Advances in Computer Science Research, volume 65

3rd International Conference on Communications, Information Management and Network Security (CIMNS 2018)

Emergency Logistics Theory, Model and Method: A

Review and Further Research Directions
Hongqian Xu, Danhui Fang and Yining Jin
China Research Center for Emergency Management, Wuhan University of Technology Wuhan Hubei 430070, China

Abstract—With the improvement of people’s understanding emergency logistics and optimize the ability of emergency
of disaster relief, emergency logistics has become an important logistics. The related theories and research models of
guarantee to improve the efficiency of relief and safeguard the emergency logistics described in this paper will mainly focus
safety of people's lives and property in recent years. In this paper, on the location, transportation and distribution of emergency
the research theories, methods and models of emergency logistics materials to rescue sites.
are classified and sorted out, and different problems such as
Location Allocation Problem, Vehicle Routing Problem,
Location-Routing Problem are classified and discussed. Under II. CHARACTERISTICS OF EMERGENCY LOGISTICS
the condition that most of the existing researches mainly focus on Emergency logistics research includes the emergency
response cost and response time, this paper argues that the allocation, transportation and distribution of materials for
primary purpose of emergency logistics should be to reduce the emergency relief in disaster areas after the occurrence of
loss of people's lives and property. Based on this, an extension sudden natural disasters and social hazards such as earthquakes,
study on emergency logistics is proposed to provide ideas for floods, hurricanes, mudslides, outbreaks, major traffic
further research in related fields.
accidents, production accidents and terrorist attacks and the
Keywords—emergency logisticst; disaster relief; methods and
delivery of personnel.
models Compared with conventional commercial logistics activities,
the cost and time of emergency logistics are contrary to each
I. BACKGROUND AND CONCEPT other, and time is more dominant in the goal. Therefore, it has
the characteristics of sudden and unpredictable, stochastic
Emergency logistics refers to the logistics activities caused
demand, urgency of time constraints, peak value, weak
by emergencies, including emergency logistics demand
economy, unconventionality, government and market
generated by emergencies and emergency logistics supply
participation. In this case, the particularity and continuity of
activities to meet these logistics needs, pursuing the purpose of
different disaster relief stages should be taken into account in
maximizing time efficiency and minimizing disaster losses.
the study of related problems and the construction of models.
Facility location and material distribution in emergency
logistics is an important part of emergency decision-making
and one of the key links of rescue operations[1]. In 2004, III. CLASSIFICATION OF EMERGENCY LOGISTICS PROBLEMS
scholars first put forward the concept of emergency logistics: Scholars generally believe that the key to effectively ensure
logistics activities for the purpose of providing materials the supply of emergency supplies lies in rational location
needed to deal with natural disasters, public health incidents, Allocation Problem (LAP) and scientific planning of the
major accidents and other emergencies[2]. According to the Vehicle Routing Problem (VRP). Moreover, there is an
definition put forward by Tomas and Fritz Institute in 2004, interdependent and interactional relationship between LAP and
humanitarian logistics is to plan, execute and control the VRP, thus it is necessary to design and optimize them as a
efficient and effective flow and storage of relief materials and whole system, that is, to study Location-Routing Problem (LRP)
related information from the place of supply to the place of in emergency logistics system.
consumption.These include preparation, planning, procurement,
transportation, warehousing, tracking and tracking, and A. Classification of LAP
customs clearance[3].
LAP problem mainly refers to the decision-maker
However, when natural disasters occur, it often causes determines the number and location of facilities in a certain
damage and damage to road facilities, and affects the efficiency geographical area according to the geographical distribution of
of emergency supplies distribution because of various obstacles. customers and the distribution of goods. It can be roughly
The loss of inefficient distribution accounts for about divided into the following eight categories. In this paper,
15%~20% of the total loss of the disaster[4]. Humanitarian typical problems and solution models are selected for analysis:
logistics for emergency rescue is a very important part of
emergency rescue operations, accounting for 80% - 90% of the
total cost of rescue, according to Fritz Research Institute[5].The
transfer speed of the rescue personnel will determine the
quality of the whole emergency rescue operation[6]. Therefore,
it is of great significance to improve the efficiency of

Copyright © 2018, the Authors. Published by Atlantis Press.

