Multi-neighborhood Local Search for the Patient Admission Problem

Authors:
Sara Ceschia;Andrea Schaerf
Affiliations:
DIEGM, University of Udine, Udine, Italy;DIEGM, University of Udine, Udine, Italy
Venue:
HM '09 Proceedings of the 6th International Workshop on Hybrid Metaheuristics
Year:
2009

Citing 5
Cited 1

Local Search in Combinatorial Optimization

Local Search in Combinatorial Optimization
EasyLocal++: an object-oriented framework for the flexible design of local-search algorithms

Software—Practice & Experience
Stochastic Local Search: Foundations & Applications

Stochastic Local Search: Foundations & Applications
The State of the Art of Nurse Rostering

Journal of Scheduling
Hybrid approaches for rostering: a case study in the integration of constraint programming and local search

HM'06 Proceedings of the Third international conference on Hybrid Metaheuristics

An investigation on the generality level of selection hyper-heuristics under different empirical conditions

Applied Soft Computing

Quantified Score

Hi-index	0.01

Visualization

Abstract

We propose a multi-neighborhood local search procedure to solve a healthcare problem, known as the Patient Admission problem. We design and experiment different combinations of neighborhoods, showing that they have diverse effectiveness for different sets of weights of the cost components that constitute the objective function. We also compute some lower bounds on benchmark instances based on the relaxation of some constraints and the solution of a minimum-cost maximum-cardinality matching problem on a bipartite graph. The outcome is that our results compare favorably with the previous work on the problem, improving on all the available instances, and are also quite close to the computed lower bounds.