symmetric and multiplicative impacts of user requirements on his satisfaction are valid. Further, the study assesses the relative efficacy of 5 methods of requirements prioritization in managing these effects as reflected by the user satisfaction with the prioritized requirement sets.MethodTo test for existence and mitigation of asymmetric effects an adaptation of the widely accepted PRCA (Penalty Reward Contrast Analysis) method was used for 5 requirements prioritization techniques. To test for existence and mitigation of multiplicative effects MHMR (Moderated Hierarchical Multiple Regression) a well-accepted technique for testing interaction effects was used.ResultsBoth asymmetric and multiplicative effects of software requirements on user satisfaction were observed for requirements prioritized using all 5 requirements prioritization methods raising questions about the efficacy of present day requirements prioritization techniques. Further, the results of the experiment led to proposing a new method for requirements prioritization for managing these effects.ConclusionThe study empirically demonstrates the complexities of prioritizing software requirements and calls for a new generation of methods to address them. Understanding and resolving these complexities will enable software providers to conserve resources by enabling them to parsimoniously selecting only those requirements for implementation in the software product that have maximum incremental impact on user satisfaction.     

Experimental investigation of conventional control strategies for a heterogeneous azeotropic distillation column

In this work, a laboratory scale sieve plate distillation column was constructed to investigate the conventional control strategies of an isopropyl alcohol (IPA), cyclohexane (CyH) and water (H₂O) heterogeneous azeotropic distillation column. Steady state process analysis showed that the optimal operation point should be located at a critical reflux, a transition point at which the distillation path switches from a route that passes through IPA+H₂O azeotrope to one that passes through IPA+CyH azeotrope. At this critical reflux, a high purity IPA product can be obtained with minimum energy consumption and maximum product recovery. However, the steady state is extremely sensitive to feed disturbances. A good control strategy must be able to maintain a steady column temperature profile that shows a plateau near 70°C to ensure passage around IPA+CyH azeotrope. In this study, an inverse double loop control strategy is recommended. Through experimental testing, the proposed control strategy was demonstrated to keep the product IPA purity at the desired high-purity level under all feed disturbance changes while other conventional control strategies fail.     

基于NMEA-BP的WSN数据流异常检测算法

针对常规BPNN(Back Propagation Neural Network)容易陷入局部最优解、收敛速度慢等问题,提出了一种基于小生境思维进化NMEA(Niche Mind Evolutionary Algorithm)及BPNN的传感器数据流异常检测算法(NMEA-BP).该算法利用NMEA的全局搜索性优化BPNN的参数,获得BPNN的最优权阈值,从而提高异常检测的准确性.为了评估算法的性能,使用因特尔伯克利实验室数据集IBRL(Intel Berkeley Research Lab)及带标记的传感网络数据集LWSNDR(Labeled Wireless Sensor Network Data Repository)完成了仿真实验,并与基于常规BPNN、支持向量机(Support Vector Machine)和极限学习机(Extreme Learning Machine)等3种异常检测算法作对比.仿真实验结果表明,与上述3种算法相比,NMEA-BP算法对各个数据集都具有较高的检测率和较低的误报率,检测率平均达到99.45％,误报率平均仅为1.45％.此外,NMEA-BP异常检测算法的模型训练时间比传统的BPNN异常检测算法平均减少30％以上.     

Comparative study on the effects of groupware and conventional technologies on the efficiency of collaborative writing

In this paper the concept of efficiency in collaborative writing is considered in detail and a definition of efficiency is proposed. The definition of efficiency leads to the development of a research framework that delineates five operational measures of efficiency: (a) writing activities efficiency, (b) coordination efficiency, (c) quality of output, (d) absence of breakdowns, and (e) satisfaction with group performance. A comparative study is subsequently presented on the effects that groupware and conventional technologies have on the effciency of collaborative writing. The hypothesis is advanced that groupware can improve the efficiency of collaborative writing over conventional technologies. The results seem to support the hypothesis and indicate that (a) the groupware system examined in this study (MUCH system) offers efficiency benefits in terms of coordination, (b) MUCH users tend to face communication breakdowns while users of conventional technologies tend to face task-related breakdowns, (c) the documents produced with MUCH are of higher content quality, more coherent, and of higher rhetorical effectiveness than the documents produced with conventional technologies, and (d) the comparison of MUCH with conventional technologies shows no significant difference in terms of their effects on group performance satisfaction.     

