Maintenance dispatch method, device, equipment and storage medium based on hierarchical clustering

Abstract

The present invention discloses a maintenance dispatching method, device, equipment and storage medium based on hierarchical clustering. The method includes: dividing the maintenance area into multiple sub-areas, and periodically performing the following steps: obtaining the order information of the current undispatched order; obtaining the number of shared order pairs between any two sub-areas based on the order information; constructing a spatial sharing graph, taking the sub-areas as vertices, and taking the number of shared order pairs of the two vertices as the weight of the connecting edge of the two vertices; clustering in the spatial sharing graph to form a sub-graph set, and in the clustering process, the cumulative sum of the weights of the connecting edges in the sub-graph set is as large as possible and the weight difference of the connecting edges in the sub-graph set is as small as possible; obtaining the location information of the maintenance personnel, and sending the order information whose service address is located in the sub-graph where the maintenance personnel is located to the maintenance personnel terminal. Constructing a spatial sharing graph is convenient for intuitively judging the hot and cold sub-areas of orders, dynamically adjusting the order information dispatched to the maintenance personnel, and reducing the time spent on the road.

Description

Technical Field

[0001] The present invention relates to the technical field of maintenance personnel scheduling, and in particular to a maintenance dispatching method, device, equipment and storage medium based on hierarchical clustering. Background Art

[0002] In home appliance repair services, orders are usually sent to home appliance repair engineers through the backend system, and the maintenance personnel provide door-to-door service based on the user information on the repair order.

[0003] Home appliance maintenance engineers have their own fixed maintenance scope, and generally only maintenance orders within the fixed maintenance scope are dispatched to engineers. This will cause an imbalance in manpower allocation when hot and cold areas of orders appear. For example, when a hot area of ​​orders appears at a certain time, maintenance engineers around the hot area cannot go to the hot area for maintenance in time due to engineer scheduling problems, or engineers around the hot area cannot receive corresponding orders due to the static division of the service scope, and maintenance engineers in the cold area of ​​orders are idle due to the small number of orders.

[0004] In the prior art, in order to solve the problem of unbalanced manpower allocation, the manpower in the cold zone is dispatched to the hot zone. However, when the cold zone and the hot zone are far apart, the engineers will spend a lot of time on the road. Summary of the invention

[0005] The present invention aims to at least solve the technical problems existing in the prior art, and in particular innovatively proposes a maintenance dispatching method, device, equipment and storage medium based on hierarchical clustering.

[0006] In order to achieve the above-mentioned purpose of the present invention, according to the first aspect of the present invention, the present invention provides a maintenance dispatching method based on hierarchical clustering, including: dividing a maintenance area into multiple sub-areas; obtaining order information that is not currently dispatched, the order information including a service address and a maintenance category; obtaining the number of shared order pairs between any two sub-areas based on the order information, the shared order pairs consisting of orders belonging to two different sub-areas but with the same maintenance category; constructing a spatial sharing graph, including: taking the sub-areas as vertices, connecting two vertices with shared order pairs, and using the number of shared order pairs of the two vertices as the weight of the connecting edge of the two vertices; clustering in the spatial sharing graph to form a sub-graph set, and in the clustering process, aiming to make the cumulative sum of the weights of the connecting edges in the sub-graph set as large as possible and the weight difference of the connecting edges in the sub-graph set as small as possible; obtaining the location information of the maintenance personnel, and sending the order information whose service address is located in the sub-graph where the maintenance personnel is located to the maintenance personnel terminal.

