*4.2. Discussion*

Figure 4 shows the overall e fficiency scores that consider both the design and e fficiency stages. As mentioned above, station areas with a score between 0.000 and 0.200 have top priority for TOD improvement due to low design and e fficiency scores. Based on the mean score for each stage, i.e., transit design and e fficiency analysis, the DMUs were also divided into four groups to indicate the type of improvement needed at these station areas. In Figure 4b, we can see that both the design and efficiency scores of group 1 are lower than the mean value, which mean that station areas in group 1 need to improve both their design and e fficiency. The station areas in group 2 need improvements in design more than e fficiency. Station areas in group 3 can improve their overall e fficiency by improving their e fficiency, and stations areas in group 4 qualify as relatively e fficient stations. The types of improvement required to achieve overall e fficiency can be determined by the scores of each stage or by group analysis.

The analysis of an administrative unit is needed to practically evaluate the TOD priorities. Using the e fficiency evaluation results of the station areas, it is also possible to analyze transit e fficiency by the administrative units in Seoul. The dong unit is the smallest unit among the administrative units. The overall e fficiency scores of the dong unit were obtained by averaging the scores of the relevant station areas.

**Figure 4.** Efficiency evaluation results with respect to design and efficiency: (**a**) efficiency results prioritized regarding TOD; and (**b**) group-wise efficiency results.

The dongs were classified with respect to four dimensions, i.e., non-scoring, low-scoring, mid-scoring, and efficient, and land-use features of each dimension were identified by the socioeconomic variables of each dong. The overall efficiency scores of the dong unit are shown in Table 5 and Figure 5.


**Table 5.** Results of overall efficiency by Dong unit in Seoul.

**Figure 5.** Visualization of the efficiency evaluation results by dong unit in Seoul.

Figure 5 shows the results of the overall transit e fficiency of Seoul by the 424 dong administrative units. Since the station areas were set within a 500-m radius, a station area could cover several dongs. Even though the no subway stations reside in the dong, the dong can be a ffected by the station area radius that spans multiple dongs. The e fficiency score of each dong is estimated by calculating the average e fficiency values of the station areas that reside or radius spans into the dong.

Eight dongs were estimated to be e fficient, having earned a 1.000 overall e fficiency score, i.e., Wirye-Dong, Gonghang-Dong, Sangam-Dong, Susek-Dong, Jinkwan-Dong, Daerim (3)-Dong, Dorim-Dong, and Yeongdeungpobon-Dong. These eight dongs can serve as reference DMUs for achieving e fficient TOD. Regarding regional characteristics, e fficient dongs have both residential and commercial land-use features. The top priority areas for TOD are the red dongs in Figure 5. These 41 dongs were estimated to be non-scoring areas since they contain no station areas. These non-scoring dongs do not have any station areas that reside or spans within the dong despite range being 500 m radius for the station area. Residents of the non-scoring dongs must first take a bus to make transit trips due to the lack of subway stations. Since only bus infrastructures have been established, these areas are at a disadvantage for improving their overall e fficiency. With respect to land use characteristics, these top-priority dongs for TOD are mostly residential with a large number of households, i.e., 9478 households. The second-priority areas are the 79 low-scoring dongs, which earned overall e fficiency scores in the 0.0 to 0.2 range. Although these 79 dongs were designed for TOD, their transit infrastructures, trips, and energy consumption must be improved to achieve e fficient TOD. Regarding land use, these second-priority dongs also have commercial features and host a large number of companies, i.e., 2083 companies.
