Taking advantage of the opportunity to carry out an in-depth project of analysis and municipal intervention in the public space of the city of Porto, it was possible to test a set of formal and digital methodologies, comparing these with more traditional ones. As the project included the collection of empirical data, as well as the use of theoretical models, it served to confront and test the theoretical models and determine ways to improve them. The paper will summarize the theoretical and methodological advances obtained. First, it will define the lines for an ontology of public space, closely relating its organic-formal aspects (geometric definition, materials, infrastructure, urban furniture, afforestation) with its functional aspects (accessibility, connectivity, visibility, intelligibility, security, sociability). Secondly, it will describe the improvement in various digital techniques to obtain empirical data. The focus will be on the possibility of monitoring and recording the individual paths of the various objects and people within space, using fully automated computer analysis of video; the improvement of results with heuristic and non-heuristic machine learning processes; the processing of information from multi-camera systems; the geo-referencing of routes; statistical and artificial intelligence treatment, with the objective of standardizing types of routes (passage, access to the periphery of space, usufruct, deambulatory / tourist) and studying the relations between the configurations of routes and spaces. Finally, it will show how the behavior of space and agents in space can be predicted through theoretical instruments such as SCAVA (space configuration, accessibility and visibility analysis) technologies, including space-syntax and agent-based methods. The advantage of theoretical solutions is that credible results can be obtained without the need to carry out empirical studies in each case study. But for this it is necessary that the theory is credible. The comparison of results from theoretical analysis with empirical data (that resulted from field studies for the same cases of public spaces in the project) allowed us to refine the theoretical methods used. Two were specially treated: Agent-based methodologies: They already include a set of configurable rules that can be adjusted. It was then a matter of defining the set of rules of individual behavior for each agent, in their relationship with space and also in their relationship with other agents, and calibrate the parameters of each of these rules. Space Syntax: The paper will describe the solutions that had to be found: (a) to delimit small areas of public space (square, street, garden) for VGA (Visibility Graph Analysis) within the global configuration of the city, without ignoring the relationships that are established between this area and the whole city; (b) to include in the analysis the characteristics of the agents in transit, in addition to the geometry in which they move, as performed in Traffic Flow Theory; and (c) to introduce the functional aspects of the use of space combining configurational attributes with urban functions such as leisure, retail, services and others, as has been done consistently by Place Syntax. All these aspects come together in a final multilayer analysis.

--

Public space,space syntax,traffic flow theory,agent based analysis,scava