logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-06142026-182835


Tipo di tesi
Tesi di laurea magistrale
URN
etd-06142026-182835
Titolo
A Version-Aware Automation Framework for Updating Environmental Impact Data in an Industrial LCA Platform
Dipartimento
INFORMATICA
Corso di studi
DATA SCIENCE AND BUSINESS INFORMATICS
Relatori
.
relatore Prof. Frangioni, Antonio
Parole chiave
  • Agribalyse
  • carbon footprint,
  • Eco labeling,
  • Ecoinent,
  • Environmental Footprint (EF 3.1),
  • environmental footprint,
  • Environmental impact assessment,
  • ETL pipeline,
  • Fuzzy matching,
  • Industrial LCA platform,
  • LCA,
  • LCI,
  • openly,
Data inizio appello
17/07/2026
Consultabilità
Non consultabile
Data di rilascio
17/07/2029
Riassunto (Inglese)
This thesis addresses the challenge of managing evolving environmental data within an operational Life Cycle Assessment (LCA) platform for the agri-food sector. Developed during an internship at Foodpilot, a Paris-based sustainability SaaS company, the work targets a concrete industrial problem: external environmental databases such as Agribalyse and ecoinvent are updated regularly, but the platform lacked a structured mechanism to absorb those updates without disrupting its existing calculation logic.
The platform computes product-level environmental footprints by combining background LCA data with client-specific inputs collected through Semi-Specific Questionnaires. Its internal repository of Impact Characterisation Values, built incrementally over time, suffered from duplicates, inconsistent naming, missing metadata, and no version-tracking making reliable automation impossible.
To address this, the thesis designs and implements a version-aware automation framework comprising four stages: data quality assurance and standardisation of the internal repository; a multi-criteria matching pipeline that aligns internal references with external database processes using structured similarity scoring; automated extraction of updated values via the openLCA IPC interface; and a meta-version concept that bundles database versions into coherent, explicitly documented data environments.
The framework is validated through a milk production case study comparing results across Agribalyse 3.1.1 and 3.2 under multiple calculation contexts. Results confirm that database updates produce consistent, attributable changes PEF increases by 33% and Climate Change by 16% between versions demonstrating that the framework enables controlled, reproducible, and interpretable environmental data evolution within an industrial platform.
Riassunto (Italiano)
File