This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). 188
 Advances in Computer Science Research, volume 65

reserve points is different, aiming at the shortest total path

distance[11].
Some scholars considered the serious shortage of
emergency materials and the shortage of emergency supplies in
the early post-earthquake period when they studied the
location-allocation problem of emergency logistics. In order to
meet the different emergency needs of multi-agent participation
after the earthquake, and at the same time consider the fairness
and efficiency, from the point of view of coordinated
FIGURE I. CLASSIFICATION OF LOCATION ALLOCATION distribution of public and private logistics resources, they
PROBLEMS constructed the maximum and total satisfaction. Multi-period,
multi-objective, multi-material dynamic location-allocation
1) Site Selection and Allocation model with minimum cost and minimum personnel loss is
In the traditional location problem, there are mainly solved by genetic algorithm with the objective of maximizing
coverage problem, center problem and median problem, and the commodity satisfaction rate at the upper level and
coverage problem is divided into set coverage model, maximizing the time satisfaction rate at the lower level. Finally,
maximum coverage model, maximum available model and so the validity of the model and algorithm is verified by a case
on. These traditional models are based on the assumption that study of the Wenchuan earthquake[12].
the conditions are determined. Compared with the traditional
location problem, the location problem of emergency supplies B. Review of VRP Researches
is stochastic and uncertain. Therefore, some people introduce VRP studies that in order to meet the customer's demand
probability theory and fuzzy theory into the location of for goods, delivery time, vehicle capacity constraints, mileage
emergency facilities[7]. constraints, time constraints, a series of customer demand
There is also an article on how to determine facilities with points for the design of appropriate vehicle routes, and to meet
uncertain demand by considering the number and distance of the shortest mileage, the least cost, the shortest time, as small
services. In the case study of Los Angeles, we use chance as possible fleet size, vehicle benefits. Optimize the target with
constrained model to solve the maximum location coverage higher utilization rate.
problem[8].For capacity facility location problem with single In the aspect of distribution routing optimization, it mainly
source constraints, some scholars have proposed a VLSN analyzes the degree of road damage and road congestion by
search algorithm and proved its effectiveness[9]. As the constructing road network, and studies the emergency material
problem of uncapacitated facility location (UFL), When the scheduling problem under the condition of uncertain vehicle
goal is to find a subset of facility locations to open and to travel time.
allocate each customer to open facilities so that the facility
opening plus customer allocation costs are minimized. Scholars The Vehicle Routing Problem (VRP) was first proposed in
use (generalized) Benders cuts to replace a huge number of 1959 to study the transportation of gasoline from petrol stations
allocation variables by a small number of continuous variables to various gas stations[13]. Under the condition of limited time,
that model the customer allocation cost directly[10]. limited quantity and quality of resources, the optimal model for
allocating and transporting resources to multiple disaster sites
2) Research on Dispatch and Allocation after an earthquake is studied with the objective of minimizing
Emergency materials dispatching and distribution is the the number of deaths[14]. Some scholars classify the
management of emergency materials after the outbreak of a distribution of rescue materials as a network flow problem with
sudden disaster event. It is usually necessary to transport all multiple modes of transportation and multi-material categories,
kinds of materials from different locations to several and construct a dynamic distribution model with time
emergency materials distribution centers and disaster areas in constraints. The objective function is to minimize the amount
the shortest time. Material allocation is the next phase of of emergency materials that do not meet the demand[15].
material dispatching, but in many cases the scheduling and Considering the Vehicle Routing Problem (VRP), some
allocation occur simultaneously, so many articles will study the scholars have studied the sub-path ejection chain method under
two together. the limit of capacity and path length[16]. A stochastic vehicle
routing model is proposed to solve the evacuation problem of
When some essays study the problem of locating flood victims[17]
emergency resource reserve sites in cities and towns, taking the
rainstorm disaster as the background and Hongshan District of C. Studies on LRP
Wuhan as an example, eight suitable emergency resource
reserve nodes in the study area are selected from the six basic Researchers on Location-Routing Problems (LRP) mainly
conditions of locating emergency resource reserve sites under focus on theoretical model and algorithm design. In the actual
the influence of the determined rainstorm disaster, and then operation of emergency logistics system, the location of
Freud is used. German algorithm calculates the distribution distribution center and the choice of distribution route are
paths with reserve points, and obtains the relationship between interdependent, mutually reinforcing. The location of
the positions of the eight reserve points when the number of distribution center will affect the planning of distribution route,
and the optimization of distribution route will affect the

