Design of Office Chair: A Quality Function Deployment Approach

This paper employs a quality function deployment (QFD) methodology to translate customer requirements into design characteristics to improve the design of an office chair. A factor analysis has been carried out on the responses obtained from a cross sectional survey directed at users through a set of questionnaires. It has been obtained that three factors with twenty two items are loaded with a threshold value above 0.7. Finally, quality function deployment is used to extract important design characteristics satisfying the customer requirements.


Introduction
Quality is a characteristic of a product which has its ability to satisfy the implied customer's needs or in other words it fulfils customer's expectation from a product.quality function deployment (QFD) provides a means of translating customer requirements into appropriate technical characteristics for each stage of product development and production (i.e.marketing strategies, planning, product design and engineering, prototype evaluation, production process development, production, and sales).A fuzzy logic based quality function deployment approach has been used to identify the e-learning service provider for effective distance education [1].To extract bottleneck techniques, some researchers have used quality function deployment to design an assistive device [2].A QFD technique has been used for management of engineering institutions to provide guidelines to prioritize the improvement policies for their organisation [3].Some researchers have proposed a fuzzy quality function deployment for determining optimum level of engineering characteristics to randomized customer attributes [4].Here, an analytical hierarchy process (AHP) has been combined with quality function deployment for the selection of software project [5].Some researchers present a systematic approach to quality function deployment by addressing the customer's voice using symmetrical triangular fuzzy member [6].A study has been done by carrying two approaches (crisp and fuzzy approach) of QFD technique to develop a new shampoo [7].Some researchers focused on QFD to the product development process of contact manufacturing [8].Some researchers have developed an integrated framework based on Fuzzy QFD and Fuzzy optimization model to determine the product technical requirements [9].

II. DATA COLLECTION
Data are collected from employees of various offices like technical institutes, banks (private and government), hospitals (private and government) and some other organization.A questionnaire is prepared including forty variables for customer requirements, dimensions for design attributes (twenty four continuous design elements and thirteen categorical design elements) regarding office chair.One hundred twenty five data are collected from respondents through cross-sectional survey on forty items.The respondents need to answer in likert-type scale (1 for strongly disagree and 5 for strongly agree).A factor analysis has been carried out to examine the validity and reliability of variables to obtain a statically proven identification of customer requirements.The validity was tested by using principal component method following varimax rotation to extract the important customer requirements for model analysis which removes the redundancy and duplication from a set of correlated variables.The most important factors are determined on the basis of absolute sample size and magnitude of factor loading [10].To decide the sample size, Kaiser-Mayer-Olkin (KMO) test is carried out.If the value of KMO is greater than 0.5 then the sample size is treated as adequate.When data are adequate, a correlation matrix is created to calculate the correlation between variables [10].The internal consistency of the data was tested by Cronbach's Alpha (α) value.A total variance was determined for principal component varimax rotated factor loading procedure to avoid the correlation between various factors.

III. QFD ANALYSIS
The customer requirements shown under three different factors are found to be in indistinct form.It needs to be converted into design characteristics through a suitable method like Quality Function Deployment (QFD).Quality function deployment (QFD) is a powerful tool for converting indistinct customer voice (customer requirements) into engineering characteristic (continuous design characteristic) [11,12].The customer ratings for each customer requirement were obtained from left correlation matrix using equation ( 1) where, ij B denotes the relationship between customer needs, i Z is the initial customer rating, The individual rating of each design characteristic is obtained from the central matrix by using eqn.(2) where and denote the relative importance of the ith design characteristics with respect to the j thcustomer need in the relationship matrix and the importance of jth customer needs (customer requirements) and n is the number of customer requirements.

IV. RESULT AND DISCUSSION
A total of one hundred twenty five data including 40 customer requirements were collected from survey analysis.The survey data were subjected to factor analysis which was carried out by SPSS 14.0.Out of 40 customer requirements, 22 were loaded under 3 factors.Total variance explained by three factors was found to be 78.5% which is acceptable value for principal component with varimax rotated factor loading procedure.Ten items were loaded under factor 1, five items under factor 2 and seven items under factor 3. Factors extracted from factor analysis are named as comfortness (Factor 1), balance (Factor 2) and luxuriousness (Factor 3).Cronbach alpha (α) has been used to assess the internal consistency of the scale.The value of alpha for all dimensions is 0.702, which is just the acceptable value for demonstrating internal consistency of the established scale.The values of α obtained are 0.878, 0.933, and 0.939 for factors 1, 2, and 3 respectively.From Kaiser-Meyer-Olkin test (KMO>0.6), it can be concluded that the matrix did not suffer from multi collinearity or singularity.With a set of continuous design characteristics as given in Table 1, (categorical design characteristics are not considered here) three different QFD models named as QFD model 1 (comfortness), QFD model 2 (balance) and QFD model 3 (luxuriousness) were considered for correlating customer requirement with design characteristic.Figure 1, 2, 3 are represented by QFD model 1 (comfortness), 2 (balance) and 3 (luxuriousness) as given below.Ten, five and seven items (customer requirements) are considered under QFD model 1, 2 and 3 respectively.Similarly for design characteristics nine, seven and nine items are used respectively for three models 1, 2, and 3 as shown in Table 2.The design attributes extracted from the experts for three models are shown in Table 3.Initial rating of customer requirements for each model is derived by using a 1-10 scale for three models as shown in Figure 1, 2 and 3 respectively.The correlation of customer requirements (left matrix), design requirements (top matrix) and customer requirements with design characteristic (central matrix) are extracted from the experts using scale of 0.8, 0.6, 0.4 and 0.2 which indicates "strong", "moderate", "weak", and "very weak" respectively.Finally, initial design requirements and correlation values from top matrix are used in equation 1 to obtain final design ratings.The normalized refined rating of design attributes are obtained by dividing each rating with maximum value.From the normalized refined rating for design attributes, "Tilt of backrest" is the most prioritized element followed by "Number of controls" in case of model 1 (Table 3).Finally, four design attributes such as tilt of back rest, number of controls, overall width and overall height are considered out of nine design attributes having normalized refined rating value of 0.85 (threshold).Similarly, other two models have been developed with prioritized design characteristics as shown in same Table 3.For QFD model 2 (balance), four design attributes such as widthheight ratio of backrest, width-height ratio of seat pan, widthheight ratio of whole body, and width-height ratio of armrest exhibiting normalized refined rating value of 0.80 (threshold) and above have been considered.Similarly, five design attributes such as seat adjustment range, use of pattern, use of cushion, use of decoration, and backrest height showing normalized refined rating value of 0.90 (threshold) and above have been considered for QFD model 3 (luxuriousness).

V. CONCLUSION
The major contribution of this paper is to provide an integrated approach for modelling design characteristic of a product (office chair) in an office environment as the interaction between the product and customer varies from customer to customer.An office chair must satisfy the customer's requirements for a wide range of population by selecting a less but important set of design characteristic.So this work provides a QFD approach to improve the quality of the product by identifying important design characteristic.

ijA
denote the relative importance of th i characteristic with respect to th j customer's needs in the relationship matrix, j X represents the importance of th j customer needs and n is the number of customer needs.