<p>　In recent years, large cities have a large number of aging public housing units and the number of vacant public housing units is increasing. In this paper, a rational reorganization method of public housing is discussed based on the data of public housing units (generally owned by local governments). The information was obtained with the cooperation of the Kobe City Housing Bureau, Hyogo Prefecture Housing Management Division, and Kyoto City Housing Policy Division.</p><p>　The first thing we considered was how to statistically derive the priority of reorganization of residential buildings. This method evaluates each residential building on two axes: overall problem level and ease of relocation. The effectiveness of the method was confirmed by applying it to all the public housing in Kobe City. Furthermore, based on the comparison with Kyoto City, the items used in this statistical processing may be useful in large cities where public housing is constructed at high density. On the other hand, in municipalities with a low density of public housing, it was considered necessary to select a separate item for statistical processing.</p><p>　Even after identifying a priority candidate public housing project, it is very difficult to derive an appropriate restructuring plan from the vast number of combinations of housing units in the project. For example, there are 29 buildings in Oshibedani housing, but it is virtually impossible for the government officials to consider all of them because there are more than 500 million combinations of buildings.</p><p>　In this study, a program was developed, using the open-source programming language Python3, to automatically rank all combinations of residential buildings that are candidates for reorganization in terms of cost of reconstruction, number of buildings that can be reduced, and number of units.</p><p>　When the program was applied to the area under consideration for reorganization by Kobe City, it led to a combination of buildings that could reorganize more units at a lower cost than the possible combinations of reorganization proposals.</p><p>　The last thing we considered was how much easier it would be to draw up a reorganization plan by making both municipal and prefectural housing in the same area candidates for reorganization at the same time.</p><p>　The study was applied to the Outuka area in Kobe City. The results show that in the Outuka area, the variation of possible combinations for restructuring increased by 50% at the present time, and the variation increases significantly as the year progresses. Thus, in order to be able to develop a more rational reorganization plan, it was found to be effective for public housing managers to cooperate with each other to develop a coordinated reorganization plan for municipal and prefectural housing.</p>