CN110490503A - A kind of logistics delivery vehicle scheduling method based on mass data - Google Patents
A kind of logistics delivery vehicle scheduling method based on mass data Download PDFInfo
- Publication number
- CN110490503A CN110490503A CN201910461469.8A CN201910461469A CN110490503A CN 110490503 A CN110490503 A CN 110490503A CN 201910461469 A CN201910461469 A CN 201910461469A CN 110490503 A CN110490503 A CN 110490503A
- Authority
- CN
- China
- Prior art keywords
- module
- data
- client
- database
- logistics
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000009826 distribution Methods 0.000 claims abstract description 42
- 238000005516 engineering process Methods 0.000 claims abstract description 14
- 230000003993 interaction Effects 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims description 18
- 230000011218 segmentation Effects 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000013459 approach Methods 0.000 claims description 8
- 238000013499 data model Methods 0.000 claims description 7
- 230000002068 genetic effect Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000013500 data storage Methods 0.000 claims description 5
- 230000008676 import Effects 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 4
- 230000007547 defect Effects 0.000 claims description 4
- 230000002452 interceptive effect Effects 0.000 claims description 4
- 108090000623 proteins and genes Proteins 0.000 claims description 4
- 230000007704 transition Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 238000013461 design Methods 0.000 claims 2
- 238000010586 diagram Methods 0.000 description 7
- 210000000349 chromosome Anatomy 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 101100070542 Podospora anserina het-s gene Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 210000001726 chromosome structure Anatomy 0.000 description 1
- 238000004043 dyeing Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000010353 genetic engineering Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0835—Relationships between shipper or supplier and carriers
- G06Q10/08355—Routing methods
Abstract
The present invention relates to logistics technology, specially a kind of logistics delivery vehicle scheduling method based on mass data, the client including database and with database realizing data interaction.In the logistics delivery vehicle scheduling method based on mass data, vehicle and logistics distribution data are uploaded to database by client, logistics private clound large data center is built with cloud computing technology, using virtualization technology, the physical resource data of dispatching are carried out pond, form total Database, logistics distribution information command is assigned by client, the logistics distribution information command assigned is uploaded in database by client, the vehicle and logistics distribution data in distribution information combined data library assigned according to client, reasonable vehicle tune information is generated by vehicle scheduling module, realize Rational Path transport, to effectively reduce transportation cost, save haulage time, it increases economic efficiency.
Description
Technical field
The present invention relates to logistics technology, specially a kind of logistics delivery vehicle scheduling side based on mass data
Method.
Background technique
All multiple features possessed by modern logistics so that loglstics enterprise seeks new technique actively to construct the object of modernization
Stream information platform improves efficiency of service, increases the level of profitability of enterprise, and a possibility that generation information technology practices is
Loglstics enterprise provides new opportunity.For the goods delivery of logistics center and Third-party Logistics Enterprise, the scheduling of haulage vehicle
It is the emphasis of work, the scheduling to hold water can effectively reduce the rate of empty ride of vehicle, Rational Path transport be realized, thus effectively
Transportation cost is reduced, haulage time is saved, increases economic efficiency.In consideration of it, it is proposed that a kind of logistics based on mass data
Optimized scheduling of distribution vehicles method.
Summary of the invention
The logistics delivery vehicle scheduling method based on mass data that the purpose of the present invention is to provide a kind of, for having
Effect reduces the rate of empty ride of vehicle, realizes Rational Path transport, to effectively reduce transportation cost, saves haulage time, improves warp
Ji benefit.
To achieve the above object, the present invention provides a kind of logistics delivery vehicle scheduling method based on mass data,
The following steps are included:
S1, it establishes database: vehicle and logistics distribution data being uploaded to database by client, with cloud computing technology
Logistics private clound large data center is built, using virtualization technology, the physical resource data of dispatching is carried out pond, form sum
According to library；
S2, service display: database carries out the system resource of logistics distribution, data resource, information resources uniformly, and with
The form of thumbnail is shown, while the information data in database being synchronized in client；
S3, task are assigned: user of service logs in client and assigns logistics distribution information command, the object that client will be assigned
Stream distribution information instruction is uploaded in database；
S4, vehicle scheduling: the vehicle and logistics distribution data in distribution information combined data library assigned according to client,
Reasonable vehicle tune information is generated, and feeds back to client.
Preferably, the database includes data segmentation module, data storage module, data modeling module and foreground exhibition
Show that module, the data segmentation module are used to do Hash segmentation by field for the physical resource data in number library, forms multiple sons
Database, for saving to the data after segmentation, the data modeling module is used for described the data storage module
Database creates model and generates .edmx file in Models catalogue, and .edmx file is an XML file, it is general for defining
The mapping between model, storage model and these models is read .edmx file also includes ADO.NET entity data model designer,
For the information of graphically present model, the foreground display module is used for the database and each client
Between realize data interaction, upload by the client and data and assign instruction.