189
 Advances in Computer Science Research, volume 65

location of distribution center in turn. Therefore, it is necessary reflects the timeliness, the maximum rescue radius reflects the
to consider the distribution of emergency logistics from the equilibrium, and the additional rescue change time under the
overall point of view from the perspective of the post-disaster situation of facility damage reflects robustness. For rod
emergency logistics distribution system. property, a target disturbance minimization model based on
ideal point is proposed, and a genetic algorithm is used to
At present, the interpretation models of LRP mainly include design and solve the model.
facility location decision model [18], robust programming [19],
two-stage random mixed integer programming model [20], etc. Further more, study on the real-time decision-making
In addition, some scholars have studied the LRP model with method of disturbance events in the location-routing problem
uncertain parameters. Such as introducing randomness into the of emergency materials distribution in the early post-
model, transformed stochastic variable demand into earthquake period. Considering the damage of road network,
deterministic demand, and constructed a LRP model with fuzzy demand of demand points and multi-mode distribution,
multiple supply points and multiple vehicle Location-Routing the LRP optimization model of post-earthquake emergency
for single objective optimization cost[21]. Many others studied logistics is constructed with the shortest total distribution time
the integrated optimization model of multi-mode, multi-stage, of emergency materials. Two kinds of disturbance events, i.e.
multi-objective emergency material allocation and multi-mode the serious damage of the original unblocked road caused by
transportation under fuzzy conditions, and realized the dynamic aftershocks, the impassability of vehicles and the failure of
location of emergency facilities and the optimal dispatch of vehicles requiring maintenance, were tested.
emergency materials through multi-objective dynamic
weighting[22]. B. Transportation of the Wounded
Choosing an objective function which is close to the
IV. OTHER STUDIES ON EMERGENCY LOGISTICS interests of the wounded and sick to evaluate the effect of
emergency rescue effectively is the research direction of many
A. Disruption Management of Logistic scholars. According to the theory of timeliness of rescue in
In the aspect of disruption management, there is little emergency medicine, the deterioration of severely injured
research on the disruption management of post-disaster LRP. patients mostly occurs in a very short time after injury. It is a
Scholars often carry out modeling and verification under ideal key to reduce the mortality of severely injured patients by
hypothesis. In fact, this research will help to improve the making them receive treatment in the "golden first aid time" as
efficiency of post-disaster rescue greatly. far as possible. The value of first aid. Because of the proximity
of the clusters of heavy and light casualties in the spatial
In addition to multi-objective interference, LRP is often distribution, it is necessary to make a comprehensive plan for
confronted with many uncertain factors in the process of the treatment and transfer of the heavy and light casualties in
emergency materials distribution after disasters. For example, the formulation of the ambulance treatment and transfer vehicle
road damage, sudden large-scale outbreak of the epidemic led routing planning.
to a sharp increase in demand for emergency supplies at some
demand points, vehicles in the course of service failure can not In the study of emergency logistics, in order to achieve a
continue to distribute, demand points require early delivery of large-scale natural and accident disaster casualties after the
emergency supplies. Such incidents are collectively referred to rapid treatment and placement, according to the casualties can
as interference events. At this point, if the decision-maker be classified into high-efficiency casualty treatment and
supplies emergency materials according to the initial plan made transfer rescue model. By introducing the theory of time-
before the interference incident, it will inevitably lead to the limited treatment of the wounded into the trauma index to
failure of some demand point services. calculate the changes of the severity of the wounded during the
waiting period, and taking the minimization of the total trauma
Therefore, it is necessary to adjust the plan according to the index of the wounded as the optimization objective, the priority
specific situation of the disturbance events, effectively deal levels of the severity of the wounded during the early and late
with the disturbance events, so as to minimize the disturbance rescue periods are dynamically adjusted, and the routes of
to the emergency logistics system after the earthquake, and ambulance treatment and transfer of the wounded are planned
ensure the normal distribution of emergency supplies. as a whole, so as to make the wounded and so on. The severity
In disruption management, the problem of site selection for of the injury was reduced to the minimum during the period.
facility damage scenarios was first proposed in 1987. Some Some scholars have established a multi-distance, demand-
researches analyzed four causes of facility failure, and then separable ambulance VRP mathematical model, and designed a
studied the p-center model and p-median model for facility distance-based neighborhood search model fast solution
damage scenarios[23]. Some others considered the different algorithm, more in line with the large-scale emergency time
failure probability of facilities in different locations, and ambulance between the hospital and the casualty cluster to
introduced the failure of facilities, established a supply network carry out rapid rescue requirements, and finally validated by
considering both reliability and economic benefits, and numerical simulation and comparative study. The effectiveness
proposed two unrestricted reliability location models[24]. of the model and algorithm[27].
Some essays considers possible damage scenarios in the
process of emergency facility location optimization, and
establishes a multi-objective optimization model for emergency
facility location[25]. The total rescue time covering the target