Applying the IPA and DEMATEL models to improve the order-winner criteria: A case study of Taiwan’s network communication equipment manufacturing industry

The purpose of this study is to use the gap-analysis method to calculate the performance of quality characteristics and apply the multiple regression analysis method to establish the overall level of satisfaction and implicit importance of quality characteristics. In addition we will use the decision making trial and evaluation laboratory (DEMATEL) method to analyze the cause-effect relationship and level of influence among different quality characteristics in order to make revisions to the traditional IPA model and find the core problems that are involved with winning orders. The methodology that we propose for this research not only makes revisions to the IPA model’s method of directly using explicit information from customer responses, we also solved the influence of the cause-effect relationships of quality characteristics. For our research we analyzed the case of a company from Taiwan’s network communication equipment manufacturing industry in order to make our corrections to the IPA model and gain the benefits of using the DEMATEL model. The findings of our research show that with the corrections to the IPA and the methodology of DEMATEL we can find and correct the core problems of the company that was analyzed, improve quality control and therefore improve the company’s ability to win and be compatible for orders.     

Understanding factors affecting trust in and satisfaction with mobile banking in Korea: A modified DeLone and McLean’s model perspective

As mobile technology has developed, mobile banking has become accepted as part of daily life. Although many studies have been conducted to assess users’ satisfaction with mobile applications, none has focused on the ways in which the three quality factors associated with mobile banking – system quality, information quality and interface design quality – affect consumers’ trust and satisfaction. Our proposed research model, based on DeLone and McLean’s model, assesses how these three external quality factors can impact satisfaction and trust. We collected 276 valid questionnaires from mobile banking customers, then analyzed them using structural equation modeling. Our results show that system quality and information quality significantly influence customers’ trust and satisfaction, and that interface design quality does not. We present herein implications and suggestions for further research.     

Optimization Design of a Webpage Based on Kansei Engineering

Webpage design has become an important component affecting user satisfaction when they surf the Internet. Interface designers are struggling to improve the quality of user experience by designing webpages that meet users’ emotional needs. An optimization design method of webpage interface is proposed in this study based on Kansei engineering theory, and a job‐hunting website homepage is taken as the research example. After determining the materials, the semantic differential (SD) method is used to extract user‐centered emotional dimensions, and the key design factors of homepages’ appearance are acquired. Next, based on the obtained semantic differential evaluation data, back propagation neural network (BPNN) is conducted to identify quantitative relations between key design factors and emotional dimensions. Finally, genetic algorithm (GA) is employed to search for a near‐optimal design. The proposed method is helpful to design webpages that can satisfy participants’ emotions. It can also be used in a variety of design cases.     

Intelligent fingerprint quality analysis using online sequential extreme learning machine

Because the quality of fingerprints can be degraded by diverse factors, recognizing the quality of fingerprints in advance can be beneficial for improving the performance of fingerprint authentication systems. This paper proposes an effective fingerprint quality analysis approach based on the online sequential extreme learning machine (OS-ELM). The proposed method is based not only on basic fingerprint properties, but also on the physical properties of the various sensors. Instead of splitting a fingerprint image into traditional small blocks, direction-based segmentation using the Gabor filter is used. From the segmented image, a feature set which consists of four selected independent local or global features: orientation certainty, local orientation quality, consistency, and ridge distance, is extracted. The selected feature set is robust against various factors responsible for quality degradation and can satisfy the requirements of different types of capture sensors. With the contribution of the OS-ELM classifier, the extracted feature set is used to determine whether or not a fingerprint image should be accepted as an input to the recognition system. Experimental results show that the proposed method performs better in terms of accuracy and time consumed than BPNN-based and SVM-based methods. An obvious improvement to the fingerprint recognition system is achieved by adding a quality analysis system. Other comparisons to traditional methods also show that the proposed method outperforms others.     

