Extended abstract
1- Introduction
Twentieth century began with a billion people with 10 percent urban dwellers. It has ended with 6 billion people and 50 percent urban dwellers. Therefore, urban planning is more important than ever before. As a result, attention to the optimal distribution of applications and service centers is needed and thus most programmers are working with this issue. Accordingly, one of the most preliminary needs of disaster victims in the initial hours proceeding man-made disasters is health utilities. Supplying and equipping health care systems and their proper distribution among urban regions is one of biggest challenges for crisis managers. Extracted maps by network analyst model shows that hospital centers distribution in Tehran 3rd district do not comply with the principle of equitable distribution and adequate access for all citizens. A Lack of equitable distribution and lack of proper servicing in case of man-made crises in the region under study is one of the main reasons of this study. In this paper the allocation of hospital centers in Tehran 3rd district is done by colonial competitive algorithm.
2- Methodology
The research method in this paper is based on a combination of descriptive and analytical procedures. The case study area included Tehran 3rd District in northern Tehran, with six regional and 11 neighborhoods and a population of 290,726 people in 2006, with an area of 29.38 hectares. Required information for this research has been collected from housing and population census data, 1:2000 maps and comprehensive and master plans of Tehran in addition to magazines and books related to the research topic. The results were analyzed using Arc/GIS, Super Decisions and Matlab software with network analysis was used for the evaluation of hospital service span and ANP model and ICA algorithm were used for hospital centers allocation.

3-Discussion
The internal flexibility of urban systems, the potential utilization of internal rules and regulations conforming to the expectations of external passive defense system are important factors that help planners and urban designers while creating a healthy and safe urban demand. According to analysis, it is revealed that the pattern of hospitals centers distribution is inappropriate in the study area and has many deficiencies. Accordingly, allocating the correct and optimal hospitals centers were based on the principles of passive defense. The allocation of hospital in this research was based on three principals: safety, consistency and efficiency. Moreover, population factors, population density, passage confinement, pathway access, suitability, and neighborhood service area were defined as sub criteria in this case study. The results are been shown in fig 1.


Figure 1: The final layer is valued for hospitals

After converting the layers to measurable criteria, the final Weights were calculated by Network Analysis Model (ANP). Then the relative weight of each factor relative to the weight of each item obtained were incorporated in the Weighted Overlay function. The resulting number indicates the region which has the highest priority is to establish hospitals.Finally, after optimizing the algorithm based on colonial competitiveness, hospitals center has been allocated in the case study area. Accordingly, each of the proposed sites is listed for the applications of weighted priorities in Table 3, and the final results are shown in Figure 2.
Table 3: Preference weights for the Tehran 3 district hospitals
Centers Numbers 1 2 3 4 5 6
weights 6.3 5.79 5.54 5.83 5.36 5.79


Map 2: Optimal sites for medical centers in Tehran 3 District
4- Conclusion and Suggestions
In this study, medical centers have been allocated with an emphasis on the principles of passive defense. This new approach to site selection for urban applications uses a heuristic algorithm colonial competition. Furthermore, network analysis model were used to examine the distribution of and access to hospital facilities. The results of network analysis showed that these centers cannot provide appropriate services in man-made crises in the region and do not follow the same distribution in the region. In order to optimize site selection and basic principles of passive defense, network analysis models, ANP model and algorithm of ICA were used for selecting of the best sites for medical centers. The results of this study showed that colonial competitive algorithm can be used as a suitable model for urban land use site selection and help urban planners to understand and prioritize and find best solutions for urban issues.