Strategies for redevelopment of gray belt objects on the basis of neural networks
The article considers the approaches for objects redevelopment in the gray belt. Information was collected about 45 objects located in different administrative districts of the city. As the criteria for clustering objects, general factors (year of construction of the building, сost of building restoration in prices of 1969, actual сost of building in prices of 1969, height, volume, number of stores, total building area, fundamental group, function) and factors on physical deterioration (wear of roof, floors, walls, foundation, finishing, MEP, total wear) were chosen. As a result of the study, SOMs with different learning parameters were created. As a result of the research, it was established how to change and select the desired redevelopment strategy for the zones of the gray belt, depending on the leaning parameters of the SOM and the individual characteristics of objects entering the gray belt.