QAOA-Based Optimization for Healthcare Resource Scheduling Under Constraints

सार

Healthcare scheduling assigns staff, rooms, and equipment to patient tasks over time while meeting clinical priorities and operational constraints. This paper formulates constrained healthcare resource scheduling as a binary optimization problem and applies the Quantum Approximate Optimization Algorithm (QAOA) as a hybrid quantum–classical solver. Hard constraints (e.g., shift limits, skill matching, resource capacity, and priority rules) are encoded as penalty terms in a QUBO/Ising cost Hamiltonian, so the optimizer searches mainly within feasible schedules. We evaluate the approach on synthetic instances that reflect variable patient arrivals, mixed acuity, and limited resources, and compare it against time-limited integer linear programming (ILP) and a genetic algorithm baseline. In simulation, QAOA improves average patient waiting time by 12.1%, increases scheduling efficiency by 15%, and raises resource utilization by 20% versus the baseline heuristic, while maintaining feasibility across larger instances where exact solvers become slow. The results indicate that QAOA can be useful for near-real-time re-scheduling on small-to-medium subproblems, while current hardware limits suggest a hybrid workflow for practical deployment. [1][3]

Keywords: QAOA, Healthcare Resource Scheduling, Scheduling Algorithms, Quantum Optimization