190
 Advances in Computer Science Research, volume 65

V. SUMMARY OF SOLUTIONS AND MODELS TO 3) Other Theories and Model Solving Methods
EMERGENCY LOGISTICS PROBLEMS In addition to the network flow model commonly used in
At present, the research structure of many papers on material transportation and distribution, there are also Bayesian
emergency logistics is mainly problem description - problem decision theory, feedback control theory, staged decision-
hypothesis - model construction (selection of objectives, making, credibility model, scenario planning method, genetic
constraints) - planning solution / algorithm design - Example algorithm and so on.
Analysis - effect evaluation (validity / rationality validation) For example, a multi-objective open LRP optimization
model is constructed to minimize the total arrival time, the total
A. Forecasting Method of Material Demand in Emergency cost and the reliability of emergency supplies, and a non-
Situation dominated sorting genetic algorithm and a non-dominated
 The demand for emergency supplies can be forecasted sorting differential evolution algorithm are designed to solve
by a series of data consisting of pre-demand, such as the model[33]. Or in the theory of stage-wise decision-making,
grey system model forecasting method; a three-stage mixed-integer stochastic programming LRP
model is constructed, which considers the uncertainties of
 Analogy Principle: Emergency supplies demand can be demand, the limitation of vehicle capacity and the condition of
predicted by looking for similar cases, the core idea is emergency logistics network in the distribution of emergency
to use the experience of previous similar problems and materials. The first stage solves the problem of locating
acquired knowledge to infer the current problem logistics center, and the second stage carries out the vehicle of
solution. For example, case-based reasoning, BP neural logistics center. The third stage is to deal with the uncertainty
network; of emergency logistics network[34].
 Relevant Principle: Emergency supplies demand is
affected by many factors, based on the linear
hypothesis between demand and factors to do VI. CORRELATION EXTENSION STUDY
correlation analysis. For example, multiple regression Through the review and Research on the field of emergency
model prediction method and indirect prediction logistics, there are still some deficiencies in the related
method. extended research from the following perspective. Scholars can
expand their research accordingly.
B. Dispatching and Allocation Solutions
1) Considering the perceived fairness of emergency
1) Research on Optimization Theory based on Operational logistics from the perspective of disaster victims' psychology.
Research
From the point of view of optimization, it can be divided 2) Considering the process of post-disaster material
into single objective optimization and multi-objective distribution as a process of game theory, using game theory to
optimization. The objectives include the minimum solve the problem of fairness of post-disaster distribution.
transportation cost, the shortest emergency time, the least 3) At present, most of the articles are carried out from the
unmet quantity, the least number of deaths, the least perspective of decision makers. Post-disaster relief has the
transportation time and so on. attribute of multi-agent participation. How to think and discuss
For example, in order to minimize the total cost and from the perspective of other subjects (disaster victims,
transport risk, scholars have established a multi-objective LRP logistics executives).
model, which solves the problem of waste recycling station 4) From the perspective of disaster chain, we should
location and arranges the routes of different types of waste to consider the inventory management of distribution centers or
these different recycling stations[28]. For LRP, a single distribution centers.
objective bi level satisfaction model[31] is constructed under
the conditions of minimizing transportation cost[29], 5) Considering the reverse logistics in emergency logistics.
minimizing transportation time[30] or time satisfaction[31].
The single objective LRP problem neglects the fixed cost of REFERENCES
facilities when calculating the cost, which is different from the [1] Guo Zixue,Guo Liang,Zhang Pei,Yang Xiaohui, “Fuzzy optimization
actual operation. For this reason, some scholars have studied model for minimizing emergency material dispatch time”. China Safety
the multi-objective LRP model. Science Journal, vol 25, no.10, pp.172-176, 2015.
[2] Ou Zhongwen, Wang Huiyun, “Emergency Logistic,” Journal of
2) Combinatorial Mathematics Method Chongqing University, 2004, pp.164-167.
When studied the LRP of emergency materials distribution [3] Tomas. A, “Leveraging private expertise for humanitarian supply
after disaster, a multi-objective LRP decision-making model chains,” Forced migration Review, vol. 21, pp. 64-65, 2004.
was constructed with the objective of minimizing warehouse [4] Gao Hongni, Zhao Yibing, Li Ning. “Study on earthquake disaster
construction cost and operation cost and maximizing demand emergency supplies scheduling model based on multiple demand
coverage, and a mathematical heuristic algorithm was designed points,” China Safety Science Journal, vol.23, no.01, pp.161-165, 2013.
to solve it[32]. [5] Taylor, D, Pettit, S, “A consideration of the relevance of lean supply
chain concepts for humanitarian aid provision,” International Journal of
Services, Technology and Management, vol.12, no.4, pp. 430-444, 2009.

