CN104866997A - Intelligent pairing method applied to online stowage of vehicles and goods - Google Patents

Abstract

The invention discloses an intelligent pairing method applied to online stowage of vehicles and goods. Information of supply of goods and platform vehicles issued by a goods owner is fully considered based on a word segmentation spatial clustering method, and the information of supply of goods issued by the goods owner is subjected to word segmentation from multiple dimensions, a vehicle information clustering space is constructed, and online intelligent pairing of the supply of goods and vehicles is achieved by the floating space and fuzzy filtering; a plurality of attributes influencing vehicle selection are divided into sequential layers associated with one another based on a multi-attribute decision making model, thus being methodized, and the matched vehicles are optimally sorted according to vehicle transaction evaluation and mass data of the platform, so that the result can be more intuitively presented. According to the intelligent pairing method applied to online stowage of vehicles and goods, the intelligent optimization pairing problem of vehicles and goods in the electronic commerce platform taking logistics service as a transaction object is well processed by the accurate big data mining analysis and the complicated model algorithm, so that the platform members or other members can quickly and accurately find the best fit stowage vehicle from the numerous vehicles and make a selection according with own requirements.

Description

A kind of intelligent matching method for the online prestowage of car goods

Technical field

The present invention relates to a kind of Intelligent Optimal matching method for the online prestowage of car goods, be applicable to logistics e-platform, from multiple dimension, participle carried out to the information of freight source that the owner of cargo issues, structure information of vehicles Cluster space, realize the on-line intelligence pairing of the source of goods and vehicle, and sequence is optimized to coupling vehicle, belong to intelligent matching technology field.

Background technology

The development of infotech, promote ecommerce to be combined with the degree of depth of logistic industry, the maximum feature of logistics e-commerce platform realizes looking for car online and looking for goods online exactly, but will suitable adaptive vehicle be found quickly and accurately to be the problem that the owner of cargo is concerned about most in platform magnanimity information of vehicles.Existing logistics e-commerce platform indiscriminately imitates internet searching method mostly, the foundation of participle system is also relatively unsound, the information of freight source that the owner of cargo issues cannot be reflected objective, fully, also unsound to the Classification Management of information of vehicles, intelligence pairing fast and accurately cannot be realized; In addition large data are another features of logistics e-commerce platform, cannot make optimal selection after the owner of cargo issues information of freight source in the intelligence pairing vehicle of magnanimity.Therefore, current logistics e-commerce platform need one be conducive to user quick, accurate, objective, reasonably select vehicle member, thus improve the intelligent matching method of the online prestowage of effective wagon goods of conevying efficiency.

Summary of the invention

Goal of the invention: for Problems existing in existing logistics e-commerce platform car goods intelligence matching method with not enough, on all information of vehicles in transaction platform and transaction record data basis, the invention provides a kind of online prestowage intelligence matching method of car goods based on participle Cluster space and Multiple Attribute Decision Model and optimal screening model.

Technical scheme: a kind of car goods based on participle Cluster space and Multiple Attribute Decision Model online prestowage intelligence matching method, is applicable to provide and carries out intelligence to car goods in the e-commerce platform of physical distribution trading service and match.Specifically comprise the steps:

(1) information of freight source issues school inspection, and carries out level participle to information of freight source

The information of freight source of owner of cargo's Online release generally must comprise: Description of Goods, cargo type, quantity of goods, weight, volume, vehicle requirement, vehicle commander's requirement, start shipment ground, destination, when loading, receive the time etc., morphology is divided to carry out participle to information of freight source with level from top to bottom, ground floor information is Description of Goods, second layer information is cargo type, goods specification, shipment month and shipping interval, third layer comprises fresh-keeping product, fragile article, dangerous material, conventional product and other items type attribute, quantity of goods, weight, volume, packaging and other items specification attribute, loading time, the time of departure, time of arrival, to receive the shipment month attributes such as time, start shipment ground, the shipping Range Attributes such as destination.

