Huff Competetition Location Problem Based On Improved Cuckoo Algorithm

With more and more competition of various industries in the 21st century,competitive thinking gradually penetrated into location problems and became one of hot problems of location at this century.Competitive location problems belonged to one branch of location problems.The research was under the environmental circumstances to let enterprise make rational solutions and make it dominant at the beginning of choice location so that enterprises and other similar company obtain more market share.Among the factors of influencing competitive location,the requirement of customers and their choice behavior are the most importance of two points.At the most parts of documents,they often ignored customers' choice behavior or attractiveness of fixed homogeneity in general,only to connect customers' choice behavior and distances in simplicity,and based on it,this essay made use of corrected huff function to connect customer' s choice behavior with features of its own facility and quantitative choice behavior of customers and establish competitive location models based on huff functions and further consideration of influences with aggregation effect to customer' s choice behavior to establish the competitive location models based on aggregation effect and put forward improved model of cuckoo algorithm.This paper mainly studies the problem of huff competition location based on the improved cuckoo algorithm.The main content of this essay as below:firstly introduced the studying background and main study result at home and abroad,then introduced the basis knowledge of competitive location problem and related knowledge and basic condition of cuckoo algorithm,and then provided main works of this essay.Firstly,improved cuckoo algorithm.Cuckoo algorithm was a kind of natural heuristic algorithm.Compared with other heuristic algorithm,it had simple structure and less control parameters and its unique Levy flight mechanism gave him the ability to jump out of local extremism and good global search and strong robustness.However,just because of Levy flight mechanism,step length is a random parameter to have efficient control of step length which led to premature convergence or long search time in the process of searching.To the problems,the third part of essay put forward a kind of cuckoo algorithm of adaptive search step length based on fitness function.In accordance with the adaptive value of solution,we would deal with differential treatment for different solutions to take different foot step for good tactics and verified that improved cuckoo algorithm was stronger than differential treatment for different solutions with faster speed of convergence and higher accuracy of the solutions.Secondly,mainly studying competitive location problems based on the corrected huff functions.Through attraction function of improved classic huff functions,the corrected huff functions were used to qualify the choice behavior of customers to make sure the probability of determining the customer's patronage facilities so as to get customer's requirement that the facilities could be captured to put the probability of customers patronizing facilities determined by the huff function to be added to the model,and established competitive location models based on corrected huff functions,and then made use of AFCS algorithm in the chapter three to get the models.Thirdly,mainly studying competitive location problems based on aggregation effect.Considering the influence of aggregation effect caused by facilities gathering on competitive location problems,i.e.when the same facilities with relationship of competition were on the same location,the facilities gathered to make the competition of regions larger and make attractiveness of facilities to customers in the region more to make customers' choice behavior changed.When the two facilities that have competitive relationships were made to choose location,facilities gathering to produce aggregation effects,The aggregation effect was described by introducing attractiveness growth rate.When the aggregation effect occurred,the attractiveness of facilities in the aggregated area was times,that is,when considering the aggregation effect of facilities,the definition of attractive functions of huff competitive location mode changed,the attractive function in the area of aggregation effect became times.The attraction of facilities remains unchanged in the area without aggregation effect to make the probability of patronage of facilities customers increased in the area of aggregation effect to influence customer's choice behavior so as to influence the enterprise profits.Based on this,the competitive location model based on the aggregation effect was established,the solutions of models used the improved AFCS algorithm of third chapter.