*5.3. Groupings and Suggestions*

We applied Ward's method to perform cluster analysis using the standardized scores for six criteria associated with villa development and the net ratio of vacant lots, the ratio of villa lots having areas ranging between 300 and 400 m2, the ratio of villa lots having areas ranging between 1000 and 1400 m2, the average land price, the total area of vacant lots, the distance between a mosque and a park, and the total road length (Figures 14–16). The 103 target areas were categorized into six groups (Figure 17). The characteristics and recommendations for reducing the net ratio of vacant lots are presented below.

**Figure 14.** Scatter diagram (1).

### 1. Group A (26 sites)

The neighborhoods included in this group evidence the following characteristics: (1) the areas of a few lots range between 300 and 400 m2; (2) the total area of vacant lots is relatively large; (3) land prices are high; and (4) the road lengths are short compared with those of other groups. Because of the relatively large total area of vacant lots, it is necessary to delineate existing lots with areas ranging between 300 and 400 m2 along with increasing road length for improving the street environment and reducing the ratio of vacant lots. In addition, increasing the number of parks to respond to the increase in population density is recommended, thereby increasing the possibility of consumers being able to purchase lots.

**Figure 15.** Scatter diagram (2).

**Figure 16.** Scatter diagram (3).

#### 2. Group B (38 sites)

The neighborhoods included in this group share the following characteristics: (1) a certain number of villas have been built in lots with areas ranging between 300 and 400 m2; (2) a few vacant areas are present; (3) the total street length is substantial; and (4) there is a short distance between the mosque and park. As the majority of lots within this group have already been developed, there are no further recommendations or changes needed.