191
 Advances in Computer Science Research, volume 65

[6] CHAIKEN, J.M. and LARSON, R.C., 1972. “METHODS FOR [26] LIU Changshi, ZHU Zheng, LIU Liyong. “Disruption management of
ALLOCATING URBAN EMERGENCY UNITS: A SURVEY,” Location-Routing Problem (LRP) for emergency logistics system in
Management Science (pre-1986), vol.19, no.4, pp. 21. early stage after earthquake,” Computer Engineering and Applications,
[7] Chen Gang, Zhang Jin, Fu Jiangyue, “A multi-objective location- 2017, vol.53, no.20, pp.224-230.
allocation model for emergency logistics in an uncertain information [27] WANG Jing;LIU Hao-tian;Huang Jun, “Research on the Route
environmen,” China Safety Science Journal, vol.26, no.12, pp.163-168, Optimization of Ambulance Treatment and Transportation after Disaster
2016. Based on the Injured Classification,” Chinese Journal of Management
[8] Murali P, Ordóñez F, Dessouky MM, “Facility location under demand Science, Vol.25, No.8, pp.114-122,2017.
uncertainty: Response to a large-scale bio-terror attack,” Socio- [28] Alumur S, Kara BY. “A new model for the hazardous waste location-
Economic Planning Sciences, vol.46, no.1, pp.78-87, 2012. routing problem. Computers and Operations Research,” vol.34, no.5. pp.
[9] Ahuja R, Orlin J, Pallottino S, Scaparra M, Scutellà M, “A Multi- 1406-23, 2007.
Exchange Heuristic for the Single-Source Capacitated Facility Location [29] EQUI L, GALLO G, MARZIALE S. “A combined transportation and
Problem,” Management Science [serial online], vol.50, no.6, pp.749-760, scheduling problem,” European Journal of Operational Research, vol.97,
2004. no.1, pp.94-104, 1996.
[10] Fischetti, Matteo, “Redesigning Benders Decomposition for Large-Scale [30] YU Wuyang, “Transshipment model for emergency materials based on
Facility Location,” Management Science, vol.63, no.7, pp. 2146-2163, time satisfaction criterion,” Journal of Systems &Management, 2013,
2017. vol.22, no.5, pp.882-887.
[11] WU Ke, SONG Yinghua, LYU Wei, “Research on siting of urban [31] KONGSOMSAKSAKUL S, YANG C, CHEN A. “Shelter location-
emergency resources depots and layout optimization considering allocation model for flood evacuation planning,” Journal of the Eastern
rainstorm disaster and distribution route,” China Safety Science Journal, Asia Society for Transportation Studies,2005,vol.6,pp.4237-4252.
Vol.27, No.9, pp. 170-174, 2017. [32] Rath S, Gutjahr W J. “A math- heuristic for the ware-house location-
[12] SONG Yinghua, NING Jingjing, LYU Wei, DU Lijing, “Research on routing problem in disaster relief,” Computers & Operations Research,
dynamic LAP model of emergency logistics public-private collaboration vol.42, pp.25-39, 2014.
in aftermath of earthquake,” China Safety Science Journal, Vol.28, No.3, [33] Wang H. “Multi-objective open location-routing model with split
