for GIS. Incidentally, in the time used for integration of 
data (the error checking and the correction etc. is 
contained) is half day per 1 areas by one operator. 
2. 4. Grasping of the blockade situation of the street 
Grasped an items shown below . 
) Street width before the earthquake disaster 
) Decrease of street width by pole falling 
) 
) 
Decrease of street width by debris 
Decrease of street width by debris and pole falling 
Dour 
Each item was classified by following standard. 
i) equal to or less than 1.5-m width (Pedestrian passage 
possible) 
ii) equal to or less than 3.0-m width (Bicycle passage 
possible) 
iii) equal to or less than 6.5-m width (1 vehicle passage 
possible) 
iv) more than 6.5-m width (2 vehicle passage possible) 
We paid notice to the possibility of vehicle's passage to 
identify the situation of blockade. And then, the authors 
grasped identify the situation of blockade. And then, the 
authors grasped what passage condition the street equal 
to or more than 3-m width by the influence of the blockade. 
Also, the passage possibility of the vehicle was found out 
from the remaining width of the street as a result of the 
blockade. The standard is based on i)iv) above- 
mentioned. 
2. 5. Implementation and evaluation of the analysis 
The passage situation of the street was simulated based 
on the structure of the street and the measurement data of 
the blockade and the pathfinding was implemented using 
network data on GIS. A pathfinding was implemented by 
following technique. 
1) Choose one node which belongs to main street on 
analysis area borders(about 6 nodes in 1 area). 
2) Implement pathfinding from chosen node to each 
node. 
3) Total frequency of pathfinding for every link. 
We can estimate the degree of the flow from the main 
street to the analysis area from above. The value of the 
load of each link was estimated from this result. 
It causes the result to have disregarded the structure of 
the street if the pathfinding is performed only based on the 
distance. Therefore, the authors set the impedance of 
each link simply as follows from width. 
Impedance = (Length of link) / (Width of link) 
The result of the pathfinding is presented as the frequency 
that the link was chosen. Also, it is inevitable that the 
invisible area occurs as long as an air-photograph is used. 
We excluded the invisible area from this analysis. 
380 
3. Results 
3. 1. Grasping of the blockade situation of the street 
The authors totaled the amount of each link according 
to the following item using GIS. 
3. 1. 1. Street width before the earthquake disaster: 
The number of the links by the width in the investigation 
area is shown in the Figure 2. The characteristic of the 
investigation area is as follows in general. 
A) There are comparatively many streets with wide 
width in Sannomiya and Nagata area where 
commerce and industry are developing. 
B) There are many street where the vehicle can not 
pass in the Hyogo area. 
C) As for the other area, the rate of the distinction at the 
street by the width is approximately the same. 
3. 1. 2. Decrease of the street width by the influence 
of falling poles: The situation of streetway blockade due 
to fallen poles is shown in Figure 3. for the streets with the 
width of 3 to less than 6.5m and in Figure 4. for the streets 
with the width of 6.5m or wider. There are few absolute 
numbers of the blockade part by the falling of poles. 
However, the possibility of the passage of the vehicle in 
the blockade occurrences is low. 
3. 1. 3. Decrease of the street width by the influence 
of debris: The situation of streetway blockade due to 
debris is shown in Figure 5. for the streets with the width of 
3 to less than 6.5m and in Figure 6. for the streets with the 
width of 6.5m or wider. Occurrence of debris is 
overwhelmingly higher comparing with the falling of poles. 
However, the hindrance to the vehicle passage due to the 
street blockade resulting from debris is less significant 
comparing with those resulting from pole falling. But the 
passage of vehicle is impossible at 60% or more of the 
streets on average in case where the street width is 3 - 
6.5m while it is around 30% on average where the width is 
6.5 m or more. Therefore, this result shows that the 
securing of width can be the basic factor of the securing of 
vehicle passage. 
3. 1. 4. Decrease of the street width by the influence 
of falling of poles and debris: The situation of streetway 
blockade due to fallen poles and debris is shown in Figure 
7. Here we roughly classified the streets into those where 
vehicles could pass and could not pass after the 
earthquake. And the streets where the vehicles could not 
pass were further divided into those with which non- 
passage was caused by the structure itself of street before 
the occurrence of earthquake and those with which 
hindrance to the passage was caused by the street 
blockade resulting from the earthquake. As a result, it was 
found that the damage of the street blockade was big in 
Rokkoumichi and Uozaki areas, and that the damage was 
small in Kasuganomichi and Hyougo area. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996 
  
Bela d ee uif 
Nunmber of LINK 
Figui 
Number of LINK 
-—— -— 
o a 
e e 
Figur 
NAG:! 
michi, 
3. 2. 
Here, 
was | 
the d. 
3.2.1. 
disas 
and tl 
patter 
the v 
width 
street