Preferably, the data modeling module includes data import modul and service view module, the data are imported
For module for the data in the database to be directed into data model, the service view module is used for the data that will be imported
Information is converted to view and is shown.
Preferably, the foreground display module includes user interactive module and vehicle scheduling module, user's interaction
Each client is connected by module with the database, so that generating number between the database and each client
According to interaction, the vehicle scheduling module is used for the distribution information assigned according to the client and in conjunction with the object in the database
Stream information generates reasonable vehicle tune information.
Preferably, the vehicle scheduling module include path planning module, freight charges planning module, constraint processing module,
Genetic operator module and scheme determining module, the path planning module is for planning optimal Distribution path, the freight charges rule
Module is drawn for planning optimal dispatching freight charges, the constraint processing module is used to guarantee the correctness of vehicle scheduling optimization, institute
Genetic operator module is stated for retaining defect individual, gene delection is avoided, improves global convergence and efficiency, the scheme determines
Module is for generating optimal transportation dispatching scheme.
Preferably, the vehicle scheduling module is designed based on neural network algorithm, the neural network algorithm includes neighbour
It connects matrix module, schedule constraints module, neural network module and scheme and forms module, the adjacency matrix module is used for vehicle
Source point, each meeting point for being passed through and rest point be abstracted into the node of network, the schedule constraints module is for guaranteeing vehicle tune
The correctness of optimization is spent, the neural network module is used to determine the transmission function and state transition equation of neuron, by net
The evolution repeatedly of network, until convergence, the scheme forms module for generating optimal transportation dispatching scheme.
Preferably, the vehicle scheduling module is designed based on heuritic approach, the heuritic approach includes establishing mould
Pattern block and solving model module, the model module of establishing is for establishing model, the solution mould for the distribution information of vehicle
Pattern block is used for solving model, and obtains optimal transportation dispatching scheme.
Preferably, the client includes user log-in block and operation module, the user log-in block is for making
User logs in the client, and the operation module is used to realize by the client and manipulate.
Preferably, the operation module includes data uploading module and task assigns module, the data uploading module
For uploading vehicle and logistics distribution data to the database by the client, the task assigns module for passing through
The client assigns logistics distribution information to the database.
Compared with prior art, beneficial effects of the present invention: should the logistics delivery vehicle scheduling based on mass data
In method, vehicle and logistics distribution data are uploaded to database by client, build logistics private clound with cloud computing technology
The physical resource data of dispatching are carried out pond, are formed total Database, pass through client by large data center using virtualization technology
Logistics distribution information command is assigned at end, and the logistics distribution information command assigned is uploaded in database by client, according to client
The vehicle and logistics distribution data in the distribution information combined data library assigned are held, reasonable vehicle is generated by vehicle scheduling module
Information is adjusted, realizes Rational Path transport, to effectively reduce transportation cost, save haulage time, increase economic efficiency.
Detailed description of the invention
Fig. 1 is overall structure diagram of the invention；
Fig. 2 is database integral module figure of the invention；
Fig. 3 is data modeling module diagram of the invention；
Fig. 4 is foreground display module schematic diagram of the invention；
Fig. 5 is vehicle scheduling module diagram of the invention；
Fig. 6 is neural network algorithm schematic diagram of the invention；
Fig. 7 is heuritic approach schematic diagram of the invention；
Fig. 8 is client modules figure of the invention；
Fig. 9 is operation module schematic diagram of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1
The present invention provides a kind of logistics delivery vehicle scheduling method based on mass data, comprising the following steps:
S1, it establishes database: vehicle and logistics distribution data being uploaded to database by client, with cloud computing technology
Logistics private clound large data center is built, using virtualization technology, the physical resource data of dispatching is carried out pond, form sum
According to library；
S2, service display: database carries out the system resource of logistics distribution, data resource, information resources uniformly, and with
The form of thumbnail is shown, while the information data in database being synchronized in client；
S3, task are assigned: user of service logs in client and assigns logistics distribution information command, the object that client will be assigned
Stream distribution information instruction is uploaded in database；
S4, vehicle scheduling: the vehicle and logistics distribution data in distribution information combined data library assigned according to client,
Reasonable vehicle tune information is generated, and feeds back to client.
