CN106709692A - Logistics center storage position allocation method - Google Patents

Abstract

The invention relates to the technical field of logistics, particularly to a logistics center storage position allocation method. The method comprises the following optimization steps: calculating goods correlation; clustering and analyzing goods; allocating clustered goods to storage areas; allocating goods to specific goods positions and finely adjusting the storage of special goods according to the characteristics of the goods. The method reduces the operating time for a warehouse and improves the operating efficiency of the shelf-type warehouse without increasing equipment investment, by optimizing a logistics center storage position allocation management system, using the shelf-type warehouse center as a research object, optimizing the goods position allocation of the warehouse and beginning with optimization management and dispatching of the warehouse.

Description

A kind of logistics center's stock's allocation method

Technical field

The present invention relates to logistlcs technology field, more particularly to a kind of logistics center's stock's allocation method.

Background technology

In recent years, flourishing with ecommerce, the demand of logistlcs technology and depot storage capacity is also being continuously increased.Enterprise The logistics management level of industry directly affects the market competitiveness of enterprise, and the warehouse-in efficiency that goes out in warehousing management is stream tube One of key link in reason.The need for traditional storage mode can not meet production and circulation, and tiered warehouse facility has been obtained Increasing application.Current many warehousing and storage activities management are still following custom for many years and experience carrys out arrangement of goods, both waste Memory space, causes that the operating efficiency in warehouse is low again.Data shows, in whole logistics Stored Procedure, the most of the time For material storage, therefore, modern enterprise spends great effort to be stored in a warehouse in link in optimization to increase economic efficiency.For many years Next, the problem for storage optimization generates a large amount of correlative study achievements, wherein storage Static Design, resource distribution, optimization tune Degree is increasingly becoming study hotspot in recent years.

But, domestic large warehoused system still generally existing that automation, intelligence degree be high, operational efficiency is relatively low The problems such as, specifically, the subject matter that generally existing and urgent need are solved in large warehoused system at present is：

L () lacks the optimal control to goods yard distribution：Still warehousing orders essentially according to goods and according to the supply of material at present The mode of business's subregion carries out random storage, so as to increased outbound required time.

(2) it is substantially to be sorted with time order and function to use the storage, outbound task order that automatic or manual mode generates, and is lacked Optimal control to task order.The optimization of task order is an important content of Automatic Warehouse optimization.Appoint due to lacking The scheduling of business order, causes that piler (storage/RetrievalMae locations nes, SRM) is in the majority with the single task cycle, and conveying sets Standby no-load ratio, semi-load rate are higher, therefore extend the outbound cycle, reduce the utilization rate of equipment.

Therefore, it is necessary to be directed to the present situation of logistics center's stock's allocation management system, a kind of logistics center's storage space is designed Distribution and access path optimization method, with meet express delivery develop the need for, adapt to market demand.

The content of the invention

To solve the above problems, the present invention proposes a kind of logistics center's stock's allocation method, by logistics center's storage space Allocation manager system is optimized, and with goods-shelf type storage center as skilful object is ground, is optimized for the goods yard distribution for closing storehouse, from The optimum management and scheduling in warehouse are started with, and in the case where equipment investment is not increased, reduce the activity duration in warehouse, improve shelf The operating efficiency in formula warehouse.

In order to achieve the above object, the present invention proposes following technical scheme：

A kind of logistics center's stock's allocation method, mainly includes the following steps that：

The first step:Goods calculation of relationship degree

Goods calculation of relationship degree is concretely comprised the following steps：

Goods i and goods j connection degree is calculated, it is general by what is together occurred to goods i in same order and goods j Rate carries out goods calculation of relationship degree, and computing formula is as follows：

Wherein, p_ijFor the order number of times that goods i and goods j occurs jointly；p_i,p_jGoods i and goods j is represented respectively The number of times occurred in order；ρ_ijIt is the correlation degree of goods i and j.

Second step：Goods cluster analysis

Assuming that the species sum of warehouse goods is N, the batch total of goods polymerization is G, G≤N, the then model of cluster analysis It is as follows：

X_ig∈ { 0,1 }, X_jg∈ { 0,1 } (6)

Wherein, formula (2) is that all of goods is aggregated to mutually similar degree of association summation；Formula (3), (4) represent each goods Thing can only be assigned in a class, and all of goods is assigned in G classes according to the relevance between them；Formula (6) is represented and determined The span of plan variable.

