This thesis develops a replicable workflow for measuring street-level indicators of physical urban decay and analysing their relationship with residential real estate prices. Physical urban decay is understood as the visible deterioration, disorder, or weak maintenance of the built environment, including elements such as graffiti, façade decay, litter, damaged street furniture, and other observable forms of physical neglect. The methodology combines street-level imagery, computer vision and hedonic price modelling. The empirical application focuses on graffiti in Milan, used as a specific and detectable case study to test the proposed workflow. Google Street View images are processed through a YOLO object detection model, and the resulting detections are aggregated into property-level exposure indicators. These indicators are then integrated with residential listing data in spatial hedonic models. The results show a stable negative association between graffiti exposure and asking prices per square metre: in the preferred specifications, a one-standard-deviation increase in exposure is associated with an approximate reduction of around 1% in price per square metre. The main contribution of the thesis is the formalisation of a transparent, scalable, and replicable method for linking street-level visual indicators of urban decay with real estate analysis.
Questa tesi sviluppa un workflow replicabile per misurare indicatori street-level di degrado fisico urbano e analizzarne la relazione con i prezzi immobiliari residenziali. Il degrado fisico urbano è inteso come il deterioramento visibile, il disordine o la debole manutenzione dell’ambiente costruito, includendo elementi quali graffiti, degrado delle facciate, rifiuti, arredo urbano danneggiato e altre forme osservabili di incuria fisica. La metodologia combina immagini street-level, computer vision e modelli edonici dei prezzi. L’applicazione empirica si concentra sui graffiti a Milano, utilizzati come caso studio specifico e rilevabile per testare il workflow proposto. Le immagini Google Street View vengono analizzate attraverso un modello di object detection basato su YOLO e le rilevazioni ottenute vengono aggregate in indicatori di esposizione associati agli immobili. Questi indicatori sono poi integrati con dati provenienti da annunci immobiliari residenziali all’interno di modelli edonici spaziali. I risultati mostrano un’associazione negativa e stabile tra esposizione ai graffiti e prezzi richiesti al metro quadrato: nelle specificazioni principali, un aumento di una deviazione standard dell’esposizione è associato a una riduzione di circa l’1% del prezzo al metro quadrato. Il principale contributo della tesi consiste nella formalizzazione di un metodo trasparente, scalabile e replicabile per collegare indicatori visivi street-level di degrado urbano e analisi immobiliare.
Measuring physical disorder and its relationship with urban Real Estate prices
FIORINA, RAFFAELE
2025/2026
Abstract
This thesis develops a replicable workflow for measuring street-level indicators of physical urban decay and analysing their relationship with residential real estate prices. Physical urban decay is understood as the visible deterioration, disorder, or weak maintenance of the built environment, including elements such as graffiti, façade decay, litter, damaged street furniture, and other observable forms of physical neglect. The methodology combines street-level imagery, computer vision and hedonic price modelling. The empirical application focuses on graffiti in Milan, used as a specific and detectable case study to test the proposed workflow. Google Street View images are processed through a YOLO object detection model, and the resulting detections are aggregated into property-level exposure indicators. These indicators are then integrated with residential listing data in spatial hedonic models. The results show a stable negative association between graffiti exposure and asking prices per square metre: in the preferred specifications, a one-standard-deviation increase in exposure is associated with an approximate reduction of around 1% in price per square metre. The main contribution of the thesis is the formalisation of a transparent, scalable, and replicable method for linking street-level visual indicators of urban decay with real estate analysis.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_07_Fiorina.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
7.99 MB
Formato
Adobe PDF
|
7.99 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/260915