pp. 173-178, 2018. delivery for optimized relief distribution inpost-earthquake,”
[13] Dantizig G, Ramser J. “The Truck Dispatching Problem,” Management Transportation Research Part E: Logistics and Transportation Review,
Science, 1959, vol.6, pp.80-91. 2014, vol.69, pp.160-179.
[14] Fiedrich, Gehbauer, Rickers. “Optimized resource allocation for [34] Rennemo S J, Ro K F, Hvattum L Ml. “A three-stage stochastic facility
emergency response after earthquake disasters,” Safety Science, 2000, routing model for disaster response planning,” Transportation Research
vol, 35, no.1, pp. 41-57. Part E:Logistics and Transportation Review, 2014,vol.62,pp.116-135.
[15] Linet Odamar, Ediz Ekinci, Beste Kucukyazici. The Dynamics of
Emergency Transshipment Supply Chains. Annals of Operations
Research, 2004, pp.217-245.
[16] Rego C. “A Subpath Ejection Method for the Vehicle Routing Problem.
Management Science [serial online],” October 1998, vol.44, no.10,
pp.1447-1459. Available from: IT Source, Ipswich, MA. Accessed June
3, 2018.
[17] DESSOUKY M, ORDONEZ F, JIA H. “Rapid distribution of medical
supplies,” Patient Flow: Reducing Delay in Healthcare Delivery, 2013,
pp.385-410.
[18] CHEN Gang, ZHANG Jin, FU Jiangyue. “A multi-objective location-
allocation model for emergency logistics in an uncertain information
environment,” China Safety Science Journal, vol.26, no.12, pp.163-168,
2016.
[19] WANG Jing, ZHANG Ling, HUANG Jun. “The robust emergency
logistics system based on uncertain demand,” Mathematics in Practice
and Theory, vol. 39, no.20, pp. 53-60, 2009.
[20] RAW LS C G, TUMQUIST M A. “Pre-positioning of emergency
supplies for disaster response,” Transportation Research Part B:
Methodological, vol. 44, no.4 pp. 521-534, 2010.
[21] Chan Y, Baker SF. “The multiple depot, multiple traveling salesmen
facility-location problem: Vehicle range, service frequency, and heuristic
implementations,” Mathematical and Computer Modelling. Vol.41, no.8,
pp.1035-53, 2005.
[22] Song yinghua, Wang lifag. “Location-multimodal transportation
problem for relief distribution in emergency logistics under fuzzy
conditions,” China Safety Science Journal, vol.27, no.7, pp.169-174,
2017.
[23] Drezner, Zvi. “Heuristic Solution Methods for Two Location Problems
with Unreliable Facilities,” The Journal of the Operational Research
Society, vol.38, no. 6 (1987), pp. 509-514.
[24] Cui, Tingting, Yanfeng Ouyang, and Zuo-Jun Max Shen. “Reliable
Facility Location Design Under the Risk of Disruptions,” Operations
Research, vol. 58, no. 4 (2010), pp. 998.
[25] Zhu Jianming. “Methods of multi-objective decision-making for
emergency facility location problem under failure scenario,” Systems
engineering-Theory & Practice, 2015, vol.35, no.3, pp.720-727.

192