(2) platform database information of vehicles extracts, based on source of goods participle information structuring Cluster space

For guaranteeing authenticity and the security of information of vehicles, the registered vehicle information of platform typing generally comprises license plate number, type of vehicle, compartment length, width and height, payload ratings, purchase phase vehicle day, pull wheel shaft, Motor Number, the vehicle essential informations such as vehicle photo, in addition vehicle can set empty wagons time and travel route according to actual needs, can to freight source at platform source of goods storehouse fast searching, improves shipping efficiency.In order to realize the intelligence pairing of car goods, being necessary to divide vehicle Cluster space for information of freight source, reaching the object of Rapid matching.Fresh-keeping product, fragile article, dangerous material, conventional product and other items type attribute is comprised based on source of goods third layer participle information, quantity of goods, weight, volume, packaging and other items specification attribute, the shipment month attributes such as loading time, the time of departure, time of arrival, the time of receiving, start shipment the shipping Range Attributes such as ground, destination.Determine vehicle Cluster space subspace comprise vehicle, load-carrying, vehicle commander, delivery availability, time of receiving, start shipment and destination, Cluster space can uniquely determine a class vehicle according to source of goods demand, realize intelligence pairing.

(3) according to source of goods determination Cluster space fluctuation area

Article one, information of freight source comprise determine vehicle demand, load-carrying demand, vehicle commander's demand, delivery availability demand, time demand of receiving, start shipment ground demand and destination demand series information, but most of vehicle can not meet these precision demand simultaneously, therefore some information can given one fluctuation tolerance interval, such as dead weight capacity is not less than 50 tons, the time of receiving is between 12:00-13:00 etc., and the vehicle that vehicle condition is in these fluctuation areas all meets source of goods demand.The tolerance interval of fluctuation area can be the owner of cargo oneself setting also can be that platform is according to the automatic default setting of vehicle historical trading data.

(4) fuzzy screening, determines intelligence pairing vehicle scheme collection

According to information of freight source and fluctuation area, the need satisfaction space of the every sub spaces of information of vehicles Cluster space can be determined, everyly be in need satisfaction space or the vehicle of need satisfaction spatial edge is all the pairing scheme vehicle to be selected met the demands, these vehicles form platform intelligent pairing vehicle scheme collection.

(5) data mining analysis attribute is chosen, and determines optional program vehicle attribute collection

The attribute of the pairing scheme vehicle to be selected that reference database and this platform of industry selecting index need, according to the mutual relationship between pairing scheme vehicle to be selected and each attribute, treats apolegamy and carries out intelligent sequencing and Optimized Matching to scheme vehicle.Property set: choose car age, driving age, transport punctuality (reaching on the time and speed of dispatching a car), Transport Safety (the guarantee service that vehicle participates in), acknowledgement of consignment business number, damage rate of goods and complaint number of times seven evaluation indexes as the attribute affecting scheme.

(6) property set standardization

The impact of apolegamy on scheme vehicle is treated, the profitable type of attribute of scheme and cost type two class according to attribute.Its property value of profit evaluation model attribute is the bigger the better, otherwise its property value of cost type attribute is the smaller the better.And the dimension of each scheme attribute and dimensional unit are also different, therefore must carry out nondimensionalization process to each attribute of scheme, the nondimensionalization process of profit evaluation model attribute and cost type attribute is as follows:

Profit evaluation model attribute process: b _ij=(a _ij-a _j ^min)/(a _j ^max-a _j ^min)

Cost type attribute process: b _ij=(a _j ^max-a _ij)/(a _j ^max-a _j ^min)

Wherein, a _ijthe property value of a jth attribute of scheme i, b _ija _ijvalue after standardization, a _j ^maxa jth attribute P _jmaximal value, a _j ^minp _jminimum value.B _ij∈ (0,1), i=1,2 ..., m, j=1,2 ..., n, normalized matrix B=(b _ij) _{m × n}.

(7) carry out multidimensional analysis, build the polynary Optimized model of attribute weight

Consider the validity of information feed back and the rationality of information processing, introduce polynary Optimized model herein, the dimension different with two, platform is evaluated from vehicle transaction, comprehensive consideration they treat the Different Effects of apolegamy to scheme vehicle, and on the basis of large data analysis, mathematics is carried out to them and portrays.Vehicle transaction is evaluated different from the emphasis of the owner of cargo with platform, therefore shows different attribute evaluation values to the emphasis of owner of cargo's prestige.

