Author Email: firstname.lastname@example.org
Recently, solicitation of Quality Function Deployment (QFD) has been extensively emerged in the product development market. Customer Requirements (CRs) and Design Requirements (DRs) are the two chief entity of the QFD process. However, variegation of marketplace directly influence CRs and DRs in making them more varying; that prompts the need of forecasting in QFD for gaining enhanced customer satisfaction. It can be attained by using Data Mining techniques of forecasting. As various forecasting techniques are available, selection of most efficient method is highly desired for evaluating out the most effective result. This paper offers an innovative software tool that precisely forecasts future values in QFD. The tool enables forecasting using several time series analysis techniques of data mining that sturdily lend a hand in carrying comparative analysis and assessing out the efficient technique. The tool is designed using VB.Net and MS-Access. Lastly, an example is shown to exhibit the workability of proposed forecasting tool.
QFD, Data Mining, Time Series Analysis, Forecasting, Tool
In todays’ innovative world owing to the rapid enhancement of technology, attaining customer satisfaction has turned to be the most challenging job for organizations. Worldwide competitiveness in the marketplace has recently become the major apprehension of many organizations, which deliberate ‘‘continuous improvement’’ to deal with the readily varying development throughout the world encouraged by technological innovations and fluctuating customers’ requirements (CRs) (Kahraman, Ertay & Buyukozkan, 2004). To deal with this issue, currently corporations primarily emphasizes on the Mantra – ‘communication with customers’, the reason behind this is customers plays key role in rising evolving market place; also realizing requirements specified by them may entail to the prosperous product. The Quality Function Deployment (QFD) has emerged as one of the most promising planning tools to deal with customer’s requirements. The QFD transforms the customer requirements (CRs’) into design requirements (DRs) for a product, prioritizing these features while simultaneously forming development objectives for it (Karsak, 2004).
Customer requirements (CRs) are one of the crucial entity that plays key role in several phases of QFD. Whereas, DRs are answers to technical implementation of these CRs in an interim semantic of designers and engineers. In consequence of prompt divergence in marketplace, it is pretty clear that CRs would be additionally fluctuating. As product designing solely depends on previous requirements, CRs may be variable while designing and be different when it is already in the market. These dynamic CRs also results in their corresponding DRs. As a result, dealing these time lag problem and dynamic requirements have become thought-provoking task for designer and engineers. Thus, there is pressing need of forecasting in QFD. Forecasting future trends in QFD can be fruitful for organizations to deliver improved products, upgrade their effectiveness in market and elevate customer satisfaction. (Hsu, Wang & Lin, 2007). In the contexture of a customer-centric products, a timely update of CRs may assist as an essential criterion to observe how things differ over time, moreover it comprises the organizations a higher ground to interoperate various tactics to encounter the future CRs (Bostaki & Roghanian, 2013). Forecasting upcoming values in QFD would prove to be profitable in framing the forthcoming tactics and can be helpful in taking the anticipatory steps for the forthcoming situations. Many experts have employed various forecasting techniques in QFD for forecasting the future values.
Due to the rapid enhancement of marketplace, forecasting the future values has become pressing need of organization to deal the dynamic requirements. To serve the purpose, a unique forecasting software tool is introduced that efficiently forecasts the CRs and DRs in QFD by applying time series techniques of forecasting. This integration assists in generating the outcomes in lesser time and improves the customer satisfaction. The tool functions in three stages: first it deals with the QFD process to generate Raw weight and Priority scores, thereafter it employs the various time-series techniques such as, SMA, WMA, SE, DEB, DEH&W, and LR on QFD data to forecast the future weights, and finally it evaluates these forecasting techniques comparatively as per their forecast accuracy. MSE is used as an evaluation criteria for identifying the apt method of forecasting. In order to test the practicability of software tool, a practical example of ERP system has been taken with data sets of 5 periods to produce forecast of succeeding period. Outcomes of study illustrates that the MSE for LR is minimum followed by SE, DEB, WMA, SMA and DEHW taking all CRs and DRs into consideration. Hence, it has been realized that the Linear Regression (LR) overtakes the other forecasting techniques and validated to be proficient technique for identifying the futuristic weights of both CRs and DRs in QFD. The result of this study can be beneficial for the designers and engineers in coping up with the futuristic CRs and DRs as well as improving the competitiveness in the global marketplace with enhanced customer satisfaction.
177 total views, 2 views today