This thesis proposes the adaptive reuse of a former industrial storage complex, the Magazzini Generali Raccordati con Silos da Cereali in Novara, Italy, as a mixed-use facility combining a data center and student housing. It responds to the environmental costs of the current AI-driven data center boom — land consumption, embodied carbon, and intensive water and energy use for cooling — by proposing adaptive reuse, rather than greenfield construction, as an alternative growth model. The research examines the structural, spatial, and infrastructural compatibility of industrial storage typologies, specifically silo clusters and long-span warehouses, with the technical requirements of data centers. It argues that the silo cluster’s vertical, cellular, load-bearing structure is inherently suited to server-rack and cooling infrastructure, while the complex’s warehouse and residential buildings accommodate human-occupied program, including student housing that addresses a documented shortage of university accommodation in Novara. Central to the design strategy is the use of the adjacent Canale Quintino Sella as a natural cooling resource, assessed through a three-loop cooling system and a quantitative feasibility study of canal capacity, thermal discharge limits, and seasonal reliability. Beyond its environmental and functional aims, the project treats the retention of the complex’s industrial heritage, character, and spatial memory as central to the design, rather than incidental to its reuse. Ultimately, the thesis demonstrates how existing agro-industrial infrastructure and natural resources can reduce the environmental footprint of new data center development while addressing local social needs and preserving industrial identity.
La presente tesi propone il riuso adattivo di un ex complesso di stoccaggio agro-industriale, i Magazzini Generali Raccordati con Silos da Cereali di Novara, come struttura a uso misto che combina un data center e una residenza universitaria. Il progetto risponde ai costi ambientali dell’attuale boom dei data center guidato dall’intelligenza artificiale — consumo di suolo, carbonio incorporato e uso intensivo di acqua ed energia per il raffreddamento — proponendo il riuso adattivo, anziché la costruzione su terreno vergine (greenfield), come modello di sviluppo alternativo. La ricerca esamina la compatibilità strutturale, spaziale e infrastrutturale delle tipologie di stoccaggio industriale, in particolare i cluster di silos e i capannoni a campata lunga, con i requisiti tecnici dei data center. Si sostiene che la struttura verticale, cellulare e portante del cluster di silos sia intrinsecamente adatta a ospitare rack server e infrastrutture di raffreddamento, mentre i magazzini e gli edifici residenziali del complesso accolgono programmi destinati alla presenza umana, inclusa una residenza universitaria che risponde a una carenza documentata di alloggi per studenti a Novara. Elemento centrale della strategia progettuale è l’utilizzo del Canale Quintino Sella, adiacente al sito, come risorsa naturale di raffreddamento, valutato attraverso un sistema a tre circuiti (three-loop) e uno studio di fattibilità quantitativo sulla portata del canale, sui limiti di scarico termico e sull’affidabilità stagionale. Oltre agli obiettivi ambientali e funzionali, il progetto considera la conservazione del patrimonio industriale, del carattere e della memoria spaziale del complesso come elemento centrale del progetto, non semplicemente incidentale al suo riuso. In definitiva, la tesi dimostra come le infrastrutture agro-industriali esistenti e le risorse naturali possano ridurre l’impronta ambientale dello sviluppo di nuovi data center, rispondendo al contempo a esigenze sociali locali e preservando l’identità industriale del sito.
Inherited future : adaptive re-use of agro-industrial storage building as data center and student housing
Effifani, Djustine
2025/2026
Abstract
This thesis proposes the adaptive reuse of a former industrial storage complex, the Magazzini Generali Raccordati con Silos da Cereali in Novara, Italy, as a mixed-use facility combining a data center and student housing. It responds to the environmental costs of the current AI-driven data center boom — land consumption, embodied carbon, and intensive water and energy use for cooling — by proposing adaptive reuse, rather than greenfield construction, as an alternative growth model. The research examines the structural, spatial, and infrastructural compatibility of industrial storage typologies, specifically silo clusters and long-span warehouses, with the technical requirements of data centers. It argues that the silo cluster’s vertical, cellular, load-bearing structure is inherently suited to server-rack and cooling infrastructure, while the complex’s warehouse and residential buildings accommodate human-occupied program, including student housing that addresses a documented shortage of university accommodation in Novara. Central to the design strategy is the use of the adjacent Canale Quintino Sella as a natural cooling resource, assessed through a three-loop cooling system and a quantitative feasibility study of canal capacity, thermal discharge limits, and seasonal reliability. Beyond its environmental and functional aims, the project treats the retention of the complex’s industrial heritage, character, and spatial memory as central to the design, rather than incidental to its reuse. Ultimately, the thesis demonstrates how existing agro-industrial infrastructure and natural resources can reduce the environmental footprint of new data center development while addressing local social needs and preserving industrial identity.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_10_Effifani_Table of Content.pdf
accessibile in internet per tutti
Descrizione: Table of Content
Dimensione
43.11 kB
Formato
Adobe PDF
|
43.11 kB | Adobe PDF | Visualizza/Apri |
|
2026_10_Effifani_Panels.pdf
accessibile in internet per tutti
Descrizione: Presentation Panels
Dimensione
9.88 MB
Formato
Adobe PDF
|
9.88 MB | Adobe PDF | Visualizza/Apri |
|
2026_10_Effifani_Thesis.pdf
accessibile in internet solo dagli utenti autorizzati
Dimensione
48.73 MB
Formato
Adobe PDF
|
48.73 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/10589/266400