Rough set and PSO-based ANFIS approaches to modeling customer satisfaction for affective product design

Facing fierce competition in marketplaces, companies try to determine the optimal settings of design attribute of new products from which the best customer satisfaction can be obtained. To determine the settings, customer satisfaction models relating affective responses of customers to design attributes have to be first developed. Adaptive neuro-fuzzy inference systems (ANFIS) was attempted in previous research and shown to be an effective approach to address the fuzziness of survey data and nonlinearity in modeling customer satisfaction for affective design. However, ANFIS is incapable of modeling the relationships that involve a number of inputs which may cause the failure of the training process of ANFIS and lead to the ‘out of memory’ error. To overcome the limitation, in this paper, rough set (RS) and particle swarm optimization (PSO) based-ANFIS approaches are proposed to model customer satisfaction for affective design and further improve the modeling accuracy. In the approaches, the RS theory is adopted to extract significant design attributes as the inputs of ANFIS and PSO is employed to determine the parameter settings of an ANFIS from which explicit customer satisfaction models with better modeling accuracy can be generated. A case study of affective design of mobile phones is used to illustrate the proposed approaches. The modeling results based on the proposed approaches are compared with those based on ANFIS, fuzzy least-squares regression (FLSR), fuzzy regression (FR), and genetic programming-based fuzzy regression (GP-FR). Results of the training and validation tests show that the proposed approaches perform better than the others in terms of training and validation errors.     

A methodology of generating customer satisfaction models for new product development using a neuro-fuzzy approach

When developing new products it is important for design teams to understand customer perceptions of consumer products because the success of such products is heavily dependent upon the associated customer satisfaction level. The chance of a new product’s success in a marketplace is higher if users are satisfied with it. In this study, a new methodology of generating customer satisfaction models using a neuro-fuzzy approach is proposed. In contrast to previous research, non-linear and explicit customer satisfaction models can be developed with the use of the proposed methodology. An example of notebook computer design is used to illustrate the methodology. The proposed methodology was measured against the benchmark of statistical regression to determine its effectiveness. Experimental results suggested that the proposed approach outperformed the statistical regression method in terms of mean absolute errors and variance of errors.     

Dynamic modelling of customer preferences for product design using DENFIS and opinion mining

Previous studies mainly employed customer surveys to collect survey data for understanding customer preferences on products and developing customer preference models. In reality, customer preferences on products could change over time. Thus, the time series data of customer preferences under different time periods should be collected for the modelling of customer preferences. However, it is difficult to obtain the time series data based on customer surveys because of long survey time and substantial resources involved. In recent years, a large number of online customer reviews of products can be found on various websites, from which the time series data of customer preferences can be extracted easily. Some previous studies have attempted to analyse customer preferences on products based on online customer reviews. However, two issues were not addressed in previous studies which are the fuzziness of the sentiment expressed by customers existing in online reviews and the modelling of customer preferences based on the time series data obtained from online reviews. In this paper, a new methodology for dynamic modelling of customer preferences based on online customer reviews is proposed to address the two issues which mainly involves opinion mining and dynamic evolving neural-fuzzy inference system (DENFIS). Opinion mining is adopted to analyze online reviews and perform sentiment analysis on the reviews under different time periods. With the mined time series data and the product attribute settings of reviewed products, a DENFIS approach is introduced to perform the dynamic modelling of customer preferences. A case study is used to illustrate the proposed methodology. The results of validation tests indicate that the proposed DENFIS approach outperforms various adaptive neuro-fuzzy inference system (ANFIS) approaches in the dynamic modelling of customer preferences in terms of the mean relative error and variance of errors. In addition, the proposed DENFIS approach can provide both crisp and fuzzy outputs that cannot be realized by using existing ANFIS and conventional DENFIS approaches.     