1. polynary weight is determined

In order to reduce single preference and the impact of cognitive limitation on result, platform is weighted average to the history evaluation marking of all owners of cargo to each index of vehicle, show that the owner of cargo is to vehicle comprehensive evaluation marking the most intuitively; In order to reduce the deviation that single dimension evaluation brings, introduce platform decision-making unit from another dimension herein---treat apolegamy based on platform historical trading data and trading activity and supplementary evaluation marking is carried out to scheme vehicle.Then air exercise is divided into the weighted value that column criterion process obtains each attribute, is finally that two decision-making units arrange rational significance level coefficient based on the integrality of data and authenticity.

The decision-making unit T that two dimensions are determined _k: k=1,2; Wherein T ₁represent carrier member's decision-making unit, T ₂represent platform decision-making unit.

The attribute weight of two decision-making unit assignment is: w ^k=(w ₁ ^k, w ₂ ^k..., w _n ^k) ^t, k=1,2

The significance level of each decision-making unit is: z=(z ₁, z ₂) ^t, wherein, z ₁+ z ₂=1, z _k>=0, k=1,2.

2. unitary weight optimization model is built

Consider vehicle transaction evaluation information may with subjective factor, herein from subjective weight determination method angle, comprehensive large data mining technology, builds attribute weight Optimized model as follows:

$\min L_1=\underset{k=1}{\overset{2}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{z}_k{(w_j-{w_j}^k)}^2$

$s.t.\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1,w_j\≥0,j=1,2,...,n$

The implication of model tries to achieve a w _j, make w _jand w _j ^kthe quadratic sum L of total partial variance ₁minimum.

3. binary weight optimization model is built

In order to reduce human factor in model as far as possible, herein from objective weight determination method angle, multidimensional analysis is carried out to data, builds attribute weight Optimized model as follows:

G-minL＝(l ₁,l ₂,...l _m)

Wherein, a _j ^*=max{a _1j, a _2j... a _mjbe attribute P _jideal value, with wait power weigthed sums approach, can be by model simplification:

$\min L_2=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{l}_i=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2{w_j}^2$

$s.t.\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1,w_j\≥0,j=1,2,...,n$

4. single object optimization model is synthesized

Above two Optimized models are carried out being integrated into G=min (L ₁, L ₂), and by linear weighting method, problem is converted into following single object optimization model:

$\min F=x\underset{k=1}{\overset{2}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{z}_k{(w_j-{w_j}^k)}^2+y\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2{w_j}^2$

$s.t.\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1,w_j\≥0,j=1,2,...,n$

x+y＝1，x＞0,y＞0

Wherein, x, y are the coefficients of polynary Optimized model, show as both relative importance.By building Lagrangian function, can obtain:

$w_j=c_j[d_j+(1-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{d}_j)/\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j]$

Wherein, $c_j=1/[x+y\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2],d_j=x\underset{k=1}{\overset{2}{\mathrm{\Σ}}}{z}_k{w_j}^k$

(8) attribute weight is solved

Due to multiple decision-making unit (T _k: k=1,2) to the comprehensive weight imparting value of attribute j be:

$w_j=c_j[d_j+(1-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{d}_j)/\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j],j=\mathrm{1,2},...,n$

The weight vectors of the polynary Optimized model of attribute weight is: w ^*=[w ₁, w ₂... w _j..., w _n], wherein $w_j=c_j[d_j+(1-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{d}_j)/\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j]$

(9) pairing scheme optimization of vehicle sequence to be selected

The each Attribute Relative Evaluations matrix of pairing scheme vehicle to be selected is A ^*=(A _ij) _{m × n}=Bw ^{* T}

Wherein, A _ijfor the final weighted value of a jth attribute of pairing scheme vehicle i to be selected, B=(b _ij) _{m × n}for property set normalized matrix.

The then relative evaluation of pairing scheme vehicle i to be selected

Relative evaluation based on pairing scheme vehicle to be selected is treated apolegamy and is carried out objective, rational Optimal scheduling to scheme vehicle, the owner of cargo can be made to optimize more fast, more convenient, more reasonably understand and select adaptive goods stock, greatly improve the shipping efficiency of platform.

