Font Size: a A A

Research On Repair Crew Scheduling On Severely Damaged Road Network

Posted on:2021-05-25Degree:MasterType:Thesis
Country:ChinaCandidate:B H TuFull Text:PDF
GTID:2416330614460407Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
Emergency repair of damaged road network is one of the basic links of post-disaster emergency response.The completion of quick emergency repair of damaged road network can provide fast and smooth road access for subsequent material and manpower dispatch.The rapid and safe evacuation has important practical significance.However,there are few studies that have taken into account the fact that many large-scale disasters actually damage the road network.The road network model and the repair team scheduling algorithm assume that the network environment is ideal,and the repair team cannot generally be given to the road network with severe damage Scheduling scheme.Based on this,on the basis of the existing reinforcement learning model,this dissertation studies the scheduling problem of the repair team of severely damaged road network and verifies its effectiveness.The main work of the dissertation is as follows:(1)The relevant background,research status,and existing problems of the road repair problem of the damaged road network in the post-disaster rescue work are elaborated,and the research motivation and main research content of this article are determined.(2)Introduced the relevant basic theoretical knowledge related to this article,including the necessity and repair principles of the damaged road network,reinforcement learning,Q learning and other related knowledge,which provided technical support for the subsequent algorithm design.(3)Facing severely damaged road network,simplified road network model and decision model,designed a more simplified and effective agent action set,state set and corresponding rewards,and designed a severely damaged road network Single repair team scheduling algorithm.It can be verified that in a severely damaged road network environment where the scale and damage rate of the road network are relatively large,the B / L repair team scheduling algorithm can ensure that all demand nodes are reachable,with higher stability and reliability.(4)In view of the situation of multiple repair teams in the road network,a model of multiple repair teams scheduling problem is constructed,and based on the distributed computing idea,all the demand points and damaged road sections in the entire emergency scene are divided into multiple to reserve Point-centered sub-scene area,a multi-repair team scheduling algorithm for multiple reserve points is proposed.The experimental results show that the proposed multi-repair team scheduling algorithm can give a scheduling plan within the repair cost,which is more close to the actual emergency rescue.
Keywords/Search Tags:disaster emergency response, severely damaged road network, repair crew scheduling, Q learning, action set tailoring
PDF Full Text Request
Related items