3rd step：Distribution clusters goods to storage area

3rd step distribution cluster goods concretely comprising the following steps to storage area：

It is determined that the order of packets of cluster goods, for each class g, calculates the average order frequency of g class goods, each The center of individual storage area l is to warehouse gateway apart from d_l, determine which cluster goods g is assigned to drag by setting up One storage area：

Wherein, F_gIt is the average order frequency of g class goods；d_lFor in l inventory areas also apart from warehouse go out storage away from From；f_iIt is the order frequency of goods i；n_gIt is the species sum of g class goods；d_eFor in the l of storage space region goods yard e apart from gateway Distance；m_lIt is goods yard number in the l of storage space region；M is depot area sum；Formula (7) is used as object function, it is desired to institute in order There is the distance of class to the gateway of goods formation most short；Formula (8) represents that each storage area at most only stores a Type of Collective class Goods；Formula (9) represents same-polymeric type cargo storage a storage area；Formula (10) represents the average of each polymeric type goods Order frequency；Formula (11) represents that l inventory area centre distances warehouse goes out the distance of storage；Formula (12) represents decision variable.

4th step：Distribute goods to specific goods yard

4th step distribution cluster goods concretely comprising the following steps to storage area：

Operated in each storage region, cluster goods is carried out from high to low according to the order-placing frequency of each goods Arrangement, by order frequency cargo storage high in the goods yard near apart from gateway, completes the distribution of all class goods：

d_aOrder completes the distance of order a pickings；O is total number of orders；Formula (13) picked all orders apart from summation.

5th step：What the storage according to goods property values to special cargo was micro-adjusted is specifically adjusted to：

A. high-risk class goods is set into secure storage areas；B. high value class goods is set into special storage area and away from discrepancy Mouthful；C. frangible class goods is placed in shelf outside；D. mobility hardly possible class goods is placed near gateway.

Present invention master has following benefit compared with prior art：

1st, the present invention proposes to be optimized by logistics center's stock's allocation management system, is with goods-shelf type storage center Skilful object is ground, is optimized for the goods yard distribution for closing storehouse, started with from the optimum management and scheduling in warehouse, thrown equipment is not increased In the case of money, the activity duration in warehouse is reduced, improve the operating efficiency in goods-shelf type warehouse.

2nd, it is of the invention by the analysis founding mathematical models to goods yard assignment problem, on the basis of clustering method, Considering warehouse historical order frequency carries out goods yard distribution, is finally finely adjusted with reference to actual demand, realizes that all orders are picked The minimum target of goods shortest path, stop frequency.

Brief description of the drawings

Fig. 1 is flow chart of the invention；

Fig. 2 is refining goods form figure in tunnel of the invention；

Fig. 3 is goods yard layout of the invention.

Specific embodiment

Below in conjunction with embodiments of the invention and accompanying drawing, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made Embodiment, belongs to the scope of protection of the invention.

As shown in figure 1, a kind of logistics center's stock's allocation method, mainly includes the following steps that：

The first step:Goods calculation of relationship degree

Goods calculation of relationship degree is concretely comprised the following steps：

Goods i and goods j connection degree is calculated, it is general by what is together occurred to goods i in same order and goods j Rate carries out goods calculation of relationship degree, and computing formula is as follows：

Wherein, p_ijFor the order number of times that goods i and goods j occurs jointly；p_i,p_jGoods i and goods j is represented respectively The number of times occurred in order；ρ_ijIt is the correlation degree of goods i and j.

Second step：Goods cluster analysis

Assuming that the species sum of warehouse goods is N, the batch total of goods polymerization is G, G≤N, the then model of cluster analysis It is as follows：

X_ig∈ { 0,1 }, X_jg∈ { 0,1 } (6)

Wherein, formula (2) is that all of goods is aggregated to mutually similar degree of association summation；Formula (3), (4) represent each goods Thing can only be assigned in a class, and all of goods is assigned in G classes according to the relevance between them；Formula (6) is represented and determined The span of plan variable.

3rd step：Distribution clusters goods to storage area

3rd step distribution cluster goods concretely comprising the following steps to storage area：

It is determined that the order of packets of cluster goods, for each class g, calculates the average order frequency of g class goods, each The center of individual storage area l is to warehouse gateway apart from d_l, determine which cluster goods g is assigned to drag by setting up One storage area：

Wherein, F_gIt is the average order frequency of g class goods；d_lFor in l inventory areas also apart from warehouse go out storage away from From；f_iIt is the order frequency of goods i；n_gIt is the species sum of g class goods；d_eFor in the l of storage space region goods yard e apart from gateway Distance；m_lIt is goods yard number in the l of storage space region；M is depot area sum；Formula (7) is used as object function, it is desired to institute in order There is the distance of class to the gateway of goods formation most short；Formula (8) represents that each storage area at most only stores a Type of Collective class Goods；Formula (9) represents same-polymeric type cargo storage a storage area；Formula (10) represents the average of each polymeric type goods Order frequency；Formula (11) represents that l inventory area centre distances warehouse goes out the distance of storage；Formula (12) represents decision variable.