Beneficial effect: compared with prior art, the present invention is applicable to provide in the e-commerce platform of physical distribution trading service and carries out intelligence pairing to car goods.Its feature carries out participle from multiple dimension to the information of freight source that the owner of cargo issues, structure information of vehicles Cluster space, realizes the on-line intelligence pairing of the source of goods and vehicle; And by affecting numerous attributes of vehicle selection by being divided into the orderly level connected each other, make it methodization, the mass data of conclude the business according to vehicle evaluation and platform, sequence is optimized to coupling vehicle, result is presented more directly perceived.According to the Optimal scheduling result of intelligence pairing, the comprehensive delivery situation of vehicle can be known objectively, solve some problem of information asymmetry in car goods on-line intelligence pairing process to a certain extent, make platform member or other members in countless vehicle, find the vehicle of the most applicable prestowage quickly and accurately and make the selection meeting oneself demand.

Accompanying drawing explanation

Fig. 1 is method flow diagram of the present invention;

Fig. 2 is the information of freight source participle model of the embodiment of the present invention.

Embodiment

Below in conjunction with specific embodiment, illustrate the present invention further, these embodiments should be understood only be not used in for illustration of the present invention and limit the scope of the invention, after having read the present invention, the amendment of those skilled in the art to the various equivalent form of value of the present invention has all fallen within the application's claims limited range.

(1) information of freight source issues school inspection, and carries out level participle to information of freight source

The information of freight source of owner of cargo's Online release generally must comprise: Description of Goods, cargo type, quantity of goods, weight, volume, vehicle requirement, vehicle commander's requirement, start shipment ground, destination, when loading, receive the time etc., morphology is divided to carry out participle as Fig. 2 to information of freight source with level from top to bottom, ground floor information is Description of Goods, second layer information is cargo type, goods specification, shipment month and shipping interval, third layer comprises fresh-keeping product, fragile article, dangerous material, conventional product and other items type attribute, quantity of goods, weight, volume, packaging and other items specification attribute, loading time, the time of departure, time of arrival, to receive the shipment month attributes such as time, start shipment ground, the shipping Range Attributes such as destination.The present embodiment information of freight source is as table 1:

Table 1 embodiment information of freight source table

Description of Goods	Cargo type		Fresh-keeping product
				Spanish mackerel	Goods specification	Quantity	10 casees
		Weight	10 tons
						Volume	15m 3
		Packaging	Congealer is packed
					Shipment month	Delivery availability	12:00±1h
		Receive the time	20:00±1h
					Shipping is interval	Start shipment ground	Zhenjiang Hui Longgang
		Destination	Formocarbam supermarket, Hankow road, Nanjing

(2) platform database information of vehicles extracts, based on source of goods participle information structuring Cluster space

For guaranteeing authenticity and the security of information of vehicles, the registered vehicle information of platform typing generally comprises license plate number, type of vehicle, compartment length, width and height, payload ratings, purchase phase vehicle day, pull wheel shaft, Motor Number, the vehicle essential informations such as vehicle photo, in addition vehicle can set empty wagons time and travel route according to actual needs, can to freight source at platform source of goods storehouse fast searching, improves shipping efficiency.In order to realize the intelligence pairing of car goods, being necessary to divide vehicle Cluster space for information of freight source, reaching the object of Rapid matching.Fresh-keeping product, fragile article, dangerous material, conventional product and other items type attribute is comprised based on source of goods third layer participle information, quantity of goods, weight, volume, packaging and other items specification attribute, the shipment month attributes such as loading time, the time of departure, time of arrival, the time of receiving, start shipment the shipping Range Attributes such as ground, destination.Determine vehicle Cluster space subspace comprise vehicle, load-carrying, vehicle commander, delivery availability, time of receiving, start shipment and destination, Cluster space can uniquely determine a class vehicle according to source of goods demand, realize intelligence pairing.The present embodiment information of vehicles Cluster space and floating space are as table 2:

Table 2 embodiment information of vehicles Cluster space and floating space

(4) fuzzy screening, determines intelligence pairing vehicle scheme collection

According to information of freight source and fluctuation area, the need satisfaction space of the every sub spaces of information of vehicles Cluster space can be determined, everyly be in need satisfaction space or the vehicle of need satisfaction spatial edge is all the pairing scheme vehicle to be selected met the demands, these vehicles form platform intelligent pairing vehicle scheme collection.The vehicle satisfied condition that the present embodiment platform searches has 5, and information is as following table 3:

Table 3 embodiment optional program vehicle scheme collection

(5) data mining analysis attribute is chosen, and determines optional program vehicle attribute collection