It is as follows that database code is established in the present embodiment, in step S1:
#include<iostream>
#include<cstdlib>
#include<cstdio>
#include<cmath>
#include<cstring>
#include<iomanip>
#include<algorithm>
#include<ctime>
#include<queue>
#definergregister
#definelstlonglong
#defineN150
#defineM1200
usingnamespacestd；
intday,n,K,m,d,cnt；
structedge{
intto,v,nxt；
}ljl[M<<1]；
inthead[N],team[N],dis[N],dp[N]；
intw[N][N],bre[N][N]；
boolvis[N],b[N]；
inlineintread()
{
Rgints=0, m=1；Rgcharch=getchar ()；
while(ch！='-' && (ch<' 0'| | ch>' 9')) ch=getchar ()；
If (ch=='-') m=-1, ch=getchar ()；
While (ch>=' 0'&&ch≤' 9') s=(s<<3)+(s<<1)+ch-'0', ch=getchar ()；
returns*m；
}
inlinevoidadd(rgintp,rgintq,rginto)
{
Ljl [++ cnt] .to=q；
Ljl [cnt] .v=o；
Ljl [cnt] .nxt=head [p]；
Head [p]=cnt；
}
inlinevoidSPFA()
{
memset(vis,0,sizeof(vis))；
memset(team,0,sizeof(team))；
For (rginti=1；I≤n；++ i) dis [i]=9999999；
Rginttop=0, tail=1；
Team [1]=1；Vis [1]=1；Dis [1]=0；
while(top<tail)
{
top++；
Rgintnow=team [top]；Vis [now]=0；
For (rginti=head [now]；i；I=ljl [i] .nxt)
{
Rgintqw=ljl [i] .to；
if(dis[qw]>dis[now]+ljl[i].v&&！b[qw])
{
Dis [qw]=dis [now]+ljl [i] .v；
if(！vis[qw])
{
Team [++ tail]=qw；
Vis [qw]=1；
}
}
}
}
}
intmain()
{
Day=read (), n=read (), K=read (), m=read ()；
For (rginti=1；I≤m；++i)
{
Rgintp=read (), q=read (), o=read ()；
add(p,q,o),add(q,p,o)；
}
D=read ()；
For (rginti=1；I≤d；++i)
{
Rgintp=read (), a=read (), b=read ()；
For (rgintj=a；J≤b；++j)
Bre [j] [++ bre [j] [0]]=p；
}
For (rginti=1；I≤day；++i)
{
For (rgintj=i；J≤day；++j)
{
memset(b,0,sizeof(b))；
For (rgintk=i；K≤j；++k)
{
For (rgintl=1；L≤bre [k] [0]；l++)
{
B [bre [k] [l]]=1；
}
}
SPFA()；
if(dis[n]！=9999999) w [i] [j]=(j-i+1) * dis [n]；
Elsew [i] [j]=dis [n]；
}
}
For (rginti=1；I≤day；++i)
{
Dp [i]=w [1] [i]；
For (rgintj=1；j<i；++j)
{
Dp [i]=min (dp [i], dp [j]+K+w [j+1] [i])；
}
}
cout<<dp[day]<<endl；
return0；
}
Embodiment 2
As second of embodiment of the invention, for the ease of the data in database are integrated and classified, this hair
Bright personnel make improvements database, and as a kind of preferred embodiment, as Figure 1-Figure 4, database includes data segmentation mould
Block, data storage module, data modeling module and foreground display module, data segmentation module are used for as the physical resource in number library
Data do Hash segmentation by field, form multiple subdata bases, and data storage module is used to save the data after segmentation,
Data modeling module is used to create database on model and generates .edmx file in Models catalogue, and .edmx file is one
XML file, it also includes for the mapping between defined notion model, storage model and these models .edmx file
ADO.NET entity data model designer, for the information of graphically present model, foreground display module is used for data
Data interaction is realized between library and each client, carries out uploading data by client and assigns instruction, data modeling module
Including data import modul and service view module, data import modul is used to the data in database being directed into data model
In, service view module shows that foreground display module includes user's friendship for the data information of importing to be converted to view
Each client is connected by mutual module and vehicle scheduling module, user interactive module with database, so that database and each visitor
Data interaction is generated between the end of family, in the distribution information that vehicle scheduling module is used to assign according to client and combined data library
Logistics information generates reasonable vehicle tune information.
In the present embodiment .edmx file is exactly entity data model in data modeling module, is repaired using EF model engineer
When changing model .edmx file can be changed, under default situations .edmx file is opened using EF model engineer, can also be according to
The following steps open .edmx file using xml editor, the specific steps are as follows:
1. ensuring that project is opened in VisualStudio；
2. the right click .edmx file in " solution Explorer " then selects " opening side
Formula ... "；
2. selecting " xml editor ", then click OK.
