Research On Resilient Recovery Of Supply Network Based On Self-Organizing Characteristics And Outer Synchronization

With the development of economic globalization,supply chains have become supply network.At the same time to emphasize improving efficiency,the supply network is more vulnerable to the risk of disruption in the uncertain environment.Many serious disruption cases have brought the problems of supply network security into the forefront of public attention.Building resilient supply network is an effective way to deal with uncertain disruption.The purpose of supply network resilience research is to enhance the ability of supply network to resist impacts of disruptions and quickly recover from disruptions.At present,there are some researches on resilience concept,resilience measurement and resilience strategy,but there are few studies on resilience recovery.However recovery is the key to resilience.The study was carried out around the resilient recovery of the supply network.From objective resilient recovery mechanism to subjective resilient recovery control,two aspects are studied,and combining them together can improve the resilient recovery effect.This study can not only provides a new idea for the supply network resilience theory and management practice,but also provides reference for other complex systems.The main researches are as follow.(1)Resilience measurement was studied,which is the basis of management decisions.Based on the Dynamic Inoperability Input-output Model(DIIM),the resilience measurement method of supply network was established.Through the inoperability,the method describes the resilience of supply networks includin the absorption,adaption and restoration,measures the resilience from time,space,and cost,displaies visually the recovery process.A case shows that the method and the corresponding measures can effectively improve the resilience of the supply network.The method could provide theoretical guidance and reference for supply network management.(2)The internal mechanism of cascade failure and resilience recovery is studied,which is an objective factor of resilience recovery.Firstly,the generation model of the supply network structure and cascade failure model are proposed,the dynamic evolution process and the effect of network characteristics were analyzed.Then the self-organizing characteristics were studied.It is found that the resilience recovery can emerge through local self-organizing reconstruction behavior.The simulation illustrates that the root of vulnerability lies in cascading failure and self-organization is the key to resilient recovery.The study of self-organization characteristic provides theoretical guidance for local control achieves the overall resilient optimization.(3)Further studies have found that the supply network has self-organizing criticality(SOC)and can be applied to resilience management.Firstly,considering enterprises as MAS,the local fitness function,neighborhood structure and interaction rules suitable for supply network are designed.Through local information guidance,the whole resilience and SOC characteristics were found.Case studies show that the micro level of self-organizing ability is the root of resilience at the macro level.Secondly,based on the logistics distribution in the emergency environment,using the SOC characteristics,the integrated resilience management framework was proposed,which includes early prevention,medium-term mitigation and later recovery.(4)Based on outer synchronization,a resilient recovery method for supply network is proposed.Firstly,the mapping relationship between outer synchronization and supply network resilient recovery is given.The outer synchronization possesses adapt ability,absorbing ability and recovery ability,which can make the disturbed network return to the planned state.Secondly,considering enterprise node as a Lorenz system,the dynamic model of supply network with uncertain parameters,time-delay and changing structure is established.Then an impulsive pinning controller and parameter identification rules are designed and the effective range of control time delay is given.Case analysis verifies the effectiveness of the dynamic model of supply network and resilient recovery method.This method breaks a new path for the subjective control of resilient recovery of supply network.(5)The control of resilient recovery is further analyzed based on outer synchronization.Different from the previous research,a resilience measurement method with memory is proposed,and a pulse controller is designed to improve the control effect.A greedy algorithm is designed to identify the supply network backbone of resilient recovery.Then,considering the recovery effect and control cost,a measurement method of control efficiency is proposed to help control strategy.Case study is used to verify the validity of the resilient recovery method and control decision(6)Two auto supply networks,Honda Accord and Acura CL/TL were studied as a case.A resilient design scheme combining internal design and external control is proposed.In internal design,we propose a model of supply network self-organization evolution.In order to maximize the satisfaction of the enterprise,the system tends to the critical state to trigger the cascade failure.A network structural design can achieve a better absorptive capacity.In external control,a pulse control based on outer synchronization can maintain the supply network in a resilient range,allowing the system to not have a large avalanche.Through internal mechanism and external control,the balance is achieved by allowing the supply network to fluctuate in efficiency and risk.