A new hybrid ANFIS–PSO model for prediction of peak particle velocity due to bench blasting

In this paper, a novel hybrid approach is proposed for predicting peak particle velocity (PPV) due to bench blasting in open pit mines. The proposed approach is based on the combination of adaptive neuro-fuzzy inference system (ANFIS) and particle swarm optimization (PSO). In this approach, the PSO is used to improve the performance of ANFIS. Furthermore, a model is developed based on support vector regression (SVR) approach. The models are trained and tested based on actual data compiled from 120 blast rounds in Sarcheshmeh copper mine. To determine the accuracy and efficiency of ANFIS–PSO and SVR models, a statistical model (USBM equation) is applied. According to the obtained results, both techniques can be used to predict the PPV, but the comparison of models shows that the ANFIS–PSO model provides better results. Root mean square error (RMSE), variance account for (VAF), and coefficient of determination (R ²) indices were obtained as 1.83, 93.37 and 0.957 for ANFIS–PSO model, respectively.     

Product form design using customer perception evaluation by a combined superellipse fitting and ANN approach

Product aesthetics plays an important role in new product design and development. Product form can deliver product images and affect customer’s impression to a product. However, it is usually difficult to apply conventional approaches to represent the product form precisely and effectively for modeling the relationship between product image and customer perception. The objective of this work is to develop a computational technique for product aesthetics design so that customer perception can be taken into product form design in a more systematic and intelligent manner. To achieve this aim, a novel parametric approach is proposed to introduce design parameters such as line, size, and ratio into product design model and the technique of generalized superellipse fitting is adopted to describe the outline pattern of a product. Since customer perception on a product is highly non-linear and very difficult to be described by any traditional mathematical approaches, an artificial neural network (ANN) model is therefore established to relate the design parameters and the perceptual values for the design of a new product. A case study of mobile phone design, in which twelve numerical parameters are defined for the conceptual model, has been conducted to explain the implementation of the proposed approach. A three-layered perceptron ANN model is developed to predict the perceptual values of stylishness based on a survey using 32 mobile phone samples. The results of the case study illustrate that the proposed approach can successfully generate an optimum design of a mobile phone by applying a genetic algorithm (GA) on the trained ANN model.     

Enhanced customer requirement classification for product design using big data and improved Kano model

The classification of customer requirements (CRs) has a significant impact on the solution of product design. Existing CRs classification methods such as the Kano model and IPA model are time-consuming and inaccurate. This paper proposes a CRs classification method for product design using big data of online customer reviews of products to classify CRs accurately and efficiently. Comments of customer reviews are matched to CRs using a hierarchical semantic similarity method. Customer satisfaction degrees are defined based on emotional levels of adjectives and adverbs of customer comments using word vectors. The function implementation degree of each product is determined by specifications crawled from online products. Fitting curves are formed by defined customer satisfaction and function implementation of CRs using polynomial modeling and least square methods. Based on the slope of the fitted curves, CRs are classified to provide the minimum and maximum function implementations of CRs in each CR group to guide a product design process. The proposed method is applied in a case study of defining CRs classifications for design of upper limb rehabilitation devices. For verifying the proposed method, CRs defined by the existing methods are compared with CRs from the proposed method in design of an upper limb rehabilitation device.     

Using Kano model to differentiate between future vehicle-driving services

Services are constantly changing with the introduction of new technologies, which affect the service systems of both conventional and autonomous driving. New theories and technologies are also key factors affecting the design and development trends of service models and solutions. Major automobile manufacturers aspire to provide customers with unique services and experiences, resulting in a growing demand for systematic approaches to characterize customer behaviors and scientific methods to accurately interpret data stored in databases. This study proposes a scientific engineering and operation framework for driving services that enables conventional automobile manufacturers to re-evaluate their service models and solutions as they expand into the domain of autonomous driving, integrating customized consumer interactions and mass production efficiency to develop new technologies, and subsequently applying these technologies to innovate their driving services, form service innovation guidelines, and accelerate the development of smart applications for the automobile industry. A Kano two-dimensional model of quality was employed. A Kano questionnaire was administered to analyze consumers' perceived satisfaction concerning different service quality elements; the elements were then ranked in the order of requiring improvement to determine the elements that are essential in conventional vehicles. Finally, suggestions were proposed for improving the service quality of driving products and evaluating driver satisfaction. A total of 56 valid questionnaires were collected from potential buyers of four-door sedans. The questionnaire evaluated respondents’ perceived value and satisfaction of 30 product elements categorized into two groups (specific functions and intangible value-added services) across eight major quality dimensions (basic safety functions, multimedia entertainment systems, information and communication systems, value-added systems, active matching, automatic service systems, hardware–software integration, and customer service and support). In addition, Kano quality categories were statistically analyzed to elucidate whether significant differences existed between groups. Using the Kano quality categories, 30 design elements were classified: 10 as “attractive,” 7 as “one-dimensional,” 3 as “must-be,” 4 as “indifferent,” and 6 as “reverse.” Enterprises can effectively reduce customer dissatisfaction and enhance customer satisfaction based on the quality category of the product and the product improvement order proposed in this study.Relevance to industryThis study determined that using the Kano quality categories, enterprises can effectively reduce customer dissatisfaction and enhance customer satisfaction based on the quality category of the product and the product improvement order proposed in this study.     

