In high-mix, low-volume manufacturing, corporate competitiveness relies on the precision of information flows. This thesis, conducted at ELFIM S.r.l., analyzes and optimizes the industrial quotation process with the aim of integrating and automating it in line with Industry 4.0 principles. Adhering to Total Quality Management (TQM) concepts, the study addresses data quality issues identified by comparing the theoretical Cost Breakdown Structure (CBS) with actual cost data extracted from the Manufacturing Execution System (MES). Through SQL queries and Excel automation, manual data entry was eliminated, establishing a standardized procedure for reconciling and normalizing shop-floor data. The application of statistical tools—such as Pareto charts, Lorenz curves, and efficiency KPIs—made it possible to isolate key operational bottlenecks and quantify cost and time variances. Consequently, this work provides the methodological foundation and the "clean" dataset required for redesigning cycle times and for the future training of predictive machine learning algorithms.
Nelle produzioni manifatturiere high-mix, low-volume, la competitività aziendale si basa sulla precisione dei flussi informativi. Il presente lavoro di tesi, svolto presso ELFIM S.r.l., analizza e ottimizza il processo di preventivazione industriale con l'obiettivo di integrarlo e automatizzarlo in ottica 4.0. In coerenza con i principi del Total Quality Management (TQM), lo studio affronta le criticità di Data Quality emerse dal confronto tra la Cost Breakdown Structure (CBS) teorica e i consuntivi reali estratti dal Manufacturing Execution System (MES). Tramite query SQL e automazioni in Excel, è stato eliminato il data-entry manuale, creando una procedura standardizzata per la riconciliazione e la normalizzazione dei dati di fabbrica. L'applicazione di strumenti statistici quali Diagramma di Pareto, Curva di Lorenz e KPI di efficienza, hanno permesso di isolare i principali colli di bottiglia operativi e di quantificare gli scostamenti economico-temporali. Il lavoro fornisce così la base metodologica e il dataset "pulito" necessari per la riprogettazione dei tempi ciclo e il futuro addestramento di algoritmi predittivi di Machine Learning.
STUDIO E ANALISI DELLE CRITICITÀ NEI FLUSSI INFORMATIVI PER UNA CBS INTEGRATA E PREDITTIVA NEI PROCESSI DI MANUFACTURING HIGH-MIX, LOW-VOLUME.
MARVULLI, LIDIA
2025/2026
Abstract
In high-mix, low-volume manufacturing, corporate competitiveness relies on the precision of information flows. This thesis, conducted at ELFIM S.r.l., analyzes and optimizes the industrial quotation process with the aim of integrating and automating it in line with Industry 4.0 principles. Adhering to Total Quality Management (TQM) concepts, the study addresses data quality issues identified by comparing the theoretical Cost Breakdown Structure (CBS) with actual cost data extracted from the Manufacturing Execution System (MES). Through SQL queries and Excel automation, manual data entry was eliminated, establishing a standardized procedure for reconciling and normalizing shop-floor data. The application of statistical tools—such as Pareto charts, Lorenz curves, and efficiency KPIs—made it possible to isolate key operational bottlenecks and quantify cost and time variances. Consequently, this work provides the methodological foundation and the "clean" dataset required for redesigning cycle times and for the future training of predictive machine learning algorithms.| File | Dimensione | Formato | |
|---|---|---|---|
|
MARVULLI_tesi.pdf.pdf
File consultabile da postazioni biblioteca
Dimensione
3.01 MB
Formato
Adobe PDF
|
3.01 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/110751