Production optimisation in the mechanical department of Scuderia Ferrari

This thesis presents the digital transformation pathway of Ferrari Gestione Sportiva’s Mechanics Department (MecSF), aiming to make decisions along the Planning–Scheduling–Execution cycle predictable, measurable, and reproducible in a high-mix, low-repeatability F1 job-shop context. Following the as-is diagnosis—characterized by heterogeneous tools (Excel/ERP extracts), information latency in the PR→PO flow, and limited representation of real constraints (shifts/presence/skills, tools/fixtures, semi-finished parts, metrology/quality checkpoints)—we motivate critical issues around ungoverned saturations, lead-time variance, and dependence on tacit expertise. The proposal defines a to-be architecture based on a Digital Thread, constrained scheduling, and native Discrete-Event Simulation (DES) to support baseline/stress what-ifs, explicitly including TTM specificity (additional stages and α = 1) and operational visibility via an interactive Gantt. We formalize functional requirements (integration with ERP/LN, Cost Control, HR, tooling; joint human–machine–tool–material constraints; reporting with scenario versioning and audit of adjusted times) and non-functional requirements (scalability to datasets ≥ 1,000 codes, ~60 s recalculation, RBAC security). Methodologically, the work structures software scouting in engineering terms (scripted C1–C5 demos, 0–5 scorecards, AHP for multi-criteria weighting) and plans a Proof-of-Concept on anonymized real data (S0–S5 scenarios) with explicit gates: constraint coverage ≥ 95%, Δρ ≤ 10 p.p., ΔLT ≤ 15%, recalculation within threshold, and end-to-end traceability. The thesis contributions are: (i) an operational meta-model for the F1 job-shop (constraints, policies, metrics), (ii) a replicable, vendor-agnostic evaluation method, and (iii) an adoption playbook toward the Production Digital Twin (shadow-mode pilot, adjusted-time feedback loop). The expected outcome is a shift from reactive to predictive, governed operations, with benefits on OTIF, plan stability, and time-to-decision.