中山市大參林連鎖藥店的配送路徑優(yōu)化【摘要】本文以中山市大參林連鎖藥店為研究對(duì)象，在實(shí)習(xí)期間向公司獲得具體的配送數(shù)據(jù)，包括倉(cāng)庫(kù)資料、配送規(guī)劃、配送時(shí)間、配送量等，根據(jù)它目前的配送現(xiàn)狀發(fā)現(xiàn)中間可能存在的問題。為了避免模型數(shù)據(jù)在求解過程中對(duì)結(jié)果產(chǎn)生影響，將配送門店的平均需求進(jìn)行比例放大得到虛擬需求，使其與對(duì)比項(xiàng)在總配送量上達(dá)到一致。以得到最短的配送路徑為優(yōu)化目標(biāo)，出于對(duì)問題的特點(diǎn)以及算法的尋優(yōu)效率考慮，選擇單目標(biāo)遺傳算法的思路建立模型，并利用MATLAB7.0軟件進(jìn)行優(yōu)化求解，最終得到一個(gè)在配送總路程以及配送車輛數(shù)量上比現(xiàn)實(shí)方案更優(yōu)的配送方案，這個(gè)優(yōu)化結(jié)果在理論上能更好地提高配送效率以及降低物流配送成本?！娟P(guān)鍵詞】醫(yī)藥物流；連鎖藥店；路徑優(yōu)化；遺傳算法注：本論文（設(shè)計(jì)）題目來(lái)源于教師的國(guó)家級(jí)（或省部級(jí)、廳級(jí)、市級(jí)、校級(jí)、企業(yè)）科研項(xiàng)目，項(xiàng)目編號(hào)為：。DistributionpathoptimizationofzhongshandashenlinchaindrugstoreAbstractThisdissertationtakesDaShenLinchaindrugstoreinZhongshanastheresearchobject.Duringtheinternshipinthecompany,obtainspecificdistributiondata,includingwarehousedata,distributionplanning,distributiontime,distributionvolume,etc..Accordingtoitscurrentdistributionstatus,themiddlemayexistproblems.Inordertoavoidtheinfluenceofmodeldataontheresultsinthesolvingprocess,theaveragedemandofdistributionstoresisscaleduptoobtainthevirtualdemand,whichisconsistentwiththecomparisontermsinthetotaldistribution.Withtheoptimizationtarget,inordertogettheshortestdeliveryrouteforthecharacteristicsoftheproblemandalgorithmoptimizationefficiencyconsideration,choosethesingleobjectivegeneticalgorithmmodel,anduseMATLAB7.0softwareoptimizationsolution,endupwithatotaljourneyindistributionanddeliveryvehiclesquantityprojectschemeisbetterthanthephysicaldistribution,theoptimizationresultsintheorytobetterimprovethedistributionefficiencyandreducelogisticscosts.KeywordsPharmaceuticallogisticsPharmacychainPathoptimizationGeneticAlgorithm目錄TOC\o"2-3"\h\z\u\t"標(biāo)題1,1"1前言 參考文獻(xiàn)國(guó)家統(tǒng)計(jì)局.批發(fā)和零售業(yè)總體規(guī)模不斷擴(kuò)大行業(yè)結(jié)構(gòu)持續(xù)優(yōu)化[DB/OL].:/tjsj/zxfb/201912/t20191204_1715272.html2019-12-04.醫(yī)藥研究僧.2020年醫(yī)藥行業(yè)展望:醫(yī)保政策決定行業(yè)發(fā)展方向[DB/OL].:/p/121667.html,2020-01-07.G.B.Dantzig,J.H.Ramser.TheTruckDispatchingProblem[J].ManagementScience:1959,6(1):80-91.G.ClarkeandJ.Wight,SchedulingofVehiclesfromacentraldepottoanumberofdeliverypoints[J],OperationResearch,1964,12:568-581.S.Shahnejat-Bushehri,R.Tavakkoli-Moghaddam,S.Momen.HomeHealthCareRoutingandSchedulingProblemConsideringTemporalDependenciesandPerishabilitywithSimultaneousPickupandDelivery[J].IFACPapersOnLine,2019,118-123.FábioNeves-Moreira,MárioAmorim-Lopes,PedroAmorim.Themulti-periodvehicleroutingproblemwithrefuelingdecisions:Travelingfurthertodecreasefuelcost?