4th step：Distribute goods to specific goods yard

4th step distribution cluster goods concretely comprising the following steps to storage area：

Operated in each storage region, cluster goods is carried out from high to low according to the order-placing frequency of each goods Arrangement, by order frequency cargo storage high in the goods yard near apart from gateway, completes the distribution of all class goods：

d_aOrder completes the distance of order a pickings；O is total number of orders；Formula (13) picked all orders apart from summation.

5th step：What the storage according to goods property values to special cargo was micro-adjusted is specifically adjusted to：

A. high-risk class goods is set into secure storage areas；B. high value class goods is set into special storage area and away from discrepancy Mouthful；C. frangible class goods is placed in shelf outside；D. mobility hardly possible class goods is placed near gateway.

Meanwhile, the cluster principle of knearest neighbour method in coupling system cluster analysis makes beeline into maximal correlation system Number, according to the order data for providing, calculates the incidence coefficient between any two goods first, then carries out the poly- of goods Class.The step of cluster is ingeniously as follows:

(1) species according to goods is divided into n classes, it is all kinds of between sequence arranged successively according to all kinds of order frequency sizes Row；

(2) coefficient of maximum is selected in the correlation matrix between the goods for calculating, by its corresponding goods Be polymerized to a class, then calculate it is all kinds of between coefficient correlation；

(3) cluster operation is continued, until all of goods is polymerized to a class.

Embodiment 1

Assuming that picking form of the picking personnel in warehouse in tunnel is as shown in Figure 2, from gateway to first distance in goods yard It is a unit, the distance between different goods yards are similarly a unit, and black color dots represent the point that picking is stopped in figure, at this The shelf that can pick tunnel both sides are pointed out, and disregards distance.10 orders are given in table 1, the type of merchandize of demand in order, Quantity is randomly generated.

The order detail list of table 1

For example, it is respectively goods (1,2,3,4) comprising 4 kinds of goods in order A, its corresponding quantity of goods is respectively 5, 4、3、2.Sum refers to the total quantity of the goods 1 included in 10 orders in table；What number of times was represented is that goods 1 is appeared in 10 times Number of times in order；Order goods number refers to the picking quantity included in each order；One is referred to comprising type of merchandize The type of merchandize included in individual order；10 orders include 124 goods of unit of 8 types, and goods total degree occurs and is 35 times.

The analysis of line frequency is entered to it by the situation of the order of table 1 respectively, according to the frequency that goods occurs in all orders Rate is ranked up, and is shown in Table 2.The goods frequency analysis of table 2

The order of goods is ranked up according to the size for total degree occur in all orders in table 2, when occurrence number phase Same situation：Such as goods 1 and goods 4, will be ranked up according to the size of quantity of goods.

Then, the correlation calculations between goods are carried out,

The distance definition of sample i and sample j is d_ij, use G₁、G₂,...,G_NClass is represented, most short method is referred in two class middle-ranges From the distance between two nearest samples D_pq, it represents two class G₁、G₂Distance, i.e.,：

D_pq=min d_ij i∈G_P, j ∈ G_P (14)

The step of its minimum distance clustering, is as follows:

1) the distance between sample is calculated, the matrix D of distance two-by-two is formed₍₀₎, each sample constitutes a class by itself during beginning, institute With D_pq=d_pq；

2) in matrix D₍₀₎The minimum element of middle selection is simultaneously set to D_pq, by G_p、G_qMerge as a new class G_r；

G_r={ G_p；G_q}；

3) the distance between new class for being formed is calculated

D_rk=min d_ij={ min d_ij i∈G_P, j ∈ G_k；min d_ij i∈G_q, j ∈ G_k} (15)

Distance relation according to being formed obtains new matrix D₍₁₎。

4) return to step 2), until all of element turns into a class, if minimum element in distance matrix is not only It it is one, then the class to these least members is merged.

