CN112183938A

CN112183938A - Logistics scheduling method and device

Info

Publication number: CN112183938A
Application number: CN202010908535.4A
Authority: CN
Inventors: 赵小宝; 王远志; 高越; 吴昊
Original assignee: Zhejiang Jicheng Yunchuang Technology Co ltd; Zhejiang Geely Holding Group Co Ltd
Current assignee: Zhejiang Jicheng Yunchuang Technology Co ltd; Zhejiang Geely Holding Group Co Ltd
Priority date: 2020-09-02
Filing date: 2020-09-02
Publication date: 2021-01-05

Abstract

The invention relates to the technical field of intelligent scheduling, in particular to a logistics scheduling method and a logistics scheduling device, wherein the method comprises the following steps: acquiring a plurality of dispatchers to be dispatched and a plurality of orders to be distributed; for each dispatcher, obtaining an influence factor related to the dispatcher; determining a relationship weight between the carrier and each of the orders based on the impact factors; determining an optimal order assignment scheme by adopting a bipartite graph matching algorithm based on the relation weight, wherein each order in the optimal order assignment scheme is matched with a dispatcher; for each of the orders, assigning the order to the matched deliverer in the optimal order assignment plan. According to the logistics scheduling method, the relation weight between the dispatcher and each order is determined through the influence factors related to the dispatcher, the bipartite graph matching algorithm is adopted, the optimal order assignment scheme is determined based on the relation weight, and the matching success rate and the logistics scheduling efficiency can be improved.

Description

Logistics scheduling method and device

Technical Field

The invention relates to the technical field of intelligent scheduling, in particular to a logistics scheduling method and a logistics scheduling device.

Background

Freight logistics is a relatively complex branch in the logistics industry, and relates to main links such as trunk logistics, urban short refute logistics and the like. Where urban short barge logistics are of great interest due to the increased concern of scheduling loading. The problem of urban short refute logistics scheduling is a multi-target and multi-constraint complex matching problem considering a plurality of factors such as transportation time, vehicle loading capacity, destinations, transportation lines, car sharing at different destinations, cost control and the like.

Most of the existing scheduling methods for urban short barge logistics are manual scheduling, namely, each scheduling personnel organizes transportation capacity offline and balances decision scheduling schemes manually according to orders on the same day. In the existing logistics scheduling process, the following problems inevitably exist:

1) the logistics vehicle dispatching pressure is high, in the actual dispatching, the constraints of customer demand satisfaction, adjustable resources (vehicle quantity), loading rate and delivery time, loading combination, earliest available time, length of car sharing and the like are considered, the processing speed and precision required by the logistics vehicle dispatching pressure cause great pressure to a dispatcher, the dispatcher is difficult to give out a dispatching scheme in a short time, the dispatching efficiency is low, and the situation that the dispatching result is not optimal is easy to occur;

2) the customer satisfaction is low, the existing order distribution points are more, the distribution quantity at one time is less, the order can be sent in a full car in a long time, the order arrival time is too long, the completion rate of the order arrival distribution (OTD) is influenced, and the customer satisfaction is influenced.

Disclosure of Invention

In view of the above problems in the prior art, an object of the present invention is to provide a method and an apparatus for logistics scheduling, which can improve the logistics scheduling effect and scheduling efficiency.

In order to solve the above problem, the present invention provides a logistics scheduling method, including:

acquiring a plurality of dispatchers to be dispatched and a plurality of orders to be distributed;

for each dispatcher, obtaining an influence factor related to the dispatcher; determining a relationship weight between the carrier and each of the orders based on the impact factors;

determining an optimal order assignment scheme by adopting a bipartite graph matching algorithm based on the relation weight, wherein each order in the optimal order assignment scheme is matched with a dispatcher;

for each of the orders, assigning the order to the matched deliverer in the optimal order assignment plan.

Further, the determining the optimal order assignment scheme by using a bipartite graph matching algorithm based on the relationship weight includes:

constructing a bipartite graph model based on the relationship weights, the plurality of dispatchers and the plurality of orders;

and determining the maximum matching of the bipartite graph model by adopting a bipartite graph matching algorithm, and determining the optimal order assignment scheme according to the maximum matching.

Specifically, the building of the bipartite graph model based on the relationship weights, the dispatchers and the orders comprises:

determining a vertex according to each dispatcher, and generating a first vertex sequence of the bipartite graph model;

determining a vertex according to each order, and generating a second vertex sequence of the bipartite graph model;

determining a vertex value of each vertex in the first vertex sequence according to the relation weight;

acquiring a preset vertex value as a vertex value of each vertex in the second vertex sequence;

and determining the edges of the bipartite graph model according to the relation weights and the superscript values of each vertex in the first vertex sequence and the second vertex sequence.