[0007] In order to achieve the above-mentioned purpose of the present invention, according to the second aspect of the present invention, the present invention provides a maintenance dispatching device based on hierarchical clustering, including: a division module, dividing the maintenance area into multiple sub-areas; an order information acquisition module, acquiring the order information of the currently undispatched order, the order information including the service address and the maintenance category; a shared order pair quantity acquisition module, acquiring the quantity of shared order pairs between any two sub-areas based on the order information, the shared order pairs consisting of orders belonging to two different sub-areas but with the same maintenance category; a spatial shared graph construction module, constructing a spatial shared graph, including: taking the sub-areas as vertices, connecting two vertices with shared order pairs, and taking the quantity of shared order pairs of the two vertices as the weight of the connecting edge of the two vertices; a subgraph set acquisition module, clustering in the spatial shared graph to form a subgraph set, and in the clustering process, taking the cumulative sum of the weights of the connecting edges in the subgraph set as large as possible and the weight difference of the connecting edges in the subgraph set as small as possible; a dispatching module, acquiring the location information of the maintenance personnel, and sending the order information whose service address is located in the subgraph where the maintenance personnel is located to the maintenance personnel terminal.

[0008] In order to achieve the above-mentioned purpose of the present invention, according to the third aspect of the present invention, the present invention provides an order dispatching device, including an order receiving module and a processing platform, the order receiving module receives order information issued by the user; the processing platform receives the order information from the order receiving module, and sends the order information to the maintenance personnel terminal according to the maintenance dispatching method based on hierarchical clustering described in the first aspect of the present invention.

[0009] In order to achieve the above-mentioned purpose of the present invention, according to the fourth aspect of the present invention, the present invention provides a storage medium storing a program, wherein the program enables a computer to execute the steps of the maintenance dispatch method based on hierarchical clustering described in the first aspect of the present invention.

[0010] To sum up, due to the adoption of the above-mentioned technical scheme, the beneficial effects of the present invention are: taking sub-regions as vertices and the number of shared order pairs between sub-regions as the weights of the sub-region connecting edges, a spatial sharing graph is constructed, which is convenient for intuitively judging the hot and cold sub-regions of orders, and is convenient for dynamically adjusting the order information sent to the maintenance personnel, as well as dynamically adjusting the number of maintenance personnel sent to the area, so that the maintenance personnel can send orders within their sub-graph. On the one hand, it can reduce the time spent by the maintenance personnel on the door-to-door service journey. On the other hand, since the shared order pairs in the sub-graph are relatively similar, the maintenance problems encountered by users are relatively similar, then the efficiency of the maintenance personnel in repeatedly handling similar problems will be improved, and more orders can be processed. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flowchart of a maintenance dispatching method based on hierarchical clustering in a specific implementation manner of the present invention;

[0012] Figure 2 is a schematic diagram of a maintenance area divided into sub-areas in a specific embodiment of the present invention;

[0013] Figure 3 yes Figure 2 The corresponding space sharing graph;

[0014] Figure 4 yes Figure 3 Schematic diagram of the maximum spanning tree structure of the space sharing graph in the middle;

[0015] Figure 5 yes Figure 4 Schematic diagram of the subgraph set formed by the maximum spanning tree clustering;

[0016] Figure 6 is with Figure 5 Schematic diagram of the corresponding two core service areas. DETAILED DESCRIPTION

[0017] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0018] In the description of the present invention, it is necessary to understand that the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0019] In the description of the present invention, unless otherwise specified and limited, it should be noted that the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the internal connection between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0020] Since the time when a maintenance order is generated and the service location are uncontrollable, maintenance personnel may be sent to a remote location for service. As a result, the maintenance personnel have a wide service range but a relatively short effective working time. Because the door-to-door journey takes more time, it is necessary to dynamically adjust the door-to-door service range of engineers according to the number of orders to improve the efficiency of maintenance personnel. To this end, the present invention discloses a maintenance dispatching method based on hierarchical clustering. In a preferred embodiment, as Figure 1 As shown, the method includes:

[0021] Step 101, divide the maintenance area into multiple sub-areas, the maintenance area is preferably but not limited to the entire urban area or a certain district in the urban area, such as the Xihu District of Hangzhou. Figure 2 As shown, a schematic diagram of the maintenance area after being divided into multiple sub-areas, and the sub-areas are interconnected. The sub-areas are preferably, but not limited to, regular shapes or irregular images, and the regular shapes are preferably, but not limited to, regular polygons, such as regular hexagons. For example, the entire city of Chongqing is divided into regular hexagons with a radius of 500 meters, and the radius of the hexagon is the distance from the center point to each endpoint, which is 500 meters. For a specific dispatch area, such as Xihu District in Hangzhou, this method can fill the administrative area with regular hexagons, and each filled regular hexagon serves as the smallest unit of the dispatch service range. Preferably, the above-mentioned sub-area division operation is performed on an electronic map so that the geographical location information of each sub-area can be easily known.

[0022] Steps 102 to 106 are executed periodically, and the execution period is preferably, but not limited to, 1 hour, 2 hours, or 3 hours.

[0023] Step 102, obtain the order information of the current unassigned order, the order information includes the service address and the repair category; the repair category preferably includes but is not limited to refrigerator repair, TV repair, washing machine repair, water heater repair, gas stove repair, air conditioner repair. The service address is the repair address in the order information.

[0024] Step 103, based on the order information, obtain the number of shared order pairs between any two sub-areas. A shared order pair consists of orders belonging to two different sub-areas but with the same maintenance category. For example, if the maintenance category of two orders located in two sub-areas is refrigerator repair, then these two orders form a shared order pair. The number of shared order pairs is linearly related to the number of orders. Therefore, using the number of shared order pairs as a weight has the same function as the number of orders, such as determining the hot and cold zones. In addition, using shared order pairs can more intuitively determine whether two orders are shared and whether they can be served by the same maintenance personnel.

[0025] Step 104, constructing a space sharing graph, including: taking the sub-region as a vertex, connecting two vertices with shared order pairs, and using the number of shared order pairs of the two vertices as the weight of the connecting edge between the two vertices; the obtained space sharing graph is as follows Figure 3 As shown, Figure 3 The vertices in Figure 2 The sub-regions in correspondence one to one, such as Figure 3 As shown, two vertices with shared order pairs are connected, and two vertices without shared order pairs are not connected. The value on the connecting edge represents the weight of the connecting edge, that is, the number of shared order pairs between the two vertices.

[0026] Step 105, clustering the spatial shared graph to form sub-graph sets, and the goal of the clustering process is to make the cumulative sum of the weights of the connecting edges in the sub-graph sets as large as possible and the weight difference of the connecting edges in the sub-graph sets as small as possible; the sub-graph sets are as follows: Figure 5 As shown, all vertices that are continuously connected are regarded as a subgraph. Obviously, Figure 5 There are two subgraphs in , one is the subgraph consisting of v1, v1, v3, v5, v8, and the other is the subgraph consisting of v4, v6, v7.

[0027] Preferably, in the clustering process, the goal is to make the cumulative sum of the weights of the connecting edges in the subgraph set as large as possible and the weight difference of the connecting edges in the subgraph set as small as possible. The following objective function can be established:

[0028]

[0029] Among them, G represents the spatial shared graph, S(G) represents the set of all subgraphs in the spatial shared graph, w(e) represents the weight of the connecting edge e in the shared spatial graph, C is the subgraph set obtained by clustering, G i represents the i-th subgraph in the subgraph set C. θ is the weight variance threshold of the connecting edge, which can ensure that the vertices in the connected subgraph have similar weight values. σ(G i ) represents the weight difference of the connecting edges in the i-th subgraph, which is calculated as follows:

[0030] |G i | represents the number of vertices in the i-th subgraph. φ represents the empty set.

[0031] Step 106, obtain the location information of the maintenance personnel, and send the order information whose service address is in the sub-graph where the maintenance personnel is located to the maintenance personnel terminal. Only the orders in the sub-graph are dispatched to the maintenance personnel. Since the sub-regions in the sub-graph are connected, the travel time of the maintenance personnel is less than the travel time when the orders are randomly assigned, and the number of shared order pairs in the region is relatively large, that is, the order similarity is relatively large, and the maintenance personnel are likely to face most orders of similar categories, which can improve the maintenance proficiency and efficiency of the maintenance personnel. In addition, the maintenance personnel are no longer fixed in one sub-region as before. In this way, when the maintenance personnel are located at the border of the sub-region, the dispatching of orders is more flexible, and they can go to the place adjacent to the sub-region in the sub-graph for maintenance.

[0032] In a preferred embodiment, clustering in a spatial shared graph to form a sub-graph set specifically includes:

[0033] The maximum spanning tree of the spatial sharing graph is obtained, preferably but not limited to Prim's algorithm. In the maximum spanning tree, a subgraph set is formed by clustering with the goal of minimizing the weight difference of the connecting edges as much as possible, and multiple vertices and connecting edges between multiple vertices in the subgraph set that are continuously connected are regarded as a subgraph. Preferably, but not limited to, the variance is used to measure the weight difference of the connecting edges in the maximum spanning tree.

[0034] In a preferred embodiment, clustering a subgraph set in a maximum spanning tree with the goal of minimizing the weight difference of connecting edges comprises:

[0035] Step A, set a subgraph variable C, initialize the subgraph variable C to the maximum spanning tree T, obtain the connection edge set E of the subgraph variable C, and set the variance threshold θ;

[0036] Step B, calculate the weight variance σ(E) of the connected edges in the connected edge set E. If σ(E)≤θ, output the subgraph set variable C. If σ(E)>θ, delete the connected edge in the subgraph set variable C that can reduce the weight variance the most, and enter step C.

[0037] Step C, return to step B.

[0038] In this embodiment, the connection edge e that can best reduce the weight variance in the subgraph set variable C can be found by the following formula: * , argmax(f(x)) is the variable point x (or the set of variable points x) corresponding to the maximum value of f(x), which means selecting a connection edge to maximize the value of σ(E)-σ(E / {e}), where σ(E / {e} represents the weight variance of the connection edge set after excluding the selected connection edge. Recursively delete the connection edge that can best reduce the weight variance until the edge weight variance in each cluster is less than the variance threshold θ, and the final subgraph set C is obtained.

[0039] In an application scenario of this embodiment, the variance threshold θ is 3. First, the maximum spanning tree of the space sharing graph is obtained, such as Figure 4 The variance of the edge weights of the maximum spanning tree T is calculated to be 5.95, which is greater than the deviation threshold θ. Then find the edge e that can reduce the variance the most. * =(v5,v7) and delete e * ,get Figure 5 The subtree set T' is shown. The thresholds of the two subtrees of T' are calculated to be 2.91 and 0.5, respectively, both lower than the variance threshold θ, thus obtaining two clusters of hierarchical clustering, i.e., two subgraphs. Figure 6 The two service areas finally presented on the map are shown. The maintenance platform can dynamically adjust the number of engineers assigned to the new service area based on the number of orders of this category in the new service area, and allow engineers to operate only in the current area, thereby reducing the time spent by engineers on door-to-door service.

[0040] In a preferred embodiment, in order to achieve optimal dynamic scheduling of maintenance personnel and reduce the travel time of maintenance personnel, the location information of the maintenance personnel is obtained, and order information is sent to the maintenance personnel terminal based on the location information of the maintenance personnel, the shared order pair density of the subgraph where the maintenance personnel is located, and the shared order pair density of the adjacent subgraph of the subgraph where the maintenance personnel is located. The shared order pair density refers to the ratio of the total number of shared order pairs in the subgraph to the number of vertices of the subgraph.

[0041] In this embodiment, in order to achieve the optimal dynamic scheduling of maintenance personnel and reduce the travel time of maintenance personnel, the above-mentioned order information is sent to the maintenance personnel terminal in combination with the location information of the maintenance personnel, the density of shared order pairs of the subgraph where the maintenance personnel is located, and the density of shared order pairs of the adjacent subgraphs of the subgraph where the maintenance personnel is located, specifically including:

[0042] If the number of maintenance personnel in the subgraph where the maintenance personnel is located matches the density of shared order pairs of the subgraph, or the number of maintenance personnel in the subgraph where the maintenance personnel is located is less than the number of personnel matched by the density of shared order pairs of the subgraph, the order information whose service address is in the subgraph where the maintenance personnel is located is sent to the maintenance personnel terminal. This indicates that the maintenance personnel in the subgraph do not need to be dispatched to other subgraphs. Obviously, the greater the density of shared order pairs in the subgraph, the greater the number of maintenance personnel required. A matching ratio can be set based on experience, and the shared order pair density of the subgraph is multiplied by the matching ratio to obtain the number of matched personnel. Further preferably, in order to reduce the travel time of the maintenance personnel, the order information whose distance from the service address to the current location of the maintenance personnel is within a preset distance threshold is sent to the terminal device of the maintenance personnel; or, the order information in the sub-area where the current location of the maintenance personnel is located is sent to the terminal device of the maintenance personnel. Preferably, the preset distance threshold is the distance traveled by the maintenance personnel within 10 minutes or 15 minutes.

[0043] If the number of maintenance personnel in the subgraph where the maintenance personnel is located is greater than the number of personnel matched by the density of shared order pairs of the subgraph, that is, there are insufficient maintenance personnel in the current subgraph, the order information whose service address is located in the first subgraph is sent to the maintenance personnel terminal, and the first subgraph is an adjacent subgraph of the subgraph where the maintenance personnel is located and the number of maintenance personnel in the first subgraph is less than the number of personnel matched by the density of shared order pairs of the first subgraph. Further preferably, in order to reduce the travel time of the maintenance personnel, the order information whose service address is within a preset distance threshold from the current location of the maintenance personnel is sent to the terminal device of the maintenance personnel.

[0044] In a preferred embodiment, in order to achieve optimal dynamic scheduling of maintenance personnel, the following steps are also included: obtaining the shared order pair density of each subgraph, the greater the shared order pair density, the greater the required maintenance personnel density, and the smaller the shared order pair density, the smaller the required maintenance personnel density; the shared order pair density refers to the ratio of the number of shared order pairs in the subgraph to the number of vertices in the subgraph; the maintenance personnel density refers to the ratio of the number of maintenance personnel dispatched to the subgraph to the number of vertices in the subgraph.

[0045] The present invention also discloses a maintenance dispatching device based on hierarchical clustering, which, in a preferred embodiment, includes: a division module, which divides a maintenance area into multiple sub-areas; an order information acquisition module, which acquires order information of currently undispatched orders, the order information including a service address and a maintenance category; a shared order pair quantity acquisition module, which acquires the quantity of shared order pairs between any two sub-areas based on the order information, the shared order pairs consisting of orders belonging to two different sub-areas but with the same maintenance category; a spatial shared graph construction module, which constructs a spatial shared graph, including: taking the sub-areas as vertices, connecting two vertices with shared order pairs, and taking the quantity of shared order pairs of the two vertices as the weight of the connecting edge of the two vertices; a sub-graph set acquisition module, which clusters in the spatial shared graph to form a sub-graph set, and in the clustering process, aims to make the cumulative sum of the weights of the connecting edges in the sub-graph set as large as possible and the weight difference of the connecting edges in the sub-graph set as small as possible; and a dispatching module, which acquires the location information of the maintenance personnel, and sends the order information whose service address is located in the sub-graph where the maintenance personnel is located to the maintenance personnel terminal.

[0046] The present invention also discloses a dispatching device, which in a preferred embodiment includes an order receiving module and a processing platform, wherein the order receiving module receives order information issued by a user; the processing platform receives the order information from the order receiving module, and sends the order information to a maintenance personnel terminal according to the above-mentioned maintenance dispatching method based on hierarchical clustering of the present invention.

[0047] The present invention also discloses a storage medium storing a program, wherein the program enables a computer to execute the steps of the maintenance dispatching method based on hierarchical clustering of the present invention.

[0048] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0049] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A maintenance dispatching method based on hierarchical clustering, characterized in that: include: Divide the maintenance area into multiple sub-areas and periodically perform the following steps: Obtaining the order information of the currently undispatched orders, the order information including the service address and the repair category; Based on the order information, the number of shared order pairs between any two sub-areas is obtained, where the shared order pairs are composed of orders belonging to two different sub-areas but having the same maintenance category; Constructing a spatial sharing graph, including: taking the sub-area as a vertex, connecting two vertices having a shared order pair, and using the number of shared order pairs of the two vertices as the weight of the connecting edge of the two vertices; In the spatial shared graph, subgraphs are formed by clustering. The goal of the clustering process is to make the cumulative sum of the weights of the connecting edges in the subgraph as large as possible and the weight difference of the connecting edges in the subgraph as small as possible. The objective function of the clustering process is: Among them, G represents the spatial shared graph, S(G) represents the set of all subgraphs in the spatial shared graph, w(e) represents the weight of the connecting edge e in the shared spatial graph, C is the subgraph set obtained by clustering, G i represents the i-th subgraph in the subgraph set C; θ is the weight variance threshold of the connecting edge, σ(G i ) represents the weight difference of the connecting edges in the i-th subgraph, which is calculated as follows: |G i | represents the number of vertices in the i-th subgraph, and φ represents the empty set; The clustering in the spatial shared graph to form a sub-graph set specifically includes: Get the maximum spanning tree of the space sharing graph; In the maximum spanning tree, clustering is performed to form a subgraph set with the goal of minimizing the weight difference of the connecting edges, and a plurality of vertices and connecting edges between the vertices in the subgraph set that are continuously connected are regarded as a subgraph; The location information of the maintenance personnel is obtained, and the order information whose service address is located in the submap where the maintenance personnel is located is sent to the maintenance personnel terminal.

2. The maintenance dispatching method based on hierarchical clustering according to claim 1, characterized in that: The step of clustering the subgraphs in the maximum spanning tree with the goal of minimizing the weight difference of the connecting edges as much as possible includes: Step A, set a subgraph variable C, initialize the subgraph variable C to the maximum spanning tree T, obtain the connection edge set E of the subgraph variable C, and set the variance threshold θ; Step B, calculate the weight variance σ(E) of the connected edges in the connected edge set E. If σ(E)≤θ, output the subgraph set variable C. If σ(E)>θ, delete the connected edge in the subgraph set variable C that can reduce the weight variance the most, and enter step C. Step C, return to step B.

3. A maintenance dispatching method based on hierarchical clustering as claimed in claim 1 or 2, characterized in that: Obtain the location information of the maintenance personnel, and send order information to the maintenance personnel terminal based on the location information of the maintenance personnel, the shared order pair density of the subgraph where the maintenance personnel is located, and the shared order pair density of the adjacent subgraph of the subgraph where the maintenance personnel is located. The shared order pair density refers to the ratio of the total number of shared order pairs in the subgraph to the number of vertices in the subgraph.

4. A maintenance dispatching method based on hierarchical clustering as claimed in claim 3, characterized in that: The sending of order information to the maintenance personnel terminal in combination with the location information of the maintenance personnel, the density of shared order pairs of the subgraph where the maintenance personnel is located, and the density of shared order pairs of the adjacent subgraphs of the subgraph where the maintenance personnel is located specifically includes: If the number of maintenance personnel in the subgraph where the maintenance personnel is located matches the density of shared order pairs of the subgraph, or the number of maintenance personnel in the subgraph where the maintenance personnel is located is less than the number of personnel matched by the density of shared order pairs of the subgraph, the order information whose service address is in the subgraph where the maintenance personnel is located is sent to the maintenance personnel terminal; If the number of maintenance personnel in the sub-graph where the maintenance personnel are located is greater than the number of personnel whose density matches the shared order pairs of the sub-graph, the order information whose service address is located in the first sub-graph is sent to the maintenance personnel terminal, where the first sub-graph is an adjacent sub-graph of the sub-graph where the maintenance personnel are located and the number of maintenance personnel in the first sub-graph is less than the number of personnel whose density matches the shared order pairs of the first sub-graph.

5. A maintenance dispatching method based on hierarchical clustering as claimed in claim 1, 2 or 4, characterized in that: Sending order information whose service address is within a preset distance threshold from the maintenance personnel's current location to the maintenance personnel's terminal device; Alternatively, the order information in the sub-area where the maintenance personnel is currently located is sent to the terminal device of the maintenance personnel.

6. A maintenance dispatch method based on hierarchical clustering as claimed in claim 1, 2 or 4, characterized in that: The following steps are also included: Get the shared order pair density of each subgraph. The larger the shared order pair density, the larger the required maintenance personnel density. The smaller the shared order pair density, the smaller the required maintenance personnel density. The shared order pair density refers to the ratio of the number of shared order pairs in the subgraph to the number of vertices in the subgraph; The maintenance personnel density refers to the ratio of the number of maintenance personnel dispatched to the subgraph to the number of vertices of the subgraph.

7. A maintenance dispatching device based on hierarchical clustering, characterized in that: include: A partitioning module divides the maintenance area into multiple sub-areas; An order information acquisition module is used to acquire the order information of the currently undispatched orders, wherein the order information includes the service address and the repair category; A shared order pair quantity acquisition module, which acquires the number of shared order pairs between any two sub-areas based on the order information, wherein the shared order pairs are composed of orders belonging to two different sub-areas but having the same maintenance category; A space sharing graph construction module is used to construct a space sharing graph, including: taking the sub-area as a vertex, connecting two vertices with a shared order pair, and using the number of shared order pairs of the two vertices as the weight of the connecting edge of the two vertices; The subgraph acquisition module clusters the spatial shared graph to form subgraphs. The goal of the clustering process is to make the cumulative sum of the weights of the connecting edges in the subgraph as large as possible and the weight difference of the connecting edges in the subgraph as small as possible. The objective function of the clustering process is: Among them, G represents the spatial shared graph, S(G) represents the set of all subgraphs in the spatial shared graph, w(e) represents the weight of the connecting edge e in the shared spatial graph, C is the subgraph set obtained by clustering, G i represents the i-th subgraph in the subgraph set C; θ is the weight variance threshold of the connecting edge, σ(G i ) represents the weight difference of the connecting edges in the i-th subgraph, which is calculated as follows: |G i | represents the number of vertices in the i-th subgraph, and φ represents the empty set; The clustering in the spatial shared graph to form a sub-graph set specifically includes: Get the maximum spanning tree of the space sharing graph; In the maximum spanning tree, clustering is performed to form a subgraph set with the goal of minimizing the weight difference of the connecting edges, and a plurality of vertices and connecting edges between the vertices in the subgraph set that are continuously connected are regarded as a subgraph; The dispatching module obtains the location information of the maintenance personnel and sends the order information whose service address is located in the submap where the maintenance personnel is located to the maintenance personnel terminal.

8. A dispatching device, characterized in that: It comprises an order receiving module and a processing platform, wherein the order receiving module receives order information issued by a user; the processing platform receives order information from the order receiving module, and sends the order information to a maintenance personnel terminal according to the maintenance dispatching method based on hierarchical clustering as described in one of claims 1-6.

9. A storage medium storing a program, wherein the program enables a computer to execute the steps of the maintenance dispatching method based on hierarchical clustering as claimed in any one of claims 1 to 6.

Images