Workshop Description
Quantum amplitude estimation for logistics risk metrics: stockout probability, service-level CVaR, and inventory tail risk across large SKU portfolios. Covers QAE variants, NISQ limitations, and realistic adoption timelines for inventory risk teams.
Classical Monte Carlo simulation remains the workhorse for computing risk metrics in logistics: stockout probabilities, fill rates, and service-level CVaR across portfolios of thousands of SKUs. The computational cost scales linearly with the number of samples needed for a given confidence level. Quantum amplitude estimation (QAE), originally proposed by Brassard et al. (2002), achieves a quadratic speedup, requiring roughly the square root of the classical sample count. Modern iterative variants (Suzuki et al. 2020, Grinko et al. 2021) reduce the qubit overhead of canonical QAE, making near-term experiments more tractable. The critical question for logistics organisations is whether the circuit depth requirements of QAE can be met on current or near-term hardware for operationally relevant portfolio sizes. This workshop examines that boundary with specific reference to inventory risk workloads, works through the encoding and estimation process, and provides an honest assessment of where the technology stands relative to classical alternatives.
What participants cover
- Classical Monte Carlo limitations: sample complexity, convergence rates, and computational cost at scale for large SKU portfolios
- QAE algorithm variants: canonical (Brassard et al.), iterative (Suzuki et al. 2020), and maximum likelihood approaches with their respective qubit and depth requirements
- Logistics risk applications: encoding demand distributions, estimating stockout probability, computing fill-rate metrics and service-level CVaR
- Circuit depth realities: why canonical QAE requires fault-tolerant hardware for practical logistics scale, and what near-term alternatives exist
- Published benchmark evidence: comparing QAE performance on simulators and current devices against classical Monte Carlo baselines
- Adoption framework: decision criteria for when QAE exploration is justified, vendor capabilities, and hybrid classical-quantum pipeline architectures