Further, the determining the maximum matching of the bipartite graph model by using a bipartite graph matching algorithm, and the determining the optimal order assignment scheme according to the maximum matching includes:

determining an augmentation path for each vertex in the first vertex sequence in sequence, and updating a matching subgraph of the bipartite graph model according to the augmentation path until the maximum matching of the bipartite graph model is obtained;

and matching the deliverers corresponding to each edge in the maximum matching with the orders to obtain the optimal order assignment scheme.

Further, the determining an augmentation path for each vertex in the first vertex sequence in sequence, and updating a matching subgraph of the bipartite graph model according to the augmentation path until obtaining the maximum matching of the bipartite graph model includes:

determining an augmentation path for the current vertex in the first vertex sequence, and updating the matching subgraph of the bipartite graph model according to the augmentation path, wherein the steps of:

acquiring a current bipartite graph model and a current matching subgraph;

searching an augmentation path for the current vertex based on the current bipartite graph model, and judging whether the augmentation path is found;

if an augmentation path is found, updating the current matching subgraph according to the augmentation path;

if an augmentation path is not found, determining a newly added edge of the current bipartite graph model according to the relation weight and the vertex mark value of each vertex in the current bipartite graph model, and updating the current bipartite graph model according to the newly added edge; and determining an augmentation path of the current vertex based on the updated current bipartite graph model, and updating the current matching subgraph according to the augmentation path.

Further, the obtaining, for each of the dispatchers, an impact factor associated with the dispatcher includes:

for each dispatcher, obtaining historical scheduling data related to the dispatcher;

determining impact factors associated with the dispatcher based on the historical scheduling data.

Further, the determining a relationship weight between the carrier and each of the orders based on the impact factors includes:

determining an influence coefficient corresponding to the influence factor according to the information of the order and the information of the dispatcher for each order;

and calculating the relation weight between the deliverer and the order according to the influence factors and the influence coefficients corresponding to the influence factors.

Further, the influence factors comprise one or more of historical data of a dispatching store where the dispatcher is located, the order taking rate of the dispatcher, the total dispatching completion occupation ratio of the dispatcher, the occupation ratio of the dispatcher in the interval from the nearest goods taking address, the occupation ratio of the order capacity in the interval of the vehicle capacity and the satisfaction degree of the customer to the dispatcher.

Further, the method further comprises:

acquiring the position information of the dispatcher;

and screening the dispatchers according to the position information of the dispatchers, and determining a plurality of dispatchers in a preset area.

Another aspect of the present invention provides a logistics scheduling apparatus, including:

the system comprises an acquisition module, a scheduling module and a scheduling module, wherein the acquisition module is used for acquiring a plurality of dispatchers to be scheduled and a plurality of orders to be distributed;

a first determining module, configured to obtain, for each of the dispatchers, an impact factor related to the dispatcher; determining a relationship weight between the carrier and each of the orders based on the impact factors;

a second determining module, configured to determine an optimal order assignment scheme by using a bipartite graph matching algorithm based on the relationship weight, where each order in the optimal order assignment scheme matches one dispatcher;

and the scheduling module is used for assigning the orders to the matched dispatchers in the optimal order assignment scheme aiming at each order.

Due to the technical scheme, the invention has the following beneficial effects:

(1) according to the logistics scheduling method, the relation weight between the deliverer and each order is determined through the influence factors related to the deliverer, the bipartite graph matching algorithm is adopted, and the optimal order assignment scheme is determined based on the relation weight, so that the matching success rate and the logistics scheduling efficiency can be improved, the transportation cost of logistics delivery is reduced, and the delivery timeliness is improved.

(2) According to the logistics scheduling method, the influence factors and the influence coefficients corresponding to the influence factors are dynamically determined by using historical scheduling data, so that a logistics scheduling scheme adaptive to customer requirements can be determined, the logistics scheduling service level is improved, and the customer satisfaction is improved.

Drawings

In order to more clearly illustrate the technical solution of the present invention, the drawings used in the description of the embodiment or the prior art will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.