Design by customer: concept and applications

Customer satisfaction can be increased by reducing the gap between what customer really needs (customer requirements) and what manufacturer can provide (product specifications). The approach of Design for Customer where products are generated by translating customer needs into product specifications (in mass production system) or into product variety (in mass customization system) is not able to give optimum satisfaction to all customers. Some customers are still forced to relax their requirements and to accept predefined product in the assortment. This study proposes a new concept of Design by Customer to increase customer satisfaction by opening maximum possible channel for customers to involve in value creation so that they are no longer only searching for goods but they can also, when necessary, involve in production cycle to specify their own design. In order to ensure the viability of the proposed concept, the integration of multi customer involvement decoupling point, product attribute analysis, crowdscreening and new manufacturing strategy are introduced in this paper. Real products of resin-based table clocks are used as practical example to verify the concept applicability and to demonstrate its merit.     

A fuzzy AHP approach to the determination of importance weights of customer requirements in quality function deployment

Quality function deployment (QFD) is an important tool in product planning that could contribute to increase in customer satisfaction and shorten product design and development time. During the QFD process, determination of the importance weights of customer requirements is a crucial and essential step. The analytic hierarchy process (AHP) has been used in weighting the importance. However, due to the vagueness and uncertainty existing in the importance attributed to judgement of customer requirements, the crisp pairwise comparison in the conventional AHP seems to be insufficient and imprecise to capture the degree of importance of customer requirements. In this paper, fuzzy number is introduced in the pairwise comparison of AHP. An AHP based on fuzzy scales is proposed to determine the importance weights of customer requirements. The new approach can improve the imprecise ranking of customer requirements which is based on the conventional AHP. Finally, an example of bicycle splashguard design is used to illustrate the proposed approach.     

A new approach to quality function deployment planning with financial consideration

Quality function deployment (QFD) is becoming a widely used customer-oriented approach and tool in product design. Taking into account the financial factors and uncertainties in the product design process, this paper deals with a fuzzy formulation combined with a genetic-based interactive approach to QFD planning. By introducing new concepts of planned degree, actual achieved degree, actual primary costs required and actual planned costs, two types of fuzzy optimisation models are discussed in this paper. These models consider not only the overall customer satisfaction, but also the enterprise satisfaction with the costs committed to the product. With the interactive approach, the best balance between enterprise satisfaction and overall customer satisfaction can be obtained, and the preferred solutions under different business criteria can be achieved through human–computer interaction.Scope and PurposeQuality function deployment (QFD) that originated in Japan in the late 1960s is a concept and mechanism for translating the ‘voice of customer’ into product through various stages of product planning, engineering and manufacturing. It has become a widely used customer-oriented approach to facilitating product design by analysing customer requirements (CRs). Determination of the target levels for the technical attributes (TAs) of a product with a view to achieving a high level of overall customer satisfaction is an important activity in product design and development.Traditional methods for QFD planning are mainly subjective, ad hoc and heuristic. They can hardly achieve global optimisation, and most of these models barely take into consideration the correlation between TAs. Moreover, most of these methods are technically one-sided without considering the design budget. However, the financial factor is also an important factor and should not be neglected in QFD planning. In addition, owing to uncertainties involved in the decision process, these deterministic methods could not formulate and solve it effectively.Taking into consideration the financial factors and uncertainties in the product design process, this paper deals with fuzzy formulation combined with a genetic-based interactive approach to QFD planning. By introducing new concepts of planned degree, actual achieved degree, actual primary costs required and actual planned costs, two types of fuzzy optimisation models are discussed in this paper. These models consider not only the overall customer satisfaction, but also the enterprise satisfaction with the costs committed to the product. With the interactive approach, the best balance between enterprise satisfaction and overall customer satisfaction can be obtained, and the preferred solutions under different business criteria can be achieved through human–computer interaction.     

