An optimization model for scheduling  patients in destination medical centers

The objectives of this research is two-fold: (1) minimize the deviation from patient’s preferred start time, and (2) minimize the inter-waiting time between procedures of patients at a destination medical center. This scheduling problem is modeled as a non-permutation hybrid flow shop with an assumption that providers have specific periods of unavailability. The research problem is known to be NP-hard. Therefore, a hybrid particle swarm optimization algorithm equipped with local search and a heuristic mining algorithm is proposed to reach a near-optimal solution efficiently. The parameters of the algorithms are first tuned using Taguchi design under different problem sizes. In the proposed Taguchi design, the parameter level with the lowest relative percentage deviation is selected as the best parameter level. Relative percentage deviation is defined as the relative difference between the obtained solution and the best solution for a given instance. Then, numerical experiments under different problem parameters’ settings are used to evaluate the performance of the proposed algorithm. Computational results confirm that the proposed algorithm has a statistically lower relative percentage deviation compared to particle swarm optimization equipped with local search and simulated annealing. Additionally, the proposed algorithm has a good performance compared to CPLEX in reaching optimal/near optimal solutions.
Source: Operations Research for Health Care - Category: Hospital Management Source Type: research