The attribute of the pairing scheme vehicle to be selected that reference database and this platform of industry selecting index need, according to the mutual relationship between pairing scheme vehicle to be selected and each attribute, treats apolegamy and carries out intelligent sequencing and Optimized Matching to scheme vehicle.Property set: choose car age, driving age, transport punctuality (reaching on the time and speed of dispatching a car), Transport Safety (the guarantee service that vehicle participates in), acknowledgement of consignment business number, damage rate of goods and complaint number of times seven evaluation indexes as the attribute affecting scheme.The correlation attribute information of the present embodiment optional program vehicle is as following table 4:

Table 4 optional program vehicle scheme attribute information table

	Car age	Driving age	Punctuality	Security	Business number	Damage rate of goods ‰	Complain number
								A	5	9	Generally	Good	15	2	2
B	6	10	Good	Generally	20	1	1
								C	4	12	Better	Good	18	2	3
D	7	8	Generally	Good	16	2	3
								E	5	15	Generally	Better	30	1	4

(6) property set standardization

The impact of apolegamy on scheme vehicle is treated, the profitable type of attribute of scheme and cost type two class according to attribute.Its property value of profit evaluation model attribute is the bigger the better, otherwise its property value of cost type attribute is the smaller the better.And the dimension of each scheme attribute and dimensional unit are also different, therefore must carry out nondimensionalization process to each attribute of scheme, the nondimensionalization process of profit evaluation model attribute and cost type attribute is as follows:

Profit evaluation model attribute process: b _ij=(a _ij-a _j ^min)/(a _j ^max-a _j ^min)

Cost type attribute process: b _ij=(a _j ^max-a _ij)/(a _j ^max-a _j ^min)

Wherein, a _ijthe property value of a jth attribute of scheme i, b _ija _ijvalue after standardization, a _j ^maxa jth attribute P _jmaximal value, a _j ^minp _jminimum value.B _ij∈ (0,1), i=1,2 ..., m, j=1,2 ..., n, normalized matrix B=(b _ij) _{m × n}.

According to upper table 4 optional program vehicle attribute information table, the value of each attribute has quantitative value and non-quantitative value two class, and wherein non-quantitative value must carry out quantification process, just has the value compared, and the quantification process of the non-quantitative value of the present embodiment is as following table 5:

Table 5 is grade quantizing table quantitatively

1	2	3	4	5	6	7	8	9
									The poorest	Very poor	Difference	Poor	Generally	Better	Good	Very well	Best

According to the property set value after quantification process, optional program vehicle attribute decision matrix can be obtained as follows:

$A=\left[\begin{array}{ccccccc}5& 9& 5& 7& 15& 2& 2\\ 6& 10& 7& 5& 20& 1& 1\\ 4& 12& 6& 7& 18& 2& 3\\ 7& 8& 5& 7& 16& 2& 3\\ 5& 15& 5& 6& 30& 1& 4\end{array}\right]$

Wherein, car age, driving age, transport punctuality (reaching on the time and speed of dispatching a car), Transport Safety (the guarantee service that vehicle participates in), acknowledgement of consignment business number are profit evaluation model attributes, and property value is the bigger the better; Damage rate of goods and complaint number of times are cost type attributes, and property value is the smaller the better.After carrying out standardization to matrix A, the normalized matrix obtained is:

$B=\left[\begin{array}{ccccccc}1/3& 1/7& 0& 1& 0& 1& 1/3\\ 2/3& 2/7& 1& 0& 1/3& 0& 0\\ 0& 4/7& 1/2& 1& 1/5& 1& 2/3\\ 1& 0& 0& 1& 0& 1& 2/3\\ 1/3& 1& 0& 1/2& 1& 0& 1\end{array}\right]$

(7) carry out multidimensional analysis, build the polynary Optimized model of attribute weight

Consider the validity of information feed back and the rationality of information processing, introduce polynary Optimized model herein, the dimension different with two, platform is evaluated from vehicle transaction, comprehensive consideration they treat the Different Effects of apolegamy to scheme vehicle, and on the basis of large data analysis, mathematics is carried out to them and portrays.Vehicle transaction is evaluated different from the emphasis of the owner of cargo with platform, therefore shows different attribute evaluation values to the emphasis of owner of cargo's prestige.

