4.2.3. Experimental Procedure

The procedure of the experiment is described as follows:


**Figure 8.** Experimental environment.

**Figure 9.** Simulation videos.

**Figure 10.** Simulation videos.

**Figure 11.** Simulation videos.

#### *4.3. Results and Analysis of the Usability Experiment*

A total of 30 participants were invited for the experiment in this study and their background information is shown in Table 5 as follows.


**Table 5.** Background information of the participants.

#### 4.3.1. Outline of the Usability Decision Model

The outline of the usability decision model that was used in this study includes several steps as follows:


#### 4.3.2. Description of Problems

A stereo system was selected as the case study. In addition to meeting the functional, physical and styling limitations, the product is required to satisfy other limitations in the specific tasks. The first step of the evaluation is to stipulate the design guidelines. The limitations could be due to the engineering, ergonomic, aesthetic, marketing, logistic, planning, design, product and economic factors.

#### 4.3.3. Constructing the objective tree

The method of objective tree [16,17] provides a clear format of the hierarchical relationship between higher levels and lower level during the decision-making process of design schemes. An objective tree that was constructed based on the above-mentioned design criteria is shown in Figure 12.

**Figure 12.** The objective tree of a stereo system.

4.3.4. Evaluation of weight function

In order to determine the influence of the lower levels on the higher levels, the weight functions should be determined by methods such as AHP [20], tabulated judgment method [21] and so forth. In this study, the tabulated judgment method was used for determining the weight functions.

The procedure is described as follows:


$$\sum\_{i=1}^{n} \, ^n\_{i=1} \mathbf{K}\_i = \left\{ \left( n^2 - n \right) / 2 \right\} \times 4 = 2 \left( n^2 - n \right) \tag{2}$$

(6) The weight function of each item can be expressed by Wi = Ki/*ni*=<sup>1</sup> *Ki*, which is the weight of the *n*th item.

By implementing the above-mentioned method, the degree of influence of the five items and their child items can be determined as shown in Figure 13 and Table 6.


**Figure 13.** Hierarchy of five items of a stereo system.



All of the weights can be calculated by the above-mentioned method and the results are summarized in Table 7.



#### 4.3.5. Measuring the Membership Functions

In order to determine the participants' subjective degree of satisfaction on these four design ideas, the membership values of all items need to be measured. The definition of the fuzzy evaluation set is as follows:

*V* = 0 Strongly disagree , 0.25 Disagree , 0.5 Neither agree nor disagree , 0.75 Agree , 1 Strongly agree (3)

After that, the average membership values of each child item on the design ideas can be obtained from the values that were calculated from the questionnaire. The results are shown in Table 8.


**Table 8.** Average measured values from the 30 questionnaire copies.

After that, the influence of the efficiency factors (accuracy of music source button, promptness of message reception, clarity of message reception, time for taking/placing CD) at the lower level on the efficiency factors at the higher levels. The influence of the four items on the four prototype stereo systems can be determined by the same approach from the lower levels up to the higher levels.

1. Efficiency factors:

$$\begin{aligned} \text{[E]} &= \begin{bmatrix} \text{w1 w2 w3 w4} \end{bmatrix} \ast \begin{bmatrix} \text{E} \end{bmatrix} = \begin{bmatrix} \text{w1 w2 w3 w4} \end{bmatrix} \ast \begin{bmatrix} \text{r1 1 r1 1 r1 } \text{r1 1 r1 }\\ \text{r1 2 r1 2 r1 2 r1 2} \\ \text{r1 3 r1 3 r1 3 r1 3} \\ \text{r1 4 r1 4 r1 4 r1 4} \end{bmatrix} \\ &= \begin{bmatrix} 0.375 & 0.25 & 0.25 & 0.25 & 0.125 \end{bmatrix} \ast \begin{bmatrix} 0.650 & 0.517 & 0.783 & 0.333 \\ 0.692 & 0.683 & 0.658 & 0.625 \\ 0.583 & 0.525 & 0.708 & 0.525 \\ 0.525 & 0.358 & 0.533 & 0.842 \end{bmatrix} \\ &= \begin{bmatrix} 0.628 & 0.628 & 0.702 & 0.518 \end{bmatrix} \end{aligned} \tag{4}$$

The membership values of efficiency factors are shown in Figure 14. It is known from the calculation of the membership functions that, Idea 3 (0.702) had the greatest influence from the aspect of the efficiency factor, followed by Idea 1, Idea 2 and Idea 4.

**Figure 14.** Membership values of efficiency factors.

The membership values of memorability factors, functional factors, emotional factors, and satisfaction factors are respectively shown in Figures 15–18 which indicated that in general Idea 2 had the grea<sup>t</sup> influence than others except from the aspect of the satisfaction factor.

2. Memorability factors:

**Figure 15.** Membership values of memorability factors.

```
3. Functional factors:
```
**Figure 16.** Membership values of functional factors.