Fig. 1 is a flowchart of a logistics scheduling method according to an embodiment of the invention;

FIG. 2 is a schematic structural diagram of a bipartite graph model according to an embodiment of the invention;

FIG. 3 is a schematic diagram of a matching subgraph provided in one example of the invention;

FIG. 4 is a schematic diagram of a matching subgraph provided in another example of the invention;

FIG. 5 is a schematic diagram of a matching subgraph provided in another example of the invention;

FIG. 6 is a schematic diagram of a matching subgraph provided in another example of the invention;

fig. 7 is a schematic structural diagram of a logistics scheduling apparatus according to an embodiment of the present invention.

Wherein reference numerals in the figures correspond to: 710-acquisition module, 720-first determination module, 730-second determination module, 740-scheduling module.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.

Referring to the specification and fig. 1, a flow of a logistics scheduling method according to an embodiment of the invention is shown. As shown in fig. 1, the method may include:

s110: a plurality of dispatchers to be dispatched and a plurality of orders to be distributed are obtained.

In the embodiment of the invention, the plurality of dispatchers to be dispatched can be dispatchers of the same dispatch store or different dispatch stores, and the information of the dispatchers can comprise position information of the dispatchers, vehicle transportation capacity information, license plate numbers and extensible label information (whether to pull refrigerated goods or not and whether to allocate carriers or not); the order to be distributed refers to an order which needs to be dispatched within a period of time, and the information of the order may include an order number, position information of the order, cargo weight, cargo volume, and extensible label information (whether to carry, whether to aquatic product, whether to throw cargo, etc.).

In one possible embodiment, the method may further include:

acquiring the position information of the dispatcher;

Specifically, the dispatchers may be screened according to their location information and their transportation capacity information, and the dispatchers meeting certain conditions are selected for order distribution, for example, the dispatchers located in the same city where the orders are located and having a transportation capacity greater than a preset threshold value may be selected.

S120: for each dispatcher, obtaining an influence factor related to the dispatcher; determining a relationship weight between the carrier and each of the orders based on the impact factors.

In the embodiment of the invention, the influence factors can comprise one or more of historical data of a dispatching store where the dispatcher is located, the order taking rate of the dispatcher, the total dispatching completion occupation ratio of the dispatcher, the occupation ratio of the dispatcher in the section of the nearest goods taking address, the occupation ratio of the order capacity in the section of the vehicle capacity and the satisfaction degree of the customer on the dispatcher. The impact factor may be determined based on information such as the dispatcher's information, the order, and historical scheduling data associated with the dispatcher.

In one possible embodiment, the obtaining, for each of the dispatchers, an impact factor associated with the dispatcher may include:

Specifically, the historical scheduling data may be statistically analyzed to calculate the impact factor. For example, the order-receiving rate of the dispatcher can be calculated according to the historical scheduling data and the order-receiving data of the dispatcher, the satisfaction degree of the customer on the dispatcher can be determined according to the evaluation level of the customer on the dispatcher, and the like. Influence factors related to the deliverers are dynamically determined based on massive historical scheduling data, so that optimal relation weight can be dynamically determined, and a scheduling scheme adaptive to customer requirements is further constructed. The embodiment of the invention does not limit the analysis method of the historical scheduling data.

In one possible embodiment, the determining a relationship weight between the carrier and each of the orders based on the impact factors includes:

Specifically, the association degree between the order and the deliverer may be determined according to information such as a matching degree between a delivery address of the order and a delivery range of the deliverer, evaluation of a customer of the order on the deliverer, and a receiving rate of the deliverer, so as to determine an influence coefficient corresponding to each influence factor. And determining a relation weight between the deliverer and the order according to the influence factor and the influence coefficient corresponding to the influence factor, wherein the relation weight can reflect the value rate of the deliverer.

It should be noted that, the method provided in the embodiment of the present invention may also intelligently adjust the influence coefficient corresponding to the influence factor according to historical scheduling data, so as to determine the relationship weight data suitable for the actual situation.

S130: and determining an optimal order assignment scheme by adopting a bipartite graph matching algorithm based on the relation weight, wherein each order in the optimal order assignment scheme is matched with one dispatcher.

The bipartite graph matching algorithm is also known as KM (Kuhn-Munkres) matching algorithm. Based on the relationship weight, the optimal order assignment scheme can be found by adopting a KM matching algorithm, wherein the weight sum in the optimal order assignment scheme is the largest. It can be understood that the KM matching algorithm can find the optimal match (i.e. the best order assignment scheme in the embodiment of the present invention) with better time complexity.

In one possible embodiment, the determining the optimal order assignment scheme using a bipartite graph matching algorithm based on the relationship weights may include:

step 1: constructing a bipartite graph model based on the relationship weights, the plurality of dispatchers and the plurality of orders.

