Simulation of urban development in the City of Rome: Framework, methodology, and problem solving

Abstract

In Italy’s case, the implementation of theUrbanSIMmodel involved the territory of Rome, including the municipalities of Rome and Fiumicino. e main goal was to build scenarios regarding the future of economic deconcentration. Rome is the largestmunicipality in Europe,with an inhabited surface area only slightly smaller than that of Greater London and almost double that of the inner Paris suburbs (thePetiteCouronne). espatialdistributionof buildingswithinthemunicipality isdistinctive. Unbuilt areas comprise 73 percent of the territory. ese voids are oen farmland (paradoxically,Rome is the largest rural municipality in Italy) or areas with high environmental, historic or cultural value. Fiumicino, previously part of themunicipality of Rome, became an independentmunicipality in 1991. Its autonomy, made all the more signicant because Fiumicino hosts the international airport, marked the start of an extensive process of economic deconcentration along the route connecting Rome to the airport. InItaly’s case, the implementationof theUrbanSIMmodel posed several challenges, notably the availability, homogeneity and completeness of data. is paper uses four specic cases (land use, travel times, accessibility, and residential land values) to propose a general methodology to solve problems related to missing or non-homogeneous data. For the land use, we simply combine two different land use data sources,while for accessibility and travel timedata,we propose the use of geostatistical methods in order to estimate missing and unavailable data, calculating also the accuracy of the predictions. For the residential land values, which are discrete data, we suggest the use of deterministic interpolation techniques. While it has not yet been possible to implement the calibration stage, some simulation outputs are presented. 