| Sumario: | Patient transportation within departments (i.e., Intra-Hospital) by hospital transporters involves frequent operations, and their outcomes impact multiple stakeholders. Stakeholders include patients, healthcare and medical professionals, and hospital managers. Hospital managers in the transportation department coordinate patient transport tasks. Each hospital has specific requirements and complies with regulations (e.g., Labor and Data Protection). Uncertain situations (i.e., task delays, new tasks, or task cancellations) affect multiple stakeholders and service quality. Thus, route planning for Intra-Hospital Patient Transport is complex, multi-objective, and non-continuous, which ensures fair workload distribution for high-quality service. The objectives are to optimize the number of transporters and the parameters (i.e., travel and idle times and their statistical measures) for efficient route plans with maximal operational flow on each shift. Our work employs multimodal methods (i.e., a combination of metaheuristics and Discrete Event Simulation) with perceptible strategies (i.e., Mixed Integer Programming and scoring strategy) to improve automated route plans. Our methods account for uncertainties and practical characteristics derived from retrospective data. Our empirical study compares the performance of multimodal methods across different data sets extracted from one-month retrospective data provided by the hospital. The empirical results show that Discrete Event Simulation provides 45% - 65% computationally fast solutions. Thus, multimodal optimization methods guided by our perceptible strategies build statistical models for each data set, considering multiple objectives to support hospital managers by automating efficient route plans.
|
|---|