In order to reduce single preference and the impact of cognitive limitation on result, platform is weighted average to the history evaluation marking of all owners of cargo to each index of vehicle, show that the owner of cargo is to vehicle comprehensive evaluation marking the most intuitively; In order to reduce the deviation that single dimension evaluation brings, introduce platform decision-making unit from another dimension herein---treat apolegamy based on platform historical trading data and trading activity and supplementary evaluation marking is carried out to scheme vehicle.Then air exercise is divided into the weighted value that column criterion process obtains each attribute, finally arranges significance level coefficient based on the integrality of data and authenticity.

To carrier member and platform, the marking to each attribute is weighted on average this example, then carries out standardization and obtains:

The attribute weight of vehicle evaluation information decision-making unit assignment is:

w ¹＝(0.53,0.13,0.1,0.6,0.1,0.64,0.64) ^T

The attribute weight of platform decision-making unit assignment is: w ²=(0.2,0.2,0.4,0.75,0.15,0.9,0.4) ^t

The significance level of each decision-making unit is: z=(0.45,0.55) ^t.

(4) attribute weight is solved

In order to make attribute weight both comprise subjective preferences, again containing objective information, subjective weight optimization model and objective weight Optimized model can be carried out being integrated into G=min (L ₁, L ₂), and by linear weighting method, problem is converted into following single object optimization model:

$\min F=x\underset{k=1}{\overset{q}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{z}_k{(w_j-{w_j}^k)}^2+y\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2{w_j}^2$

$s.t.\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1,w_j\≥0,j=1,2,...,n$

x+y＝1，x＞0,y＞0

Wherein, x, y are the coefficients of polynary Optimized model, show as both relative importance.Now making x=0.5, y=0.5, by building Lagrangian function, can obtain:

c＝(0.89,0.89,0.33,0.67,0.4,0.22,0.1) ^T

d＝(0.15,0.1,0.11,0.31,0.06,0.38,0.25) ^T

According to formula, then can obtain and reflect that the weight vectors of carrier member and platform information is simultaneously:

w ^*＝(0.23,0.19,0.07,0.28,0.07,0.11,0.04) ^T

(9) pairing scheme optimization of vehicle sequence to be selected

The each Attribute Relative Evaluations matrix of pairing scheme vehicle to be selected is A ^*=(A _ij) _{m × n}=Bw ^{* T}

Wherein, A _ijfor the final weighted value of a jth attribute of pairing scheme vehicle i to be selected, B=(b _ij) _{m × n}for property set normalized matrix.

The then relative evaluation of pairing scheme vehicle i to be selected

Obtain the evaluation collection of pairing scheme vehicle to be selected so have so scheme 4 is optimal cases, namely pairing scheme D to be selected is the optimum pairing vehicle that the online prestowage of the present embodiment car goods realizes intelligence pairing, secondly be vehicle C, vehicle E, vehicle A and vehicle B, so the intelligence pairing result of the online prestowage of the present embodiment car goods is as following table 6:

The intelligence pairing Output rusults of the online prestowage of table 6 embodiment car goods

Claims (1)

1., for an intelligent matching method for the online prestowage of car goods, it is characterized in that, specifically comprise the steps:

(1) information of freight source issues school inspection, and carries out level participle to information of freight source

(2) platform database information of vehicles extracts, based on source of goods participle information structuring Cluster space

(3) according to source of goods determination Cluster space fluctuation area

(4) fuzzy screening, determines intelligence pairing vehicle scheme collection

(5) data mining analysis attribute is chosen, and determines optional program vehicle attribute collection

(6) property set standardization

The impact of apolegamy on scheme vehicle is treated, the profitable type of attribute of scheme and cost type two class according to attribute; The nondimensionalization process of profit evaluation model attribute and cost type attribute is as follows:

Profit evaluation model attribute process: b _ij=(a _ij-a _j ^min)/(a _j ^max-a _j ^min)

Cost type attribute process: b _ij=(a _j ^max-a _ij)/(a _j ^max-a _j ^min)

Wherein, a _ijthe property value of a jth attribute of scheme i, b _ija _ijvalue after standardization, a _j ^maxa jth attribute P _jmaximal value, a _j ^minp _jminimum value; b _ij∈ (0,1), i=1,2 ..., m, j=1,2 ..., n, normalized matrix B=(b _ij) _{m × n};

