Integrated Dynamic Optimization Of Blood Collection Point Location-resource Allocation Under Uncertain Environment

As a scarce resource for medical assistance,blood plays an irreplaceable role in the medical system.It has the characteristics of single source and extremely high shortage cost.The characteristics of blood determine multiple uncertainties of the influencing factors of blood collection site selection and resource allocation,which makes decision-making more difficult.At present,the unreasonable setting of blood collection points in our country makes it inconvenient for people to donate blood,which greatly reduces the amount of blood collection.At the same time,the blood donation rate is far lower than the demand of blood consumption.Once an emergency occurs,it is easy to face the dilemma of blood shortage.At present,there are few researches on the location and resource allocation of blood collection points,and there are almost no researches considering the location models of two types of blood collection facilities.Therefore,studying the location and resource allocation of blood collection points from the perspective of multiple facilities,multiple cycles and considering uncertainty is of great practical significance for increasing blood collection volume and enhancing the level of blood security.Firstly,the thesis starts with the model research,and guarantees the reasonable utilization of personnel and equipment resources under the premise of considering the maximum total coverage.Based on the generalized maximum coverage model,full consideration of various influencing factors in coverage weights(regional flow of people,blood donation activity,personnel accessibility,blood collector service level,blood collection interval,etc),supplemented by grey cluster assessment,Constructing a deterministic parameter optimization model that maximizes the total coverage under the weight dynamic evaluation.On this basis,considering the uncertainty of the number of available blood collection vehicles and the number of available personnel and equipment in each cycle in reality,the robust control idea is used to process the data,so as to construct the robust optimization corresponding model under the uncertain level parameters.Secondly,on the basis of the introduction of dynamic weighting factor,an improved grey wolf optimization algorithm is designed in combination with the simulated annealing algorithm and the 3opt local optimization algorithm.Based on the data of Chongqing Blood Center and other places,a simulation case was designed to solve the two models.By comparing the operation results of the improved grey Wolf algorithm with the original grey Wolf algorithm and particle swarm optimization algorithm,it is found that the improved grey Wolf algorithm has great advantages in terms of the convergence speed and the stability of the solution for dealing with uncertain parameters.Finally,In order to verify the influence of uncertainty level parameters and coverage recovery coefficient on the decision results,sensitivity analysis is conducted respectively.According to the analysis results,it is found that the uncertainty level parameter is too small,which will cause waste of resources.Appropriate selection of the uncertainty level parameters can realize the reasonable planning of the blood collection vehicle and help the decision maker make the optimal decision.The value of the coverage recovery coefficient will have an impact on the final blood collection vehicle planning and the number of personnel and equipment configuration.Reasonable setting of the coverage recovery coefficient can maximize the coverage,collect more blood,and ensure the reasonable utilization of resources.