Further, data segmentation module program operation code is as follows:
mporttensorflowastf
importscipy.ioassio
importnumpyasnp
defget_Batch(data,label,batch_size):
print(data.shape,label.shape)
Input_queue=tf.train.slice_input_producer ([data, label], num_epochs=
1, shuffle=True, capacity=32)
X_batch, y_batch=tf.train.batch (input_queue, batch_size=batch_size,
Num_threads=1, capacity=32, allow_smaller_final_batch=False)
returnx_batch,y_batch
Data=sio.loadmat (' data.mat')
Train_x=data [' train_x']
Train_y=data [' train_y']
Test_x=data [' test_x']
Test_y=data [' test_y']
X=tf.placeholder (tf.float32, [None, 10])
Y=tf.placeholder (tf.float32, [None, 2])
W=tf.Variable (tf.truncated_normal ([10,2], stddev=0.1))
B=tf.Variable (tf.truncated_normal ([2], stddev=0.1))
Pred=tf.nn.softmax (tf.matmul (x, w)+b)
Loss=tf.reduce_mean (- tf.reduce_sum (y*tf.log (pred), reduction_indices
=[1]))
Optimizer=tf.train.AdamOptimizer (2e-5) .minimize (loss)
Correct_prediction=tf.equal (tf.argmax (y, 1), tf.argmax (pred, 1))
Accuracy=tf.reduce_mean (tf.cast (correct_prediction, tf.float32), name
=' evaluation')
X_batch, y_batch=get_Batch (train_x, train_y, 1000)
# training
withtf.Session()assess:
# initiation parameter
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
# opens coordinator
Coord=tf.train.Coordinator ()
# is filled using start_queue_runners starting queue
Threads=tf.train.start_queue_runners (sess, coord)
Epoch=0
try:
whilenotcoord.should_stop():
# obtains batch_size sample and label in each batch of training
Data, label=sess.run ([x_batch, y_batch])
Sess.run (optimizer, feed_dict={ x:data, y:label })
Train_accuracy=accuracy.eval ({ x:data, y:label })
Test_accuracy=accuracy.eval ({ x:test_x, y:test_y })
Print (" Epoch%d, Trainingaccuracy%g, Testingaccuracy%g " % (epoch,
train_accuracy,test_accuracy))
Epoch=epoch+1
Excepttf.errors.OutOfRangeError:#num_epochs times number is finished this exception that can dish out
print("---Trainend---")
finally:
# coordinator coord issues all Thread Termination signals
coord.request_stop()
print('---Programmend---')
Main thread is added in the thread of unlatching by coord.join (threads) #, and threads is waited to terminate
Further, tensor_list is input, and format is the list of tensor；Generally [data, label], i.e., by
The data set of feature and label composition.
In addition, num_epochs this be that you extract the number of batch, if nothing will be extracted without given value
Batch for several times will report the mistake of OutOfRange if given value after reaching number.
In addition to this, whether shuffle is to upset at random, is to extract in order if it is False, batch；If it is
True, batch are to randomly select.
Embodiment 3
As the third embodiment of the invention, for the ease of optimizing scheduling to vehicle, vehicle is arranged in the present invention staff
Scheduler module, as a kind of preferred embodiment, as shown in figure 5, vehicle scheduling module includes path planning module, freight charges planning
Module, constraint processing module, genetic operator module and scheme determining module, path planning module is for planning optimal dispatching road
Diameter, for freight charges planning module for planning optimal dispatching freight charges, constraint processing module is used to guarantee the correct of vehicle scheduling optimization
Property, genetic operator module avoids gene delection, improves global convergence and efficiency, scheme determine mould for retaining defect individual
Block is for generating optimal transportation dispatching scheme.
In the present embodiment, path planning module using natural number to can row line encode, as length be l+m dyeing
Body is writeable are as follows:
(0,i11,i12,…,i1s,0,i21,…,i2t,0,…,0,im1,…,imn)
Wherein, ikjIndicate i-thkjItem task, such chromosome structure can be regarded as vehicle from parking lot 0, by appointing
Be engaged in i11, i12..., i1sAfter return to parking lot 0, form subpath 1；Then again from parking lot 0, by task i21..., i2tAfter return
Turn-round forms path 2, repeatedly, until all m item tasks are all completed.I is exchanged in subpath 111With
i12Position indicate walking path change, also change function target, in this way, following heredity, which iterates, can make function target most
It is small, namely it is intended to best or preferable path.
Further, freight charges planning module function is
In formula, CijFor from source point i to the unit costs of meeting point j each car, XijIt is per tour fully loaded from source point i to meeting point j
The quantity of vehicle, m, n are the number of source point and meeting point.