Integrating customer's preferences in the QFD planning process using a combined benchmarking and imprecise goal programming model

Quality function deployment (QFD) is a customer-oriented design tool for developing new or improved products to achieve higher customer satisfaction by integrating various functions of an organization. The engineering characteristics (ECs) affecting the product performances are designed to match the customer attributes (CAs). However, from the viewpoint of the QFD team, product design processes are performed in imprecise environments, and more than one factor must be taken into account in determining the target levels of ECs, especially the limited resources and increased market competition. This paper presents an imprecise goal programming (GP) approach to determine the optimum target levels of ECs in QFD for maximizing customer satisfaction under resource limitation and considerations of market competition. Based on benchmarking data of CAs, the concept of satisfaction functions is utilized to formulate explicitly the customer's preferences and to integrate the competitive analysis of target market into the modelling and solution process. In addition, the relationships linking CAs and ECs and the ECs to each other are integrated by functional relationships. The proposed approach will be illustrated through a car door design example.     

A fuzzy QFD approach to determine supply chain management strategies in the dairy industry

The aim of this study is to identify the crucial logistics requirements and supply chain management (SCM) strategies for the dairy industry. For product or service development, quality function deployment (QFD) is a useful approach to maximize customer satisfaction. The determination of design requirements and supply chain management strategies are important issues during QFD processes for product or service design. For this reason, a fuzzy QFD methodology is proposed in this study to determine these aspects and to improve customer satisfaction. Qualitative information is converted firstly into quantitative parameters, and then this data is combined with other quantitative data to parameterize two multi-objective mathematical programming models. In the first model, the most important logistic requirements for the company are determined based on total technical importance, total cost, total feasibility and total value increment objectives, and in the second model, based on these objectives, appropriate supply chain management strategies are determined. Finally, a case study from the Turkish dairy industry is given to illustrate the proposed approach.     

Virtual design of multiengineering electronics systems

The authors describe a virtual reality prototyping (VRP) and virtual product design approach that uses a modeling strategy to design, test, and evaluate concepts and products in advance-before creating any physical design models. Their design decisions reflect solutions and practices derived from systems engineering and software engineering research. Close competition has forced companies to focus on tailoring technological products to meet customer preferences in user interface design, usability, and appearance. Often, implementing new features has assumed less importance than creating an optimal product variation for different customer segments in international markets. Modern computer-based systems result from a multidomain development process, and organizing this work requires developers to communicate, cooperate, and coordinate effectively with nonengineering teams. Early linking of product development to other company operations supports this activity. Geographically dispersed development teams must communicate and coordinate effectively. Extending the use of interactive, functional, and photorealistic 3D product models into other company domains can facilitate effective business practices that cover the entire value chain, from simulated product idea to Web-based customer support of the final product. This wider adoption of the VRP technology and process offers ample scope for further research     

Product family modeling for mass customization

With growing reliance on modeling in product development, it is imperative to describe product families in a cohesive way. In particular, mass customization calls for a close integration of product life cycle from customer recognition to delivery and services. This paper proposes a triple-view scheme for modeling product families. Technical challenges are discussed by comparing product family modeling with modeling single products. Individual modeling formalisms for different views are discussed. An example of product family modeling in power supply design is presented to illustrate the feasibility and potential of the proposed approach.     