[J].TransportationResearchPartE:LogisticsandTransportationReview,2020,133.MahdiAbbasi,MiladRafiee,MohammadR.Khosravi.Anefficientparallelgeneticalgorithmsolutionforvehicleroutingproblemincloudimplementationoftheintelligenttransportationsystems[J].JournalofCloudComputing:Advances,systemsandApplications,2020,9(2),87-102.郭耀煌,李軍,車輛優(yōu)化調(diào)度[M]成都:成都科技大學(xué),1994.譚穎.中小型醫(yī)藥配送中心實(shí)施連鎖藥店配送存在問題及對(duì)策研究[J].物流技術(shù).2010(Z2).鄭國(guó)華,周小強(qiáng),張力敏.基于時(shí)間窗的城市醫(yī)藥品動(dòng)態(tài)配送路徑優(yōu)化模型與算法[J].鐵道科學(xué)與工程學(xué)報(bào),2011,8(4):80-85.黃惠春,馬海英.醫(yī)藥配送中心排隊(duì)論運(yùn)用研究[J].物流技術(shù).2015(11).王雨寧.“兩票制”下KH醫(yī)藥經(jīng)營(yíng)公司藥品市內(nèi)配送庫(kù)存—路徑管理研究[D].北京:北京交通大學(xué),2018.宋曉昭.農(nóng)資物流配送路徑優(yōu)化研究——以H企業(yè)農(nóng)資配送為例[D].鄭州:鄭州大學(xué),2018.羅威,張曉蓉,張?zhí)?基于改進(jìn)Dijkstra算法的醫(yī)藥物流配送網(wǎng)絡(luò)優(yōu)化[J].價(jià)值工程,2019,21:121-122.陳愛軍.民營(yíng)藥店“四大家族”半年報(bào)大比拼[DB/OL].:/p/112790.html,2019-09-07.制藥在線.半年?duì)I收52億！大參林公布2019年度半年報(bào)[DB/OL].:/news/show-173519.html.胡潤(rùn)研究院.2019胡潤(rùn)中國(guó)500強(qiáng)民營(yíng)企業(yè)[DB/OL].:/CN/Article/Details?num=956B3DE566B6,2020-01-09.火石創(chuàng)造.我國(guó)醫(yī)藥物流產(chǎn)業(yè)發(fā)展現(xiàn)狀分析[DB/OL].:/intelligence/insight119008.html.胡斐,趙治國(guó).基于自適應(yīng)模擬退火遺傳算法的非線性方程組求解[J].微型電腦應(yīng)用,2010,26(09):25-27+66.張建民.基于改進(jìn)蟻群算法的物流配送路徑問題研究[D].新疆:新疆農(nóng)業(yè)大學(xué),2010.致謝一轉(zhuǎn)瞬，大學(xué)四年的時(shí)光就要過去了，四年的學(xué)習(xí)讓我獲益匪淺，在充實(shí)的校園生活中，不僅讓我學(xué)到了各種各樣的知識(shí)，也磨練了我的心智，讓我愈發(fā)沉穩(wěn)，即將踏入社會(huì)，開始新的征程，此刻心情難以言表，感謝這幾年一直幫我的同學(xué)和老師，萬(wàn)分感激！本文是在王景峰老師的悉心指導(dǎo)下完成的，感謝老師在我進(jìn)行模型選擇以及算法代碼書寫上給予的專業(yè)的幫助，也感謝大參林醫(yī)藥集團(tuán)股份有限公司在所需數(shù)據(jù)上的無(wú)私提供，他們高尚的人格、負(fù)責(zé)的態(tài)度、嚴(yán)謹(jǐn)?shù)那髮W(xué)作風(fēng)讓我欽佩，在此，我對(duì)它們表達(dá)由衷的敬意和真摯的感謝！附錄A遺傳算法的代碼clc;clear;closeall;%%參數(shù)配置populationSize=100;%種群規(guī)模maxGeneration=2000;%最大進(jìn)化代數(shù)crossoverRate=0.6;%交叉概率mutationRate=0.02;%變異概率model=initModel();%問題定義%%初始化population=initialPopulation(populationSize,model);%初始化種群popFitness=getFitness(population,model);%計(jì)算種群適應(yīng)度numOfDecVariables=size(population,2);%決策變量維度bestIndividualSet=zeros(maxGeneration,numOfDecVariables);%每代最優(yōu)個(gè)體集合bestFitnessSet=zeros(maxGeneration,1);%每代最高適應(yīng)度集合avgFitnessSet=zeros(maxGeneration,1);%每代平均適應(yīng)度集合%%進(jìn)化fori=1:maxGeneration[popFitness,index]=sort(popFitness);%根據(jù)適應(yīng)度從小到大排序population=population(index,:);bestIndividualSet(i,:)=population(end,:);%第i代最優(yōu)個(gè)體bestFitnessSet(i)=1./popFitness(end);%第i代最高適應(yīng)度avgFitnessSet(i)=sum(1./popFitness)./populationSize;%第i代種群平均適應(yīng)度showIndividual(population(end,:),model);bestFitnessSet(i)newPopulation=selectionOperation(population,popFitness);%選擇操作newPopulation=crossoverOperation(newPopulation,crossoverRate);%交叉操作newPopulation=mutationOperation(newPopulation,mutationRate);%變異操作newPopFitness=getFitness(newPopulation,model);%子代種群適應(yīng)度