In addition, constraint processing module handles constraint using the method for punishment, if the corresponding solution of a chromosome violates
Some constraint, gives certain punishment according to its violation degree, makes it have lesser fitness value, do not losing group in this way
On the basis of body number, with the progress iterated, keep the number of infeasible solution proportion in group smaller and smaller, feasible solution
Number then gradually increase, and tend to optimal solution.
In addition to this, genetic operator module includes duplication, intersects, variation.The purpose of duplication operator is to retain defect individual,
Gene delection is avoided, global convergence and efficiency are improved, effect is to be combined into new individual, is carried out in chromosome space effective
Search, while the failure probability to effective model is reduced, when chromosome uses natural number coding, crossover operator generally has part
With intersection, sequence crossover, circle intersects etc., when chromosome uses binary coding, frequently with crossover operator have single point crossing, it is double
Point intersection etc., the crossing-over rate used in crossover operator is generally between 0.75-0.95.
It is worth noting that scheme determining module generates the chromosome string of best performance, root by above-mentioned genetic manipulation
The string is decoded into optimal scheduling scheme according to initial coding regulation.
Embodiment 4
As the 4th kind of embodiment of the invention, vehicle scheduling module can also select neural network algorithm to be realized,
As a kind of preferred embodiment, as shown in fig. 6, vehicle scheduling module is designed based on neural network algorithm, neural network algorithm packet
It includes adjacency matrix module, schedule constraints module, neural network module and scheme and forms module, adjacency matrix module is used for vehicle
Source point, each meeting point for being passed through and rest point be abstracted into the node of network, schedule constraints module is for guaranteeing that vehicle scheduling is excellent
The correctness of change, neural network module are used to determine the transmission function and state transition equation of neuron, repeatedly by network
Develop, until convergence, scheme forms module for generating optimal transportation dispatching scheme.
In the present embodiment, neural network algorithm is solved using Hopfield network and self-organizing feature map neural network
The Optimal Scheduling of vehicle, in Hopfield network, system can be from original state, by the transfer of a series of state
Equilibrium state is gradually converged on, this equilibrium state is local minimum point.
Specifically, the source point of vehicle, each meeting point passed through and rest point are abstracted into the knot of network by adjacency matrix module
Point, the directed walk between them are abstracted into the side of network, thus constitute a digraph G=(N, L, D), and wherein N indicates knot
Points, L indicate number of edges, and D is the matrix of N × N, referred to as adjacency matrix, if there are path, adjacency matrix phases between two nodes
Answer the value of element for the length in path；If path is not present between two nodes, the value of adjacency matrix respective element is ∞.
Further, schedule constraints module is handled as an energy term of neural network, is applied one
It is added to after penalty term in the energy equation of network, in this way with the convergence of network, the energy of constraint also gradually tends to stable state,
Embody constraint.
In addition, neural network module calculation are as follows: it sets each element in adjacency matrix and corresponds to a neuron, it is fixed
Adopted position is Vxi in the output of the neuron of position (x, i).The energy function of network is determined first, which includes network
Output energy function and it is each constraint conversion energy function
E=E5+E1+E2+E3+E4
In formula, E5For the most short target of distance, E1For active path constraint, E2For input and output path constraint, E3For network receipts
Hold back constraint, E4For the constraint of defined starting and terminal point.
In turn, the transmission function and state transition equation for determining neuron, by the evolution repeatedly of network, until restrain,
When network is finally restrained by developing, the transposition battle array being made of 0 and 1 can be formed, 1 position in battle array indicates institute
The node of process, the sum of the distance between these nodes is the shortest distance.
Embodiment 5
As the 5th kind of embodiment of the invention, vehicle scheduling module can also select heuritic approach to be realized, make
For a kind of preferred embodiment, as shown in fig. 7, vehicle scheduling module is designed based on heuritic approach, heuritic approach includes establishing
Model module and solving model module establish model module for the distribution information of vehicle to be established model, solving model module
For solving model, and obtain optimal transportation dispatching scheme.
In the present embodiment, model module method is established are as follows: numbering parking lot is 0, car number k, mission number 1,
2 ..., l consider the constraints such as freight volume constraint, the constraint of stop number of vehicles, consolidating the load and unloading time constraint, can be defined as follows
Basic model:
ETi≤si≤LTiI=1 ..., l
In formula, cijIndicate the transportation cost from point i to j, it can be embodied as transportation range according to the target of optimization
Or freight or haulage time.xijkAnd ykiFor variable, is defined as:
In formula, ETiAnd LTiThe late finish time of earliest start time and permission that respectively task i allows；giIt is i-th
The volume of goods transported of point, q are the nominal load capacity of haulage vehicle.
