An extended QFD planning model is presented for selecting design requirements (DRs) that consider longitudinal effect. In the proposed model, the longitudinal effect is incorporated by introducing a time dimension into the existing house of quality structure. As a consequence of explicitly considering the longitudinal effect, the proposed model yields not only an optimal set of DRs but also the timing of their selection. The proposed model is demonstrated through a case study for improving customer loyalty in the high-speed internet service.
Quality function deployment (QFD) is a mechanism for translating the “voice of the customer” (called customer requirements; CRs) into the language of the engineers (called design requirements; DRs). DRs are subsequently translated into parts characteristics, process plans, and detailed production requirements. One of the acknowledged advantages of QFD is its ability to promote organizational consensus building and decision making (Cohen, 1995).
QFD has been widely used across various industries to improve customer value (CV) such as customer satisfaction or customer loyalty. Such an effort is called QFD planning. QFD planning is defined as a proactive “customer-driven planning” based on two major pieces of QFD information, namely, the DRs’ impact on CRs and the CRs’ effect on CV (Day, 1993). Through QFD planning, the decision-maker (DM) identifies the DRs that are essential to achieving the desired CV level.
The problems in QFD planning may be boiled down to two types: project selection type problems and resource allocation type problems. The project selection type problem is encountered in finding an optimal set of DRs to be undertaken among all DRs to improve the CV in the most efficient way (Ahn and Kim, 1999 and Park and Kim, 1998). In this type, DRs can be regarded as projects for improving CRs, and thus are represented as 0–1 decision variables. The resource allocation type problem is encountered in determining the level of resources to be invested in DRs. As such, the DRs are represented as continuously controllable decision variables (Chen et al., 2005, Fung et al., 2002, Fung et al., 2003 and Tang et al., 2002). Except for the different types of decision variables, both problems are fundamentally equivalent in the sense that they attempt to determine the optimal level of DRs to maximize the CV.
The existing QFD planning models have a common assumption – the CV at a certain time point is affected by the levels of CRs at the same time point only. In other words, the CRs’ effect on CV is cross-sectional. However, this is not always the case in reality. The CV at a certain time point can be affected not only by the levels of CRs at the same time point but also by the levels of CRs at the previous time points collectively. This means that the CRs’ effect on CV is longitudinal. For example, customer loyalty, one of the popular CVs, is known to be formed by longitudinal effect. Customer loyalty is defined as the customer’s attitude to a product/service formed over a period of time (Ramaswamy, 1996 and Stank et al., 1999). Namely, it is determined by a series of effects over a certain period of time (viz. longitudinal effect), rather than by the effect at a specific point of time.
The high-speed internet service is an internet-access service based on the ADSL (asymmetric digital subscriber lines) or VDSL (very high-speed digital subscriber lines) technology. The notion of customer loyalty in the high-speed internet service is very important because this market is highly competitive and nearly saturated. Kim et al. (2007) demonstrate that the customer loyalty in this service is, in fact, formed by the longitudinal effect. More specifically, the customer loyalty is shown to be affected by the levels of CRs for the recent two-month period.
This paper proposes an extended QFD planning model which takes the longitudinal effect into consideration. In the proposed model, a primary focus is placed on incorporating the longitudinal effect of CRs on CV. This is accomplished by introducing a time dimension into the CRs. As a consequence of explicitly considering the longitudinal effect, the proposed model yields not only an optimal set of DRs but also the timing of their selection.
A brief review of the house of quality chart and the previous developments in QFD planning is given in Section 2. An extended QFD planning model with longitudinal effect is proposed in Section 3. In Section 4, the proposed model is demonstrated through a case study in the high-speed internet service. Finally, conclusions are given in Section 5.
An extended QFD planning model which takes the longitudinal effect into consideration has been proposed. The longitudinal effect is incorporated by introducing a time dimension into the existing HOQ structure. The proposed model provides not only an optimal set of DRs but also the timing for their selection. By considering the longitudinal effect, the proposed model allows the DM more decision choices than the conventional QFD planning models, and thus leads to a better performance.
The proposed model has also been demonstrated through a case in high-speed internet service to improve customer loyalty. The proposed model may be applied to various project selection type problems not just in the internet service industry, but also in any other industry where the longitudinal effect is particularly important. The relative efficacy of the proposed model is expected to increase as the significance of the longitudinal effect becomes higher.
There are several issues that should be investigated further. The proposed model in this paper implicitly assumes that the DRs are uncorrelated. There may be specific interactions among DRs such as technical correlations (Fung et al., 2002) or financial correlations (Park & Kim, 1998). A future research is called for on the effect of the technical and financial correlations among DRs in conjunction with the longitudinal effect.
Secondly, the proposed model aims to achieve the target CV at the final point of time. However, in reality, the interim targets as well as the final target are often of concern, especially in the long-term planning. The proposed model may be extended to accommodate a series of interim targets at various points of time.