(7) carry out multidimensional analysis, build the polynary Optimized model of attribute weight

1. polynary weight is determined

In order to reduce single preference and the impact of cognitive limitation on result, platform is weighted average to the history evaluation marking of all owners of cargo to each index of vehicle, show that the owner of cargo is to vehicle comprehensive evaluation marking the most intuitively; In order to reduce the deviation that single dimension evaluation brings, introduce platform decision-making unit from another dimension herein---treat apolegamy based on platform historical trading data and trading activity and supplementary evaluation marking is carried out to scheme vehicle.Then air exercise is divided into the weighted value that column criterion process obtains each attribute, is finally that two decision-making units arrange rational significance level coefficient based on the integrality of data and authenticity;

The decision-making unit T that two dimensions are determined _k: k=1,2; Wherein T ₁represent carrier member's decision-making unit, T ₂represent platform decision-making unit;

The attribute weight of two decision-making unit assignment is: w ^k=(w ₁ ^k, w ₂ ^k..., w _n ^k) ^t, k=1,2

The significance level of each decision-making unit is: z=(z ₁, z ₂) ^t, wherein, z ₁+ z ₂=1, z _k>=0, k=1,2;

2. unitary weight optimization model is built

Build attribute weight Optimized model as follows:

$\begin{array}{cc}\min & L_1=\underset{k=1}{\overset{2}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{z}_k{(w_j-{w_j}^k)}^2\end{array}$

$\begin{array}{ccc}s.t.& \underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1& w_j\≥0\end{array},j=\mathrm{1,2},...,n$

The implication of model tries to achieve a w _j, make w _jand w _j ^kthe quadratic sum L of total partial variance ₁minimum.

3. binary weight optimization model is built

Build attribute weight Optimized model as follows:

G-minL＝(l ₁,l ₂,...l _m)

Wherein, a _j ^*=max{a _1j, a _2j... a _mjbe attribute P _jideal value, with wait power weigthed sums approach, can be by model simplification:

$\begin{array}{cc}\min & L_2=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{l}_i=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2{w_j}^2\end{array}$

$\begin{array}{ccc}s.t.& \underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1& w_j\≥0\end{array},j=\mathrm{1,2},...,n$

4. single object optimization model is synthesized

Above two Optimized models are carried out being integrated into G=min (L ₁, L ₂), and by linear weighting method, problem is converted into following single object optimization model:

$\begin{array}{cc}\min & F=x\underset{k=1}{\overset{2}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{z}_k{(w_j-{w_j}^k)}^2+y\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2{w_j}^2\end{array}$

$\begin{array}{ccc}s.t.& \underset{j=1}{\overset{n}{\mathrm{\Σ}}}{w}_j=1& w_j\≥0\end{array},j=\mathrm{1,2},...,n$

x+y＝1，x＞0,y＞0

Wherein, x, y are the coefficients of polynary Optimized model, show as both relative importance.By building Lagrangian function, can obtain:

$w_j=c_j[d_j+(1-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{d}_j)/\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j]$

Wherein, $c_j=1/[x+y\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{({a_j}^{*}-a_{\mathrm{ij}})}^2],d_j=x\underset{k=1}{\overset{2}{\mathrm{\Σ}}}{z}_k{w_j}^k$

(8) attribute weight is solved

Due to multiple decision-making unit (T _k: k=1,2) to the comprehensive weight imparting value of attribute j be:

$w_j=c_j[d_j+(1-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{d}_j)/\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j],j=\mathrm{1,2},...,n$

The weight vectors of the polynary Optimized model of attribute weight is: w ^*=[w ₁, w ₂... w _j..., w _n], wherein $w_j=c_j[d_j+(1-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{d}_j)/\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j]$

(9) pairing scheme optimization of vehicle sequence to be selected

The each Attribute Relative Evaluations matrix of pairing scheme vehicle to be selected is A ^*=(A _ij) _{m × n}=Bw ^{* T}

Wherein, A _ijfor the final weighted value of a jth attribute of pairing scheme vehicle i to be selected, B=(b _ij) _{m × n}for property set normalized matrix.

The then relative evaluation of pairing scheme vehicle i to be selected