Loading...

Improving Patient Flow in the Emergency Department with Dynamic Scheduling Using Phase-Type Distribution Function

Eshghi, Alireza | 2023

98 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56311 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Varmazyar, Mohsen
  7. Abstract:
  8. The Emergency Department is one of the most important departments in a hospital, which is continuously active and plays a crucial role in saving patients' lives. For this reason, using research methods in operations and optimization can greatly help improve the performance of the Emergency Department. In this thesis, the emergency scheduling problem is examined with the objective function of minimizing the total weighted sum of job completion times, considering the random nature of service times. Since the objective function of the problem is to minimize the total job completion time, the job completion time is obtained as the sum of several random variables. Considering the additivity property of the normal distribution function, this distribution is considered a suitable option for obtaining the patient's job completion time. However, this distribution does not provide accurate results for real data that exhibit skewness. For this reason, in this thesis, the phase-type distribution is used to fit the service time data. Due to the use of the phase-type distribution, the structure of this distribution function and its types are introduced. Additionally, the Expectation Maximization algorithm is used to estimate the parameters of the phase-type distribution function that is fitted to the emergency service data. The phase-type distribution is used to estimate the completion time of operations for patients in the emergency department due to its additivity property and high accuracy. For this purpose, this distribution function is compared with other distribution functions such as normal, log-normal, and gamma distributions in terms of goodness of fit. In the next stage, a two-stage scheduling problem is solved. In the first step, the problem under investigation is a parallel machine problem with the objective of minimizing the weighted waiting time of patients, where the machines are the existing beds in the emergency unit. The next step is a flexible flow shop problem with the objective of minimizing the weighted sum of operation completion times for patients, while maximizing resource utilization. Due to the NP-hardness of the problem, an heuristic algorithm is used to generate an initial solution, and then in the next step, the tabu search and simulated annealing metaheuristic algorithms are used to refine this solution. Finally, the algorithm is executed in several different scenarios to investigate the effect of resources and patient arrival rate on the completion time of patients' operations, and the results are examined
  9. Keywords:
  10. Dynamic Scheduling ; Heuristic Algorithm ; Meta Heuristic Algorithm ; Emergency Department Overcrowing ; Emergency Department ; Continuous Phase-Type Distribution ; Expectation Maximazation Algorithm ; Minimizing Total Task Completion Time ; Emergency Department Scheduling

 Digital Object List

 Bookmark

No TOC