3 are shown in Table by correlation matrix between each goods that formula is calculated

The cluster result that K means method according to example analyzes G=1 to G=8 is shown in Table 4, is then arrived according to formula (3) Formula (7) carries out the sequence of classification according to the average frequency size of every class goods, and the big goods class of frequency of occurrence in order is put Above, the ground nearer apart from gateway is placed on according to the size of the different goods frequencies of occurrences in the packet of of a sort goods Side；

From the cluster result of table 4 it can be seen that during as G=2, i.e., when 8 kinds goods are divided into the storage of two classes on shelf, according to this The distance that picking form completes the outbound task walking of all orders is most short, and stop frequency is minimum.

The path distance of table 5 is contrasted

Goods yard distribution as G=1 with G=8 is identical with the goods yard distribution according to shipment frequency in principle, from result Compare both different methods to find, main difference occurs being this reason for this in the 3rd kind of distribution of goods In the process of cluster analysis of literary grace, it is contemplated that goods 3 and the correlation of goods 1,2,4 are placed than larger together with 1,2,4 In the front end in goods yard.Because the frequency of goods 3 is lower than the frequency of goods 5, therefore clustering method ratio is distributed according to order frequency Certain picking distance can be increased during the outbound of goods 3.

Embodiment 2

The sequence information statistical analysis in company A-year, to being ranked up from big to small for order volume, 25 before selection ranking Order, such as table 6, in each order comprising type of merchandize quantity between 5 to 11, all of order contains 20 kinds of goods.

The order detail list of table 6

Algorithm design is carried out using the thought binding model of cluster for this case, is finally carried out on the basis of result Fine setting, final the results are shown in Table 7.

Tab.2-8 Result ofClustering

Result according to classification finds:

(1) 20 kind of goods is divided into 1 class and to be divided into the result of 20 classes be that consistent reasons are that goods yard distribution is all according to order What frequency size was carried out；

(2) it is gradually increased from the total picking distance of 1 class to 4 classes, because the order frequency of goods 4,6 is smaller, But itself and the big goods degree of relevancy of frequency are higher, but order for goods frequency is higher to be assigned to larger distance；

(3) when 20 kinds of goods are assigned to 14 class, total goods distance of plunderring is 426 minimum, and its stop frequency is 138 times；

(4) stop frequency in all of cluster result is respectively less than the stop frequency based on frequency, illustrates cluster result Validity.

In summary allocation result, when optimal allocation result is 14 class, its goods yard distribution layout is as shown in Figure 3.

By the classification results of table 7, in the case of G=1 and G=20, its allocation order is entirely according to History Order The result that is ranked up of frequency size, it is conceivable most herein by the think of using cluster on the basis of being distributed based on frequency Excellent result is the situation as G=14.The Comparative result of two kinds of situations is shown in.

8 two kinds of Comparative results of situation of table

By the comparative analysis of data above, the goods yard distribution layout that optimal cluster result is formed has picked all and has ordered Single goods improves 2.35% in the distance of walking, does not increase significantly；But the stop frequency of picking car is more Few equally to influence overall refining goods efficiency, stop frequency improves than the goods yard distribution based on frequency in the result of cluster 16.67%, so as to demonstrate the validity of cluster result.

The present invention is comprehensive on the basis of clustering method by the analysis founding mathematical models to goods yard assignment problem Close and consider that warehouse historical order frequency carries out goods yard distribution, be finally finely adjusted with reference to actual demand, realize all order-pickings The minimum target of shortest path, stop frequency.

Finally, above example and accompanying drawing are merely illustrative of the technical solution of the present invention and unrestricted, although by above-mentioned Embodiment is described in detail to the present invention, it is to be understood by those skilled in the art that can in form and carefully Various changes are made to it on section, without departing from claims of the present invention limited range.

Claims (6)

1. a kind of logistics center's stock's allocation method, it is characterised in that mainly include the following steps that：

The first step:Goods calculation of relationship degree；Second step：Goods cluster analysis；3rd step：Distribution clusters goods to storage area； 4th step：Distribute goods to specific goods yard；5th step：Storage according to goods property values to special cargo is micro-adjusted.

2. a kind of logistics center's stock's allocation method according to claim 1, it is characterised in that：Goods calculation of relationship degree Concretely comprise the following steps：

Goods i and goods j connection degree is calculated, is entered by the probability that goods i in same order and goods j together occur Crudely-made articles thing calculation of relationship degree, computing formula is as follows：

${\mathrm{\ρ}}_{ij}=\frac{p_{ij}}{p_i+p_j-p_{ij}}---\left(1\right)$

Wherein, p_ijFor the order number of times that goods i and goods j occurs jointly；p_i,p_jRepresent that goods i and goods j is being ordered respectively The number of times occurred in list；ρ_ijIt is the correlation degree of goods i and j.