TFT‐LCD driver IC with embedded non‐volatile memory for portable applications

Abstract— To improve the display quality and yield of the TFT‐LCD driver IC, non‐volatile multiple‐time‐programmable (MTP) memory, which consists of an EEPROM cell and our proposed sense amplifier and power control circuit (SP), was integrated into a TFT‐LCD driver IC. The proposed SP has a 30% smaller layout area and a 18% faster response time compared to that of the conventional SP. The proposed SP also has lower power consumption because it does not use a static current. The TFT‐LCD quality was also improved by tuning the characteristics of the driver IC and the panel with the VREF, OSC, and VCOM blocks, using non‐volatile MTP memory. When the display quality improved, the yield also improved, along with a reduction in the failure ratio of the display module, which consists of the driver IC and the panel. As a result, the TFT‐LCD driver IC with the non‐volatile MTP memory demonstrated improved display quality and a higher yield compared to conventional driver ICs without such a memory.     

A new cloud computing method for establishing asymmetric cycle time intervals in a wafer fabrication factory

Estimating the cycle time of each job in a wafer fabrication factory is critical. An equally critical problem is to estimate the range of a cycle time. This topic has rarely been discussed because most existing methods for range calculation establish only a symmetric range. However, a symmetric range does not meet the requirements of managerial applications based on the lower and upper bounds of a cycle time. Recently, a few researchers have attempted to establish the asymmetric bounds of a cycle time. However, these methods either have overly complex computations or do not consistently perform well. This study proposes a new approach for effective cycle-time bounding. First, construction of a back propagation network predicts the cycle time of a job. Second, two linear functions of the output are formed, and the threshold on the output node is fuzzified to derive the lower and upper bounds of the cycle time. In theory, such a treatment tightens the lower and upper bounds and improves the forecasting precision. Third, theorems for the feasibility of the two linear functions are proved. Fourth, a cloud computing scheme is proposed to improve the bounds in an effective manner. Finally, a real case illustrates the applicability of the proposed methodology. Experimental results show that this methodology narrows the range of cycle times for untrained data by 31 %, while maintaining a considerably high hit rate of 92.5 %.     

Asymmetric Principal Component and Discriminant Analyses for Pattern Classification

This paper studies the roles of the principal component and discriminant analyses in the pattern classification and explores their problems with the asymmetric classes and/or the unbalanced training data. An asymmetric principal component analysis (APCA) is proposed to remove the unreliable dimensions more effectively than the conventional PCA. Targeted at the two-class problem, an asymmetric discriminant analysis in the APCA subspace is proposed to regularize the eigenvalue that is, in general, a biased estimate of the variance in the corresponding dimension. These efforts facilitate a reliable and discriminative feature extraction for the asymmetric classes and/or the unbalanced training data. The proposed approach is validated in the experiments by comparing it with the related methods. It consistently achieves the highest classification accuracy among all tested methods in the experiments.     

Working for one penny: Understanding why people would like to participate in online tasks with low payment

The development of Internet technology has facilitated the emergence of online marketplace for various kinds of tasks (e.g., Amazon’s Mechanical Turk in USA and Taskcn.com in China). Although the payment is relatively low, numerous people participate in the tasks in these online marketplaces. Drawing upon shopping value literature, we develop a research model by considering various value perceptions so as to better understand this interesting phenomenon. Specifically, it is proposed that hedonic value and utilitarian value (including benefit-cost ratio and net benefit) influence satisfaction and continuance intention. A field survey with 205 participants in an online working website in China was carried out to examine the proposed research model and hypotheses. The key findings are: (1) benefit-cost ratio significantly influences satisfaction, while net benefit does not; (2) there is a significant substitutive effect or negative interaction effect between benefit-cost ratio and net benefit; (3) hedonic value directly influences continuance intention, as well indirectly influences continuance intention via satisfaction. Implications, limitations, and future research directions are discussed.     

