Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/513291
Title: | Jumping Particle Swarm Optimisation algorithm for Nurse Rostering Problems |
Authors: | Yahya Zakaria Yahya Al-Arajy (P60048) |
Supervisor: | Salwani Abdullah, Prof. Dr. |
Keywords: | Schedule Nurse Mathematical optimization |
Issue Date: | Jan-2017 |
Description: | Hospital and healthcare services are recognised as one of the most important needs for every living person. Efficient management of healthcare resources is part of the essential factors to improve the quality of healthcare system. Nurse Rostering Problem (NRP) is one of the challenging problems in healthcare resources management. NRP is a scheduling problem that deals with assigning nurses to their duties according to their preferences, skill and experience subject to a specified constraint. Although many published works in the literature have covered a good angle to challenge this problem, the problem becomes more difficult when various types of constraints need to be satisfied. In addition, treating NRP as a multi-objective optimisation problem has not been given sufficient concerns in the literature, due to the complex nature of its real-world applications. This research proposed a population-based method called Jumping Particle Swarm Optimisation (JPSO) algorithm to solve single and multi-objective variants of NRP. Firstly, a various improvements for JPSO to effectively handle single objective NRP are proposed. The proposed JPSO uses multiple neighbourhood structures to enhance the exploration at the solutions regions and aid on escaping from local optimum. In order to maintain the balance between the exploration and exploitation in producing better quality solutions and overcome premature convergence, the proposed JPSO is hybridised with a new local search that uses an acceptance criterion that allows worse solutions to be accepted to jump from a local optima point. Secondly, the single objective NRP is formulated as a multi-objective combinatorial optimisation problem and adapted the proposed JPSO to address it. Two variants of multi-objective JPSO where proposed: weighted sum multi-objective JPSO and Pareto based multi-objective JPSO. Computational experiments are tested on 60 instances with three levels of complexity that are based on real world constraints. The comparison between the proposed approach with other state-of-the-art approaches shows that the JPSO is able to find an optimal solution with high frequency (73.3%) for most of the tested instances. On average, the proposed algorithm displayed relatively competitive results and outperformed some of the existing approaches.,Certification of Master's/Doctoral Thesis" is not available |
Pages: | 167 |
Call Number: | QC20.7.M27A733 2017 3 tesis |
Publisher: | UKM, Bangi |
Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ukmvital_97951+SOURCE1+SOURCE1.0.PDF Restricted Access | 614.5 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.