3. a kind of logistics center's stock's allocation method according to claim 1, it is characterised in that：The tool of goods cluster analysis Body step is：

Assuming that the species sum of warehouse goods is N, the batch total of goods polymerization is G, G≤N, then the model of cluster analysis is as follows It is shown：

$\max imize\underset{g=1}{\overset{G}{\Σ}}\underset{i=1}{\overset{N-1}{\Σ}}\underset{j>i}{\overset{N}{\Σ}}{\mathrm{\ρ}}_{ij}{X}_{ig}{X}_{jg}---\left(2\right)$

$s.t:\underset{g=1}{\overset{G}{\Σ}}{X}_{ig}=1,i=1,2,3,\mathrm{...},N---\left(3\right)$

$\underset{g=1}{\overset{G}{\Σ}}{X}_{jg}=1,j=1,2,3,\mathrm{...},N---\left(4\right)$

X_ig∈ { 0,1 }, X_jg∈ { 0,1 } (6)

Wherein, formula (2) is that all of goods is aggregated to mutually similar degree of association summation；Formula (3), (4) represent each goods only Can be assigned in a class, all of goods is assigned in G classes according to the relevance between them；Formula (6) represents that decision-making becomes The span of amount.

4. a kind of logistics center's stock's allocation method according to claim 1, it is characterised in that：3rd step distribution cluster goods Thing is concretely comprised the following steps to storage area：

It is determined that the order of packets of cluster goods, for each class g, calculates average order frequency, each storage of g class goods The center of region l is deposited to warehouse gateway apart from d_l, determine which cluster goods g is assigned to drag by setting up Storage area：

$\min imize\underset{g=1}{\overset{G}{\Σ}}\underset{l=1}{\overset{M}{\Σ}}{F}_g{d}_l{Y}_{gl}---\left(7\right)$

$s.t:\underset{g=1}{\overset{G}{\Σ}}{Y}_{gl}\≤1,l=1,2,3,\mathrm{...},M---\left(8\right)$

$\underset{l=1}{\overset{G}{\Σ}}{Y}_{gl}=1,g=1,2,3,\mathrm{...},G---\left(9\right)$

$F_g=\underset{i\∈N}{\Σ}{f}_i/n_g,g=1,2,3,\mathrm{...},G---\left(10\right)$

$d_l=\underset{e\∈M_l}{\overset{g}{\mathrm{\Σ}}}{d}_e/m_l,l=1,2,3,\mathrm{...},M---\left(11\right)$

Wherein, F_gIt is the average order frequency of g class goods；d_lAlso to go out the distance of storage in l inventory areas apart from warehouse； f_iIt is the order frequency of goods i；n_gIt is the species sum of g class goods；d_eFor in the l of storage space region goods yard e apart from gateway away from From；m_lIt is goods yard number in the l of storage space region；M is depot area sum；Formula (7) is used as object function, it is desired to which institute is in stockit is available in order The distance of class to the gateway that thing is formed is most short；Formula (8) represents that each storage area at most only stores a Type of Collective class goods； Formula (9) represents same-polymeric type cargo storage a storage area；Formula (10) represents the average order of each polymeric type goods Frequency；Formula (11) represents that l inventory area centre distances warehouse goes out the distance of storage；Formula (12) represents decision variable.

5. a kind of logistics center's stock's allocation method according to claim 1, it is characterised in that：4th step distribution cluster goods Thing is concretely comprised the following steps to storage area：

Operated in each storage region, cluster goods is arranged from high to low according to the order-placing frequency of each goods Row, by order frequency cargo storage high in the goods yard near apart from gateway, complete the distribution of all class goods：

$D=\underset{a\∈O}{\Σ}{d}_a---\left(13\right)$

d_aOrder completes the distance of order a pickings；O is total number of orders；Formula (13) picked all orders apart from summation.

6. a kind of logistics center's stock's allocation method according to claim 1, it is characterised in that：5th step is according to goods characteristic What the storage of confrontation special cargo was micro-adjusted is specifically adjusted to：

A. high-risk class goods is set into secure storage areas；B. high value class goods is set into special storage area and away from gateway； C. frangible class goods is placed in shelf outside；D. mobility hardly possible class goods is placed near gateway.