In an embodiment of the present invention, the building a bipartite graph model based on the relationship weight, the plurality of dispatchers, and the plurality of orders may include:

Specifically, a maximum value of each relationship weight corresponding to the dispatcher may be obtained as a vertex value of a vertex corresponding to the dispatcher, and the vertex value of each vertex in the second vertex sequence may be set to 0. The vertex value of the vertex i in the first vertex sequence may be denoted as label (i), i ═ 1, …, m, the vertex value of the vertex j in the second vertex sequence may be denoted as label (j), j ═ 1, …, n, the relationship weight between the dispatcher corresponding to the vertex i in the first vertex sequence and the order corresponding to the vertex j in the second vertex sequence may be denoted as weight (i, j), i ═ 1, …, m, j ═ 1, …, n, and the edge between the vertex i and the vertex j of the weight (i, j) ═ label (i) + label (j) may be used as the edge of the bipartite graph model, so that the bipartite graph obtained may be referred to as a relative subgraph.

In one example, assume that the dispatcher is 5: x0, X1, X2, X3, and X4, with 5 orders: y0, Y1, Y2, Y3 and Y4. The weight of the relationship between the dispatcher and the order is shown in table 1:

TABLE 1 relationship weight Table

	Y0	Y1	Y2	Y3	Y4
						X0
	3	4	6	4	9
						X1	6	4	4	3	8
X2	7	6	3	4	2
						X3	6	3	2	2	5
X4	8	4	5	4	7

By the above method, a bipartite graph model as shown in fig. 2 can be constructed, and it can be determined that vertex values of vertices X0, X1, X2, X3, and X4 are 9, 8, 7, 6, and 8, respectively, and vertex values of vertices Y0, Y1, Y2, Y3, and Y4 are 0, and 0, respectively; the edges of the bipartite graph model may be determined to include X0Y4, X1Y4, X2Y0, X3Y0, and X4Y 0.

Step 2: and determining the maximum matching of the bipartite graph model by adopting a bipartite graph matching algorithm, and determining the optimal order assignment scheme according to the maximum matching.

In an embodiment of the present invention, the determining the maximum matching of the bipartite graph model by using a bipartite graph matching algorithm, and the determining the optimal order assignment scheme according to the maximum matching may include:

Specifically, the sequentially determining an augmentation path for each vertex in the first vertex sequence, and updating the matching subgraph of the bipartite graph model according to the augmentation path until the maximum matching of the bipartite graph model is obtained may include:

acquiring a current bipartite graph model and a current matching subgraph;

In an example, with reference to fig. 3 to 6 in the specification, the maximum matching of the bipartite graph model may be obtained by finding an augmentation path for vertices X0, X1, X2, X3 and X4 in sequence, starting from vertex X0 of the bipartite graph model shown in fig. 2. For example, for vertex X0, X0Y4 may be found, and a matching subgraph as shown in FIG. 3 may be determined; for vertex X1, if no augmented path can be found, the bipartite graph model needs to be updated and the topmark needs to be adjusted.

The method for updating the bipartite graph model may comprise: for each vertex on the searched path, assuming that the X vertex set on the path is S, Y vertex set as T, calculating d (Xi, Yj) ═ label (Xi) + label (Yj) -weight (Xi, Yj) for all points Xi in S and Yj not in T; adding an edge XiYj formed by a vertex Xi and a vertex Yj corresponding to d min (Xi, Yj) into the bipartite graph model to obtain an updated bipartite graph model; and d is subtracted from the X mark post in the S set and added into the mark post of the Y in the T set to obtain an updated topmark value.

In the above example, three nodes X1, Y4, X0 are accessed, then the vertex set S includes X0 and X1, the vertex set T includes Y4, so d can be calculated as 2, and the corresponding edge is X1Y0 (from the greedy edge selection point of view, we can choose a new edge for X0 and discard the matching edge in the original matching subgraph, or choose a new edge for X1 and discard the matching edge in the original matching subgraph, because we cannot choose X0Y4 and X1Y4 at the same time, because this is an illegal match, and the meaning of choosing an edge consisting of the vertex corresponding to d mind (Xi, Yj) is that we choose a new edge, which will be added to the matching subgraph so that the match is legal, choose a matching subgraph formed by this edge, which will be less than the original matching subgraph formed by adding this illegal matching edge (if it is a legal, its weight and its weight will be the smallest and the smallest weight), i.e., the sum of the weights is largest in a legal match). At this time, as shown in fig. 4, an augmentation path is searched for X1 again in the updated bipartite graph, resulting in X1Y 0.