Specifically, solving model module is based on C-W algorithm Solving Vehicle Scheduling Problem, its step are as follows:
(1) calculates the cost savings value s (i, j) of route between each point i and point j first, forms set M, and according to from
It arrives greatly and small s (i, j) is ranked up, in which: s (i, j)=ci0+c0j-cij；
(2) if M be sky, termination iterate, otherwise in M first item s (i, j) investigation whether meet following condition it
One, it is walked under turning if meeting, otherwise goes to step (6)；
(a) point i and j is not on the route constituted；
(b) point i and j is connected on the route constituted, but not with parking lot；
(c) point i and j is located on the different routes constituted, is not connected with parking lot, and one is starting point, and one is eventually
Point.
(3) investigates total volume of goods transported Q on the route after point i is connected with j and walks under turning if Q≤q, otherwise goes to step (6)；
(4) after the route where calculating tie point i and j, the time that vehicle reaches j point reaches j point than vehicle on former route
Time variable quantity are as follows: EFj=si+Ti+tij-sj；
If (a) EFj=0, go to step (5)；
If (b) EFj < 0, Δ j is calculated-, when | EFj |≤Δ j-, (5) are gone to step, (6) are otherwise gone to step；
If (c) EFj> 0, then calculate Δ j+, when | EFj|≤Δj+, (5) are gone to step, (6) are otherwise gone to step.
In formula, Δ j-The maximum allowable of j point time is reached for what is withouted waiting at the subsequent each task of j point on route
Lead；Δj+For the maximum allowable postponement of the arrival j point time that time-to-violation does not constrain of the subsequent each task of j point on route
Amount, in which:
(5) tie point i and point j calculates new time when vehicle reaches each task；
(6) enables M=M-s (i, j), goes to step (2).
Embodiment 6
As the 6th kind of embodiment of the invention, for the ease of passing through client and database realizing data interaction, this hair
Bright personnel make improvements client, and as a kind of preferred embodiment, as shown in Figure 8 and Figure 9, client includes that user logs in mould
Block and operation module, user log-in block log in client for user, and operation module is used to realize by client and manipulate,
Operation module includes data uploading module and task assigns module, and data uploading module is used to upload by client to database
Vehicle and logistics distribution data, task assign module for assigning logistics distribution information to database by client.
In the present embodiment, the data uploaded in data uploading module include information of vehicles, map datum and logistic resources
Information.
Further, task assigns the starting point, destination and goods information that logistics is assigned in module.
The basic principles, main features and advantages of the present invention have been shown and described above.The technology of the industry
For personnel it should be appreciated that the present invention is not limited to the above embodiments, described in the above embodiment and specification is only the present invention
Preference, be not intended to limit the invention, without departing from the spirit and scope of the present invention, the present invention also has various
Changes and improvements, these changes and improvements all fall within the protetion scope of the claimed invention.The claimed scope of the invention is by institute
Attached claims and its equivalent thereof.
Claims (9)
1. a kind of logistics delivery vehicle scheduling method based on mass data, comprising the following steps:
S1, it establishes database: vehicle and logistics distribution data being uploaded to database by client, built with cloud computing technology
The physical resource data of dispatching are carried out pond, are formed total data by logistics private clound large data center using virtualization technology
Library；
S2, service display: the system resource of logistics distribution, data resource, information resources are carried out unification by database, and with breviary
The form of figure is shown, while the information data in database being synchronized in client；
S3, task are assigned: user of service logs in client and assigns logistics distribution information command, and client matches the logistics assigned
Information command is sent to be uploaded in database；
S4, vehicle scheduling: the vehicle and logistics distribution data in distribution information combined data library assigned according to client generate
Reasonable vehicle tune information, and feed back to client.
2. the logistics delivery vehicle scheduling method according to claim 1 based on mass data, it is characterised in that: institute
Stating database includes data segmentation module, data storage module, data modeling module and foreground display module, the data segmentation
Module is used to do Hash segmentation by field for the physical resource data in number library, forms multiple subdata bases, and the data save
Module for being saved to the data after segmentation, the data modeling module be used to create the database model and
Models catalogue generate .edmx file .edmx file is an XML file, it for defined notion model, storage model and
Mapping between these models .edmx file also includes ADO.NET entity data model designer, for being in graphically
The information of existing model, the foreground display module are used to realize that data are handed between the database and each client
Mutually, it carries out uploading data by the client and assigns instruction.
3. the logistics delivery vehicle scheduling method according to claim 2 based on mass data, it is characterised in that: institute
Stating data modeling module includes data import modul and service view module, and the data import modul is used for the database
Interior data are directed into data model, and the service view module is shown for the data information of importing to be converted to view
Show.