A modified Smith predictive scheme based back-propagation neural network approach for FOPDT processes control

In this paper, a novel control scheme to deal with process uncertainties in the form of disturbance loads and modelling errors, as well as time-varying process parameters is proposed by applying the back-propagation neural network (BPNN) approach. A BPNN predictive controller that replaces the entire Smith predictor structure is initially trained offline. Lyapunov direct method is used to prove that the convergence of this BPNN is guaranteed by selecting a suitable learning rate during the learning process. However, the Smith predictor based BPNN control is an off-line training based algorithm, which is a time consuming method and requires a known process plant input from the controller. A desired control input to the process is difficult to obtain for the training of the network. As a result a group of proper training data (target control inputs and outputs) can hardly be provided. In order to overcome this problem, a BPNN with an on-line training algorithm is introduced for the control of a First Order plus Dead Time (FOPDT) process. The stability analysis is carried out using the Lyapunov criterion to demonstrate the network convergence ability. Simulation results show that this proposed online trained neural Smith predictor based controller provides excellent robustness to process modelling errors and disturbance loads, and high adaptability to time varying processes parameters.     

An efficient CMAC neural network for stock index forecasting

Stock index forecasting is one of the major activities of financial firms and private investors in making investment decisions. Although many techniques have been developed for predicting stock index, building an efficient stock index forecasting model is still an attractive issue since even the smallest improvement in prediction accuracy can have a positive impact on investments. In this paper, an efficient cerebellar model articulation controller neural network (CAMC NN) is proposed for stock index forecasting. The traditional CAMC NN scheme has been successfully used in robot control due to its advantages of fast learning, reasonable generalization capability and robust noise resistance. But, few studies have been reported in using a CMAC NN scheme for forecasting problems. To improve the forecasting performance, this paper presents an efficient CMAC NN scheme. The proposed CMAC NN scheme employs a high quantization resolution and a large generalization size to reduce generalization error, and uses an efficient and fast hash coding to accelerate many-to-few mappings. The forecasting results and robustness evaluation of the proposed CMAC NN scheme were compared with those of a support vector regression (SVR) and a back-propagation neural network (BPNN). Experimental results from Nikkei 225 and Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) closing indexes show that the performance of the proposed CMAC NN scheme was superior to the SVR and BPNN models.     

An on-line learning neural controller for helicopters performing highly nonlinear maneuvers

This paper presents an on-line learning adaptive neural control scheme for helicopters performing highly nonlinear maneuvers. The online learning adaptive neural controller compensates the nonlinearities in the system and uncertainties in the modeling of the dynamics to provide the desired performance. The control strategy uses a neural controller aiding an existing conventional controller. The neural controller is based on a online learning dynamic radial basis function network, which uses a Lyapunov based on-line parameter update rule integrated with a neuron growth and pruning criteria. The online learning dynamic radial basis function network does not require a priori training and also it develops a compact network for implementation. The proposed adaptive law provides necessary global stability and better tracking performance. Simulation studies have been carried-out using a nonlinear (desktop) simulation model similar to that of a BO105 helicopter. The performances of the proposed adaptive controller clearly shows that it is very effective when the helicopter is performing highly nonlinear maneuvers. Finally, the robustness of the controller has been evaluated using the attitude quickness parameters (handling quality index) at different speed and flight conditions. The results indicate that the proposed online learning neural controller adapts faster and provides the necessary tracking performance for the helicopter executing highly nonlinear maneuvers.