Industry Case Study on Digitized Counterdiabatic Quantum Optimization for a Combinatorial Optimization Problem

In this presentation, we discuss quantum computing outcomes from an engagement with a German manufacturing customer, focusing on a combinatorial optimization challenge: the job shop scheduling problem (JSSP). Our work concentrates on specific sub-JSSP instances involving 16 variables, equivalent to a quantum formulation with 16 qubits. We utilize both pure and hybrid quantum strategies, drawing on Kipu’s digitized counterdiabatic quantum optimization (DCQO) approach to address these instances. The quantum circuits developed are executed on IBMQ's superconducting hardware and IonQ’s trapped-ion devices. We evaluate the effectiveness of our methods through success probability and approximation ratio metrics, demonstrating that both pure and hybrid DCQO methods surpass the Quantum Approximate Optimization Algorithm (QAOA) across all JSSP instances tested. To be specific, we find a four-fold improvement in success probability with DCQO in comparison with QAOA, and up to three orders of magnitude with hybrid DCQO. Notably, in experiments with IonQ hardware, we consistently achieve the optimal solution with a significant probability, indicating a highly efficient solution approach.