基于机器学习的区域性综合资源调度与优化算法设计

打开文本图片集

中图分类号：TN915.04-34；TP391.9 文献标识码：A 文章编号：1004-373X（2025）16-0128-05

Machine learning based design of regional comprehensive resource scheduling andoptimizationalgorithm

LEIChengtao¹，ZHANGLi²，HANTengfei1 （1.College of Architecture and Environment，Sichuan University，Chengdu 61oo65，China; 2.Sichuan University Engineering Design and Research Institute Co.，Ltd.， Chengdu 61O065，China）

Abstract：Inorderto enhancethe emergencyresource scheduling capabilityinurbandisaster preventionandreduction，a regional comprehensiveresourceschedulingandoptimizationalgorithmbasedonmachinelearning isproposed.Inthisalgorithm， thecontinumapproximation（CA）modelisused todeterminetheoptimalreplenishmentquantityandinventorylevelforsevering a specific location，and K -means clustering algorithm is used to cluster the pre allocated routes，making emergency resource schedulingmoreeficient.ByintroducingtheChristofiedesalgorithm，theschedulingpathisfurtheroptimized toensurefinding anaproximateoptimal solutionin polynomial time.Theexperimentalresultsshowthatthe proposed scheduling methodcannot onlyfindefectiveresourcescheduling strategiesinurban disasteremergencyresponse toimproveresourceutilizationand responsespeed，butalsooutperform themethodbasedonlocalobservationinboth motioncostandtotalcost，verifyingits advantages in emergency resource scheduling.

Keywords：emergencyresource scheduling；path optimization;machine learning;continuousapproximation model; K-means clustering algorithm;Christoffedesalgorithm

0 引言

自然灾害通常会对人们造成一定的安全威胁和财产损失，迅速向受灾区域部署必要的救济物资是救灾行动的首要任务[1-2]。（剩余5661字）