Abstract
The nurse scheduling problem (NSP) is multi-criteria decision problem concerned with allocation of shift schedules to available nurses over a planning horizon of one week to one month. Developing interactive, multi-objective, and fast optimization approaches for solving the NSP is imperative. The NSP has posed continued challenges to decision makers in healthcare organizations. This paper presents a fuzzy simulated metamorphosis algorithm (FSM), inspired by biological metamorphosis evolution. The algorithm mimics the metamorphosis process by going through three phases, namely, initialization, growth, and maturation. Initialization randomly generates a single candidate solution using a guided constructive heuristic. Subsequently, the algorithm goes through growth and maturation loops, till termination criteria are satisfied. Computational results based on benchmark problems in the literature demonstrate that, compared to related metaheuristic algorithms, FSM is more efficient and effective, producing better solutions within reasonable computation times.
| Original language | English |
|---|---|
| Article number | 14 |
| Pages (from-to) | 222-231 |
| Number of pages | 10 |
| Journal | Engineering Letters |
| Volume | 23 |
| Issue number | 3 |
| Publication status | Published - 10 Jul 2015 |
Keywords
- Algorithm
- Evolution
- Metaheuristics
- Multi-criteria decision methods
- Optimization
- Simulated metamorphosis
ASJC Scopus subject areas
- General Engineering