An optimum design selection approach for product customization development

Consumer preferences and information on product choice behavior can be of significant value in the development processes of innovative products. In this paper, product customization evaluation and selection model is introduced to support imprecision inherent of qualitative inputs from customers and designers in the decision making process. Focusing on customer utility generation, an optimum design selection approach based on fuzzy set decision-making is proposed, where design attributes priority is identified from customer preferences using an analytical hierarchy process. A multi-attribute analysis diagram is developed to visualize the preference of each attribute from the expert’s group decision. Conjoint analysis is used in the product customization to focus on customer utility generation in terms of multiple criteria. The use of the decision-making method is illustrated with a case example that highlights the utility of the proposed method.     

A method for product form design of integrating interactive genetic algorithm with the interval hesitation time and user satisfaction

Due to the continuous release of new products, manufacturers are paying attention to customer-oriented design of products that meet user needs to minimize the risk of their products being rejected by the market. Due to the ambiguity of user cognition, it is difficult to accurately obtain the user's preference for individual productions. To respond to the challenge, we propose an engineering scientific research method of interactive genetic algorithm with the interval arithmetic based on hesitation and fuzzy kano model(FKM) to explore the emotional needs of users for product forms and drive product modeling evolution design. Through expert interviews, the morphological characteristics and perceptual images factors of the products attracting users are investigated. In order to identify the user's satisfaction relationship with the perceptual images, we use FKM to analyze the product image style that meets the user's kansei needs accurately and selects 5 factors which is attractive attributes. Meanwhile, we attempt to transform this 5 factors into evaluation carrier to guide the evolution direction of product styling in HIIF-IGA, and then optimized four electric bikes with scores over 8.8 so that it could realize user demand-driven product evolution design. To handle users' ambiguity, the FAHP method is used to quantify the user's emotional imagery criterion and create a product evolution design system platform, which can automatically generate product styling design scheme in line with user preferences. This experimental results show that the proposed method can help enterprises effectively improve customer satisfaction and reduce the cost and time of product development.     

A new ARIMA-based neuro-fuzzy approach and swarm intelligence for time series forecasting

Time series forecasting is an important and widely interesting topic in the research of system modeling. We propose a new computational intelligence approach to the problem of time series forecasting, using a neuro-fuzzy system (NFS) with auto-regressive integrated moving average (ARIMA) models and a novel hybrid learning method. The proposed intelligent system is denoted as the NFS–ARIMA model, which is used as an adaptive nonlinear predictor to the forecasting problem. For the NFS–ARIMA, the focus is on the design of fuzzy If-Then rules, where ARIMA models are embedded in the consequent parts of If-Then rules. For the hybrid learning method, the well-known particle swarm optimization (PSO) algorithm and the recursive least-squares estimator (RLSE) are combined together in a hybrid way so that they can update the free parameters of NFS–ARIMA efficiently. The PSO is used to update the If-part parameters of the proposed predictor, and the RLSE is used to adapt the Then-part parameters. With the hybrid PSO–RLSE learning method, the NFS–ARIMA predictor may converge in fast learning pace with admirable performance. Three examples are used to test the proposed approach for forecasting ability. The results by the proposed approach are compared to other approaches. The performance comparison shows that the proposed approach performs appreciably better than the compared approaches. Through the experimental results, the proposed approach has shown excellent prediction performance.     

CFNN-PSO: An Iterative Predictive Model for Generic Parametric Design of Machining Processes

ABSTRACT

Every production process consists of a large number of dependent and independent variables, which substantially influence the quality of the machined parts. Due to the large impact of process variabilities, it is difficult to design optimal models for the machining processes. Mathematical or numerical models for production processes are resource driven, which are not cost effective approaches in terms of computation and economical production. In this paper, a new artificial neural network (ANN) based predictive model is introduced, which exploits particle swarm optimization (PSO) algorithm to minimize the root mean square errors (RMSE) for the network training. This approach can effectively obtain an optimized predictive model that can calculate precise output responses for the production processes. In order to verify the proposed approach, two case studies are considered from literature and shown to produce significant improvements. Furthermore, the proposed model is validated on abrasive water jet machining (AWJM) with industrial garnet abrasives and optimal machining conditions have been obtained with optimized responses, which are substantially improved while compared with gray relational analysis (GRA).