KETs2026 reloaded

24 Sept 2026 | Germany

ExpertiseUpdated on 10 September 2026

Modelling, simulation & predictive analytics with preregistered validation: Monte Carlo, causal inference, uncertainty budgets, quantum-inspired engines

Founder & CEO at SOVEREIGNPIALPHA FRANCE LTD

PARIS, France

About

SOVEREIGNPIALPHA France Ltd offers modelling, simulation and predictive-analytics expertise to research and innovation consortia, industrial partners and public bodies, either as a work-package contributor or as a subcontracted expert.

METHODS

Monte Carlo and scenario simulation for techno-economic, supply-chain and policy questions (for example CRMA 2030 benchmark feasibility, processing-concentration and diversification scenarios); causal inference for R&D and programme evaluation (directed acyclic graphs, difference-in-differences, instrumental variables, regression discontinuity, sensitivity to unmeasured confounding); Bayesian and frequentist experimental design (expected information gain, D-optimality, sample size and power, sequential designs); measurement uncertainty budgets under the GUM; dimensional analysis and scaling laws as first-line sanity checks; predictive-maintenance models (calendar ageing, remaining useful life, contextual impedance diagnostics for batteries); quantum-inspired optimisation and inference engines (QAOA, variational methods, tensor-network and open-quantum-system formalisms simulated on classical hardware, Qiskit / IBM Quantum where relevant).

VALIDATION DISCIPLINE

Preregistered refutation campaigns: acceptance thresholds fixed before execution, entity-level data partitions, declared recalibrations retested at the same thresholds, independent triple review of every deliverable, SHA-256 manifests of data and code, Harvard-referenced reports under a zero-fabrication policy. Results that fail are reported as such: NO-GO verdicts are documented rather than hidden.

DELIVERABLES

Model specifications and code (Python: numpy, scipy, scikit-learn, XGBoost, PyTorch; Qiskit), validation reports, uncertainty and sensitivity analyses, decision-ready dashboards, open-science deposits (Zenodo DOIs) and publication-quality manuscripts.

TYPICAL USES IN A PROJECT

Techno-economic and life-cycle scenario modelling; evaluation of pilot results with confidence bounds; predictive models for recycling, sorting, battery and materials processes; policy-impact modelling for CSA and RIA actions; independent verification of partner claims.

Contact: Stephane Bobe Engueleguele, Founder & CEO, via b2match messaging or a B2B meeting slot on 24 September 2026 and during the long-term matchmaking until January 2027.

Field

  • Modelling, simulation, predictive technologies
  • Raw Materials
  • Sustainability
  • AI-GenAI /Data/Robotics
  • Quantum and High Performance Computing
  • Batteries
  • Testing & Analysis

Organisation

SOVEREIGNPIALPHA FRANCE LTD

SME

PLACHY BUYON, France

Similar opportunities