Application Of Improved Bacterial Foraging Optimization Algorithm In Nurse Scheduling Problems

As the basic component of hospital workings,nurses are indispensable to the hospital normal operation.Having an efficient and reasonable working time can not only reduce the pressure of work,improve the quality of work,but also can reduce the human cost of the hospital,so the nurse scheduling has become one of the important content of hospital management.However,actual nurse scheduling needs to considering labor rules,nurse skills and personalized requests,etc.The nurse scheduling problem is a complex combinatorial optimization problem,known as NP-hard,which is deeply concerned about in abroad.Most of the domestic research of nurse scheduling problem is to simulate simple manual scheduling methods,not conform to various constraint conditions,and lack of effective optimization mechanism,so how to make a fast and efficient nurse scheduling is the most urgent problem currently.And in recent years,some researchers have begun to use intelligent algorithm to solve this problem.This article is aimed at this issue for study.This thesis summarized the state of Nurse Scheduling Problem(NSP)and Bacteria Foraging Optimization(BFO)Algorithm,and described the purpose and significance of the study,introduced the related theoretical basis and added nurse preferences to improve the NSP standard model,meanwhile,proposed the improved BFO algorithm,and applied the improved BFO into the nurse scheduling problem at the end.This paper has carried out the following research under the support of the National Natural Science Foundation of China(71571120,71271140,71471158)and the Guangdong Provincial Natural Science Foundation(2016A030310074):(1)In order to reduce the total cost of NSP function,we first made a survey of nurses’ scheduling in a real hospital department;Basing on these,we considered specific working shifts,partners,and personalized requirements,added the nurse preference constraints to the standard NSP model,making the problem more close to the hospitals real scheduling situation.(2)Because of the three nested loop structure,traditional BFO has a slow convergence speed,this paper introduces a hierarchical communication mechanism(SRBFO-RC)to solve the nurse scheduling problem.In order to verify the validity of the proposed algorithm,we selected the BFO and PSO algorithm to compare with SRBFO-RC and test on the standard test functions.The results show that the improved SRBFO-RC algorithm has greatly improved this problem.(3)Based on the proposed SRBFO algorithm,this thesis gave a solution to solve the NSP model with nurse preference constraints,and discussed the individual initialization and encoding.In addition,a real hospital nurse data were used to have an experiment,and compared with the traditional BFO,PSO.Results have verified that the proposed algorithm has a good application potential in solving this kind of problem.