Specifically, the same method is adopted to find an augmented path for X2, and the augmented path cannot be found, and the bipartite graph model also needs to be updated to adjust the topmark. According to the above method, the sides X0Y2, X2Y1 may be added to the bipartite graph model, and d is 1; searching an augmentation path for the X2 again on the new bipartite graph model, and if a depth-first method is adopted and referring to FIG. 5, obtaining the augmentation path of X2Y0- > Y0X1- > X1Y4- > Y4X0- > X0Y2, and negating the matching result at the moment, obtaining three matches of X2Y0, X1Y4 and X0Y 2; if the breadth first method is employed, referring to FIG. 6, the matching results obtained are X0Y4, X1Y0, X2Y 1. The same method can be used to find the augmented paths for X3 and X4, resulting in the maximum match of the bipartite graph model.

In the embodiment of the present invention, the maximum matching is found to ensure that the number of matched dispatchers and orders in the maximum order assignment scheme is the largest. After the maximum matching result is determined, the dispatcher and the order corresponding to each edge in the maximum matching can be matched to obtain the optimal order assignment scheme. For example, assuming that the maximum matching includes X0Y4, the dispatcher corresponding to vertex X0 and the order corresponding to vertex Y4 may be matched to obtain a matching result of the best order assignment scheme.

S140: for each of the orders, assigning the order to the matched deliverer in the optimal order assignment plan.

In summary, the logistics scheduling method of the invention has the following beneficial effects:

Referring to fig. 7 in conjunction with the description, a logistics scheduling apparatus according to an embodiment of the invention is shown, and as shown in fig. 7, the logistics scheduling apparatus may include:

an obtaining module 710, configured to obtain a plurality of dispatchers to be scheduled and a plurality of orders to be distributed;

a first determining module 720, configured to obtain, for each of the dispatchers, an impact factor related to the dispatcher; determining a relationship weight between the carrier and each of the orders based on the impact factors;

a second determining module 730, configured to determine an optimal order assignment scheme by using a bipartite graph matching algorithm based on the relationship weight, where each order in the optimal order assignment scheme matches one dispatcher;

a scheduling module 740, configured to assign the order to the matched delivery person in the optimal order assignment plan for each order.

In one possible embodiment, the second determining module 730 may include:

the construction unit is used for constructing a bipartite graph model based on the relation weight, the dispatchers and the orders;

and the determining unit is used for determining the maximum matching of the bipartite graph model by adopting a KM algorithm and determining the optimal order assignment scheme according to the maximum matching.

The foregoing description has disclosed fully preferred embodiments of the present invention. It should be noted that those skilled in the art can make modifications to the embodiments of the present invention without departing from the scope of the appended claims. Accordingly, the scope of the appended claims is not to be limited to the specific embodiments described above.

Claims

1. A logistics scheduling method, comprising:

2. The method of claim 1, wherein determining an optimal order assignment scheme using a bipartite graph matching algorithm based on the relationship weights comprises:

3. The method of claim 2, wherein said constructing a bipartite graph model based on the relationship weights, the plurality of dispatchers and the plurality of orders comprises:

4. The method of claim 3, wherein said employing a bipartite matching algorithm to determine a maximum match for the bipartite model, and wherein determining the optimal order assignment plan based on the maximum match comprises:

5. The method of claim 4, wherein the sequentially determining an augmented path for each vertex in the first sequence of vertices, and updating the matching subgraph of the bipartite graph model according to the augmented path until a maximum match of the bipartite graph model is obtained comprises:

acquiring a current bipartite graph model and a current matching subgraph;

6. The method of claim 1, wherein said obtaining, for each of said dispatchers, an impact factor associated with said dispatcher comprises:

7. The method of claim 1, wherein determining a relationship weight between the carrier and each of the orders based on the impact factors comprises:

8. The method of claim 1, 6 or 7, wherein the impact factors include one or more of historical data of a dispatch store where the dispatcher is located, a rate of order pickup by the dispatcher, a total dispatch completion percentage of the dispatcher, a percentage of the dispatcher in the interval from the nearest pickup address, a percentage of order capacity in the interval from vehicle capacity, and customer satisfaction with the dispatcher.

9. The method according to claim 1 or 2, characterized in that the method further comprises:

acquiring the position information of the dispatcher;

10. A logistics scheduling apparatus, comprising: