A hybrid genetic algorithm with 2D encoding for the scheduling of rehabilitation patients. (November 2018)

• Record Type: Journal Article
• Authors: Huynh, Nhat-To; Huang, Yi-Chao; Chien, Chen-Fu
• Is Part Of: Computers & industrial engineering. Volume 125(2018)
• Page Start: 221
• Page End: 231
• Publication Date: 2018-11
• Subjects: Rehabilitation scheduling -- Genetic algorithm -- 2D encoding -- Timetabling -- Hospital management; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2018.08.030
• Languages: English
• ISSNs: 0360-8352

Abstract

Graphical abstract: An effective approach that integrates genetic algorithm and simulated annealing is developed for rehabilitation scheduling that is a hybrid shop scheduling problem. The proposed approach employs 2D encoding and adapted timetabling with better performance than conventional approaches. The proposed approach is validated with a number of designed cases in different sizes and has shown its practical viability via empirical study. Highlights: An approach is developed for rehabilitation scheduling that is a hybrid shop. An approach integrating genetic algorithm and simulated annealing is proposed. 2D encoding and adapted timetabling are developed to enhance this approach. This approach is validated with realistic data and designed cases of different sizes. An empirical study has shown practical viability of this approach. Abstract: Owing to aging population, rehabilitation treatments are increasing important, yet the scheduling of rehabilitation patients has been received little research attention. Focusing on real needs for hospital management, this study aims to develop a hybrid genetic algorithm that integrates genetic algorithm and simulated annealing to solve the present problem effectively and efficiently. A novel 2-dimension encoding method was designed to represent for the sequence of patients and their operations. In addition, a timetabling algorithm was developed for decoding a sequence into the best solution efficiently. To validate the proposed approach, a number of experiments based on the real data in a medical center were designed for comparison. The results have shown the proposed approach with better performance than conventional approaches. An empirical study has shown practical viability of this approach.