4. the logistics delivery vehicle scheduling method according to claim 2 based on mass data, it is characterised in that: institute
Stating foreground display module includes user interactive module and vehicle scheduling module, and the user interactive module is by each client
It is connected with the database, so that generating data interaction, the vehicle scheduling between the database and each client
Module is used for the distribution information assigned according to the client and in conjunction with the logistics information in the database, generates reasonable vehicle
Adjust information.
5. the logistics delivery vehicle scheduling method according to claim 4 based on mass data, it is characterised in that: institute
Stating vehicle scheduling module includes that path planning module, freight charges planning module, constraint processing module, genetic operator module and scheme are true
Cover half block, the path planning module are optimal for planning for planning optimal Distribution path, the freight charges planning module
Freight charges are dispensed, the constraint processing module is used to guarantee the correctness of vehicle scheduling optimization, and the genetic operator module is for protecting
Defect individual is stayed, gene delection is avoided, improves global convergence and efficiency, the scheme determining module is for generating optimal transport
Scheduling scheme.
6. the logistics delivery vehicle scheduling method according to claim 5 based on mass data, it is characterised in that: institute
It states vehicle scheduling module to design based on neural network algorithm, the neural network algorithm includes adjacency matrix module, schedule constraints
Module, neural network module and scheme form module, the adjacency matrix module be used for by the source point of vehicle, passed through it is each
Meeting point and rest point are abstracted into the node of network, and the schedule constraints module is used to guarantee the correctness of vehicle scheduling optimization, described
Neural network module is used to determine the transmission function and state transition equation of neuron, by the evolution repeatedly of network, until receiving
It holds back, the scheme forms module for generating optimal transportation dispatching scheme.
7. the logistics delivery vehicle scheduling method according to claim 6 based on mass data, it is characterised in that: institute
It states vehicle scheduling module to design based on heuritic approach, the heuritic approach includes establishing model module and solving model mould
Block, for the model module of establishing for the distribution information of vehicle to be established model, the solving model module is used for solving model,
And obtain optimal transportation dispatching scheme.
8. the logistics delivery vehicle scheduling method according to claim 7 based on mass data, it is characterised in that: institute
Stating client includes user log-in block and operation module, and the user log-in block logs in the client for user,
The operation module is used to realize by the client and manipulate.
9. the logistics delivery vehicle scheduling method according to claim 8 based on mass data, it is characterised in that: institute
State that operation module includes data uploading module and task assigns module, the data uploading module be used for by the client to
The database uploads vehicle and logistics distribution data, the task assign module for by the client to the data
Assign logistics distribution information in library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910461469.8A CN110490503A (en) | 2019-05-30 | 2019-05-30 | A kind of logistics delivery vehicle scheduling method based on mass data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910461469.8A CN110490503A (en) | 2019-05-30 | 2019-05-30 | A kind of logistics delivery vehicle scheduling method based on mass data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110490503A true CN110490503A (en) | 2019-11-22 |
Family
ID=68546301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910461469.8A Pending CN110490503A (en) | 2019-05-30 | 2019-05-30 | A kind of logistics delivery vehicle scheduling method based on mass data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110490503A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110942271A (en) * | 2019-12-11 | 2020-03-31 | 江苏佳利达国际物流股份有限公司 | Intelligent transportation scheduling management method based on neural network genetic algorithm |
| CN111199354A (en) * | 2019-12-31 | 2020-05-26 | 广东电网有限责任公司 | Hopfield neural network-based power communication network operation and maintenance work order scheduling method |
| CN111199375A (en) * | 2020-01-13 | 2020-05-26 | 江苏佳利达国际物流股份有限公司 | Intelligent logistics transportation system |
| CN112836846A (en) * | 2020-12-02 | 2021-05-25 | 红云红河烟草(集团)有限责任公司 | Multi-depot and multi-direction combined transportation scheduling double-layer optimization algorithm for cigarette delivery |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170270448A1 (en) * | 2016-03-18 | 2017-09-21 | Jusda International Logistics (TAIWAN) CO.,LTD | Vehicle scheduling device and method for transportation systems |
| CN107220731A (en) * | 2017-05-23 | 2017-09-29 | 南京邮电大学 | A kind of logistics distribution paths planning method |
| CN109816246A (en) * | 2019-01-28 | 2019-05-28 | 江苏大地物流有限责任公司 | Logistic dispatching system based on GIS and intelligent algorithm |
-
2019
- 2019-05-30 CN CN201910461469.8A patent/CN110490503A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170270448A1 (en) * | 2016-03-18 | 2017-09-21 | Jusda International Logistics (TAIWAN) CO.,LTD | Vehicle scheduling device and method for transportation systems |
| CN107220731A (en) * | 2017-05-23 | 2017-09-29 | 南京邮电大学 | A kind of logistics distribution paths planning method |
| CN109816246A (en) * | 2019-01-28 | 2019-05-28 | 江苏大地物流有限责任公司 | Logistic dispatching system based on GIS and intelligent algorithm |
Non-Patent Citations (6)
| Title |
|---|
| 刘镇等: "基于云计算的冷链物流配送车辆路径优化方法研究", 《电子设计工程》 * |
| 彭志忠等: "《物流管理学》", 31 August 2005, 山东大学出版社 * |
| 徐优香: "基于云计算的冷链物流配送车辆路径优化方法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李知杰等: "基于EF的数据建模方法", 《软件导刊》 * |
| 高晓亮等: "《仓储与配送管理》", 31 October 2006, 北京交通大学出版社 * |
| 齐二石: "《物流工程 修订版》", 31 January 2005, 中国科学技术出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110942271A (en) * | 2019-12-11 | 2020-03-31 | 江苏佳利达国际物流股份有限公司 | Intelligent transportation scheduling management method based on neural network genetic algorithm |
| CN111199354A (en) * | 2019-12-31 | 2020-05-26 | 广东电网有限责任公司 | Hopfield neural network-based power communication network operation and maintenance work order scheduling method |
| CN111199375A (en) * | 2020-01-13 | 2020-05-26 | 江苏佳利达国际物流股份有限公司 | Intelligent logistics transportation system |
| CN111199375B (en) * | 2020-01-13 | 2021-01-01 | 江苏佳利达国际物流股份有限公司 | Intelligent logistics transportation system |
| CN112836846A (en) * | 2020-12-02 | 2021-05-25 | 红云红河烟草(集团)有限责任公司 | Multi-depot and multi-direction combined transportation scheduling double-layer optimization algorithm for cigarette delivery |
| CN112836846B (en) * | 2020-12-02 | 2022-07-08 | 红云红河烟草(集团)有限责任公司 | Multi-depot and multi-direction combined transportation scheduling double-layer optimization algorithm for cigarette delivery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110490503A (en) | A kind of logistics delivery vehicle scheduling method based on mass data | |
| CN111862579B (en) | Taxi scheduling method and system based on deep reinforcement learning | |
| Xu et al. | Multi-objective artificial bee colony algorithm for multi-stage resource leveling problem in sharing logistics network | |
| CN108880663B (en) | Space-ground integrated network resource allocation method based on improved genetic algorithm | |
| Sungur et al. | A model and algorithm for the courier delivery problem with uncertainty | |
| CN112418497B (en) | Material distribution path optimization method for manufacturing Internet of things | |
| Ahmadizar et al. | Two-level vehicle routing with cross-docking in a three-echelon supply chain: A genetic algorithm approach | |
| CN105787595B (en) | Emergency logistics transfer station site selection method based on improved ant colony algorithm | |
| Dechampai et al. | A differential evolution algorithm for the capacitated VRP with flexibility of mixing pickup and delivery services and the maximum duration of a route in poultry industry | |
| Sivanandam et al. | Genetic algorithm optimization problems | |
| CN107798388B (en) | Measurement and control resource scheduling and allocation method based on Multi-Agent and DNN | |
| Liu et al. | Novel multi-objective resource allocation and activity scheduling for fourth party logistics | |
| CN112766813A (en) | Air-space cooperative observation complex task scheduling method and system | |
| CN112987757B (en) | Path planning method for multi-mode intermodal transportation of goods | |
| CN101714228A (en) | Single step flight schedule optimization | |
| CN109214756A (en) | Based on ant group algorithm and the complete vehicle logistics dispatching method and device of hierarchy optimization, storage medium, terminal | |
| Eltoukhy et al. | Heuristic approaches for operational aircraft maintenance routing problem with maximum flying hours and man-power availability considerations | |
| Sherali et al. | Two-stage fleet assignment model considering stochastic passenger demands | |
| CN104537446B (en) | Two layers of band fuzzy stochastic time window vehicle routing optimization method | |
| CN113496313A (en) | General assembly workshop material demand prediction method | |
| Wang et al. | Solving task scheduling problems in cloud manufacturing via attention mechanism and deep reinforcement learning | |
| Wu et al. | Managing pharmaceuticals delivery service using a hybrid particle swarm intelligence approach | |
| Jia et al. | Dynamic container drayage with uncertain request arrival times and service time windows | |
| Jin et al. | Container port truck dispatching optimization using Real2Sim based deep reinforcement learning | |
| CN109345008A (en) | Automatic row's ship's method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |