Three-dimensional warehouse inventory taking and dynamic parking space distribution method

Abstract

The invention discloses a three-dimensional warehouse stocktaking and dynamic parking space distribution method, which comprises the following steps of: judging whether workpieces exist in parking spaces of an RGV type three-dimensional warehouse or not, and if the workpieces exist, carrying out information acquisition on the workpieces and acquiring identification data of the workpieces; uploading the identification data to a server background; wherein a vehicle body information base is stored in the server background, and the vehicle body information base comprises complete information, corresponding skid body numbers and warehousing time of the workpieces, which are in one-to-one correspondence with the identification data; comparing and matching the identification data with a vehicle body information base to obtain complete information of the workpiece; according to the invention, the cost is reduced because the requirements on read heads and wiring are reduced. And the flexibility and expansibility of the system also provide convenience for future technology upgrade and storage location increase. And the reliability of inventory management is further improved by an automatic anomaly detection and quick response mechanism. On the whole, the invention provides an efficient, safe, economical and easy-to-maintain inventory management method.

Description

Technical Field

[0001] The present invention relates to the technical field of stereoscopic warehouses, and in particular to a stereoscopic warehouse inventory and dynamic parking space allocation method. Background Art

[0002] A high-bay warehouse is also called a high-bay warehouse or an elevated warehouse. It generally refers to a warehouse that uses shelves with several, dozens or even dozens of layers to store unit goods, and uses corresponding material handling equipment to carry out warehousing and outbound operations.

[0003] In related technologies, visual recognition technology can be used to accurately identify information such as numbers, letters, and text on workpieces, and then index and match the specific inventory data of the workpieces in the storage location.

[0004] When the vertical warehouse equipment works abnormally, the workpieces in the warehouse are manually transferred or the internal data of the scheduling system is disordered, the existing manual inventory or RFID system identification methods have the following problems: manual inventory requires stopping the line and cannot be carried out quickly in real time, which affects the production rhythm, and requires people to enter the equipment operation area for inventory, which poses certain safety hazards. The RFID system requires the installation of readers and wiring at each warehouse, which is costly. Furthermore, the vertical warehouse equipment, as the body storage area, is an important part of the production system. Since the traditional storage area uses a fixed partition mode, the warehouse location allocation cannot be adjusted after it is built. When production demand changes, there is waste or insufficient location allocation in the storage area, affecting production efficiency and stability. Summary of the invention

[0005] In order to solve the above problem, the present invention provides a three-dimensional parking garage and a dynamic parking space allocation method to solve the problem.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] A method for storing a three-dimensional warehouse, characterized in that it is applied to a RGV type three-dimensional warehouse, and comprises the following steps:

[0008] S101: Determine whether there is a workpiece in the storage location of the RGV-type three-dimensional warehouse. If there is a workpiece, execute step S102;

[0009] S102: Collect information on the workpiece to obtain identification data of the workpiece;

[0010] S103: uploading the identification data to a server backend; wherein the server backend stores a vehicle body information database, and the vehicle body information database includes complete information of the workpiece corresponding to the identification data, the corresponding skid body number, and the storage time;

[0011] S104: Compare and match the identification data with the vehicle body information database to obtain complete information of the workpiece.

[0012] It is further configured as follows: the complete information of the workpiece includes the order number of the vehicle body, the vehicle model, the manufacturing route, the color of the vehicle body mid-coat, the color of the vehicle body topcoat, and whether the vehicle body is a color-matching vehicle.

[0013] A three-dimensional warehouse and disk system comprises: a communication module, a workpiece detection module, a data acquisition module, a data comparison and processing module, an alarm module, a warehouse location determination logic module and a motion strategy control module.

[0014] The communication module is used for communication between each module and the server backend;

[0015] The workpiece detection module is used to identify whether there is a workpiece in the storage location;

[0016] The data acquisition module is used to acquire identification data of the workpiece;

[0017] The data comparison processing module is used to compare and match the identification data with the vehicle body information database. If the comparison is successful, the complete information of the workpiece is successfully obtained and the matching result is output. If the comparison fails, the signal is transmitted to the alarm module, and the alarm module issues an alarm.

[0018] The storage location determination logic module controls the mobile vehicle to the next storage location for determination after the workpiece detection module identifies that there is no workpiece in the storage location or the data processing module outputs a matching result;

[0019] The motion strategy control module is used to control the running state of the mobile car.

[0020] It is further configured as follows: it also includes a plurality of mobile carts, which are respectively located at the entrance of each layer of the RGV-type vertical warehouse; the workpiece detection module is an ultrasonic sensor or a diffuse reflection sensor installed on the mobile cart, which determines whether there is a workpiece in the storage location through ultrasonic or diffuse reflection recognition; the data acquisition module is a camera installed on the mobile cart, which is used for visual disk storage.

[0021] A dynamic parking space allocation method is applied to an RGV-type three-dimensional parking garage with a visual parking function, comprising the following steps:

[0022] Establish RGV-type three-dimensional warehouse information system;

[0023] Construct a decision cost matrix based on the stereo library information system;

[0024] Establish cost expectations based on the decision cost matrix based on Bayesian decision theory;

[0025] Construct the minimum risk decision cost model based on cost expectation;

[0026] Execute the minimum risk decision cost model to obtain the minimum risk decision result;

[0027] Perform dynamic parking space allocation based on the minimum risk decision result.

[0028] The stereo library information system is represented as:

[0029] S=(U,At,{V a |a∈At},{I a |a∈At});

[0030] Among them, Y is the domain of the system; At is the attribute set of the system; V a is the conditional attribute set of the system; a is the decision attribute set of the system.

[0031] The domain U includes a state set Ω and an action set Δ, where the state set Ω = {C P ,C N}, action set Δ={A P ,A B ,A N}, the decision cost matrix of the state set and action set is constructed as follows:

[0032]

[0033] Among them, C P Indicates that the target storage location during the warehousing or outbound process can meet the demand; C N Indicates that the target storage location during the warehousing or outbound process cannot meet the demand; A P Indicates normal execution of storage and outgoing actions; A B Indicates a fuzzy boundary; A N Indicates that the storage entry and exit actions are refused.

[0034] The cost expectation is expressed as:

[0035]

[0036] Among them, R(A|x) represents the expected cost of object x taking different actions A; Pr(C P |x) represents the probability that the target storage location can meet the demand during the process of object x entering or leaving the warehouse; Pr(C N |x) represents the probability that the target storage location cannot meet the demand during the storage or outbound process of object x; the object x represents the storage location of the three-dimensional warehouse and the inventory at the storage location.

[0037] It is further set as follows: The calculation steps of the probability that the object x meets different states C include:

[0038] Step 1: Construct the fuzzy decision matrix D, D = (r ij ) m×n ,r ij =wΘy ij ;

[0039] Where D is the fuzzy decision matrix; r is the fuzzy decision matrix element, i is the serial number of object x, j is the serial number of attribute item, m is the number of matrix rows, n is the number of matrix columns; w is the weight coefficient of attribute item; y is the fuzzy index;

[0040] Step 2: according to the distance between different objects and the fuzzy positive ideal and the fuzzy negative ideal;

[0041] Among them, the distance between different objects and the fuzzy positive ideal

[0042] Distances of different objects from the fuzzy negative ideal

[0043] In the formula, is the distance between different objects and the fuzzy ideal, is the distance between different objects and the fuzzy negative ideal, are fuzzy positive ideals of different indicators, It is a fuzzy negative ideal for different indicators;

[0044] Step 3: According to the distance between different objects and the blur ideal and the distances of different objects from the fuzzy negative ideal Calculate the membership degree μ; the membership degree

[0045] Among them, the probability Pr(C P |x)=μ;

[0046] The probability Pr(C N |x)=1-μ.

[0047] It is further set as follows: The minimum risk decision cost model is expressed as:

[0048]

[0049] Compared with the prior art, the beneficial technical effects of the present invention are:

[0050] 1. The present invention realizes an automated inventory counting process by combining a mobile car, a high-speed camera and a vehicle body information database. It can automatically conduct online inventory counting in real time, improve inventory counting efficiency, and ensure that the production rhythm is not affected. There is no need for people to enter the equipment operation area to count inventory, which protects personal safety to the greatest extent. There is no need for too many people to participate in the inventory counting work, which reduces the inventory counting cost.

[0051] Therefore, the present invention not only improves inventory efficiency and data accuracy, but also reduces errors and safety risks caused by manual operations. Since inventory work can be performed during non-production hours, it avoids interference with the production line and ensures the continuity of production. In addition, compared with the RFID system, the present invention reduces costs because the need for readers and wiring is reduced. The flexibility and scalability of the system also facilitate future technology upgrades and increased storage locations. Automated anomaly detection and rapid response mechanisms further improve the reliability of inventory management. Overall, the present invention provides an efficient, safe, economical and easy-to-maintain inventory management method.

[0052] 2. The present invention realizes efficient and flexible management of the stereoscopic warehouse through dynamic parking space allocation and minimum decision cost algorithm, and integrates the traditional fixed workpiece storage areas into the entire interconnected storage area of ​​the whole workshop, so that the storage location types of each workpiece stored in the warehouse can be dynamically and flexibly adjusted in the order of storage and outbound storage; each storage and outbound storage process is integrated into a modular configuration, and different storage and outbound storage methods can be enabled according to specific needs;

[0053] The present invention integrates multiple information sources, optimizes the decision-making process by using fuzzy decision-making and Bayesian theory, reduces the risk of wrong decisions, and improves the accuracy of warehouse allocation and production efficiency. At the same time, through modular configuration and comprehensive consideration of multiple allocation logics, the production process is further optimized, costs are saved, and production risks are effectively managed. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 This is a schematic diagram of the steps in Example 1 of the present invention;

[0056] Figure 2 This is a schematic diagram of the system framework in Example 2 of the present invention;

[0057] Figure 3Schematic diagram of the steps in Example 3 of the present invention. DETAILED DESCRIPTION

[0058] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0060] Example 1

[0061] Reference Figure 1 , is a three-dimensional library storage method disclosed in the present invention, comprising:

[0062] S101: Determine whether there is a workpiece in the storage location of the RGV-type three-dimensional warehouse. If there is a workpiece, execute step S102;

[0063] S102: Collect information on the workpiece to obtain identification data of the workpiece;

[0064] S103: uploading the identification data to a server backend; wherein the server backend stores a vehicle body information database, and the vehicle body information database includes complete information of the workpiece corresponding to the identification data, the corresponding skid body number, and the storage time;

[0065] The complete information of the workpiece includes the order number, vehicle model, manufacturing route, mid-coat color, topcoat color, and whether the vehicle body is a color-matching vehicle.

[0066] S104: Compare and match the identification data with the vehicle body information database to obtain complete information of the workpiece.

[0067] Example 2

[0068] Reference Figure 2 , is a three-dimensional library disk library system disclosed in the present invention, comprising:

[0069] The communication module is used for communication between each module and the server backend;

[0070] The workpiece detection module is used to identify whether there is a workpiece in the storage location;

[0071] The data acquisition module is used to acquire identification data of the workpiece;

[0072] The data comparison processing module is used to compare and match the identification data with the vehicle body information database. If the comparison is successful, the complete information of the workpiece is successfully obtained and the matching result is output. If the comparison fails, the signal is transmitted to the alarm module, and the alarm module issues an alarm.

[0073] The storage location determination logic module controls the mobile vehicle to the next storage location for determination after the workpiece detection module identifies that there is no workpiece in the storage location or the data processing module outputs a matching result;

[0074] The motion strategy control module is used to control the running state of the mobile car.

[0075] A mobile cart, wherein multiple mobile carts are respectively located at the entrance of each layer of the RGV-type vertical warehouse. The workpiece detection module is an ultrasonic sensor or a diffuse reflection sensor installed on the mobile cart, which determines whether there is a workpiece in the warehouse through ultrasonic or diffuse reflection recognition; the data acquisition module is a camera installed on the mobile cart, which is used for visual inventory.

[0076] Example 3

[0077] The vertical warehouse information system is the perception and record of the actual status of the vertical warehouse. The vertical warehouse information system in the prior art includes database information and PLC information. The vertical warehouse information system in this technical solution adds visual collection information.

[0078] Since each information source has uncertainty, there may be omissions or errors. For example, the confidence of database information is α1, the confidence of PLC information is α2, and the confidence of visual information is α3. The existing process is to reject the decision result when the information is inconsistent. It is a relatively extensive decision-making method. The decision cost is not optimal. In fact, the introduction of visual information increases the decision cost.

[0079] Based on this, refer to Figure 3 The present invention discloses a dynamic parking space allocation method, comprising:

[0080] S201, establish an RGV type three-dimensional warehouse information system, the three-dimensional warehouse information system is expressed as:

[0081] S=(U,At,{V a |a∈At},{I a |a∈At});

[0082] Wherein, U is the domain of the system; At is the attribute set of the system, and the attribute set represents the collection of attribute items of the stereo library information system. For specific attribute items of the system, see Table 1; V ais the conditional attribute set of the system, which is for individual objects and represents the specific information of each storage location (including the above inventory) in each attribute item; a is the decision attribute set of the system, wherein the decision attribute set represents the state affiliation of each object according to its condition attribute set;

[0083] Furthermore, the domain U includes a state set Ω and an action set Δ, where the state set Ω = {C P ,C N}, action set Δ={A P ,A B ,A N}, C P Indicates that the target storage location during the warehousing or outbound process can meet the demand; C N Indicates that the target storage location during the storage or outbound process cannot meet the demand. In the present invention, C P With C N That is the state affiliation; A P Indicates normal execution of storage and outgoing actions; A B Indicates a fuzzy boundary; A N Indicates that the storage entry and exit actions are refused.

[0084] Table 1, representative specific attribute items (decision-making indicators):

[0085]

[0086]

[0087] In Table 1, the present invention represents various situations as quantitative state values. For example, there are multiple situations for the signals of 2 induction switches to judge whether there is stock in the RGV type three-dimensional warehouse:

[0088] Case 1: Both are true;

[0089] Case 2: 1 is true and 1 is false;

[0090] Case 3: 1 is true and 1 is no signal.

[0091] The present invention represents case 1, case 2 and case 3 as quantitative state values.

[0092] S202, construct a decision cost matrix based on the stereo library information system, and the decision cost matrix is ​​expressed as:

[0093]

[0094] S203, based on the Bayesian decision theory, the cost expectation is established according to the decision cost matrix, and the cost expectation is expressed as:

[0095]

[0096] Where R(A|x) represents the expected cost of object x taking different actions A. In the present invention, different actions A include but are not limited to normal execution of storage in and out actions A. P , at the fuzzy boundary A B , refuse to execute the storage action A N ; Pr(C P |x) represents the probability that the target storage location can meet the demand during the process of object x entering or leaving the warehouse; Pr(C N |x) represents the probability that the target storage location cannot meet the demand during the storage or outbound process of object x; the object x represents the storage location of the three-dimensional warehouse and the inventory at the storage location.

[0097] Furthermore, the probability Pr(C P |x) and the probability Pr(C N The calculation steps of |x) are as follows:

[0098] Step 1: Construct the fuzzy decision matrix D. Where D is the fuzzy decision matrix; r is the fuzzy decision matrix element, i is the serial number of object x, j is the serial number of attribute item, m is the number of matrix rows, n is the number of matrix columns; w is the weight coefficient of attribute item; y is the fuzzy index;

[0099] Step 2: according to the distance between different objects and the fuzzy positive ideal and the fuzzy negative ideal;

[0100] Among them, the distance between different objects and the fuzzy positive ideal

[0101] Distances of different objects from the fuzzy negative ideal

[0102] In the formula, is the distance between different objects and the fuzzy ideal, is the distance between different objects and the fuzzy negative ideal, are fuzzy positive ideals of different indicators, It is a fuzzy negative ideal for different indicators;

[0103] Step 3: According to the distance between different objects and the blur ideal and the distances of different objects from the fuzzy negative ideal Calculate the membership degree μ; the membership degree

[0104] Among them, the probability Pr(C P|x) = μ;

[0105] The probability Pr(C N |x)=1-μ.

[0106] S204, constructing a minimum risk decision cost model based on the cost expectation. The minimum risk decision cost model is expressed as:

[0107]

[0108] S205, executing the minimum risk decision cost model to obtain the minimum risk decision result;

[0109] S206, performing dynamic parking space allocation according to the minimum risk decision result;

[0110] For example, for a certain outbound decision process, there are n candidate goods to choose from, and the goods are numbered i=1, 2, ..., n.

[0111] For each cargo, assign a value to each decision indicator based on the information obtained, denoted as y ij , taking object 1 as an example, see Table 2.

[0112] Table 2, Decision indicator assignment for object 1:

[0113]

[0114] The weighting coefficients of each indicator are as follows:

[0115]

[0116] Calculate the weighted decision index for object 1:

[0117]

[0118] Similarly, the weighted decision indicators of objects 2, 3, ... are calculated to form a fuzzy decision matrix:

[0119]

[0120] Determine the fuzzy positive ideal and the fuzzy negative ideal. For the positive index, the fuzzy positive ideal and the fuzzy negative ideal of the jth index are:

[0121]

[0122] Calculate the membership degree (taking object 1 as an example):

[0123]

[0124] μ1=0.86;

[0125] Get the probability of each object belonging to different states (taking object 1 as an example):

[0126] Pr(C P |x1)=0.86,Pr(C N |x1)=0.14;

[0127] Obtain decision cost assignments, for example:

[0128]

[0129] Calculate the expected cost (taking object 1 as an example):

[0130]

[0131] Therefore, according to the calculation of the existing information sources, the minimum expected cost of object 1 is R(A B |x1), the decision with the least risk is to take action A B , additional actions are required to confirm the status of the goods.

[0132] In the dynamic parking allocation mode, the storage type can be flexibly changed. For example, the topcoat parking space can be changed to the midcoat parking space. Each time the target storage space is selected for entry, an entry decision needs to be made. Each time the target goods are selected for exit, an exit decision needs to be made. If a storage location changes its type when there is no goods, an entry decision will be made according to the new type; if a storage location changes its type when there is goods, its goods will continue to be retained and the exit decision will be made according to the original type until the empty storage location after the goods are exited makes an entry decision according to the new type.

[0133] The present invention proposes a minimum decision cost model that includes visual acquisition information and is suitable for the dynamic parking space allocation mode of the stereoscopic warehouse. It effectively supplements the information source in the entry and exit decision-making of the visual recognition method. The dynamic parking space allocation model significantly improves the flexibility of the stereoscopic warehouse, obtains the minimum risk decision result, and reduces the possibility of decision-making errors.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional warehouse inventory method, characterized in that: Applied to RGV type stereoscopic warehouse, including the following steps: S101: Determine whether there is a workpiece in the storage location of the RGV-type three-dimensional warehouse. If there is a workpiece, execute step S102; S102: Collect information on the workpiece to obtain identification data of the workpiece; S103: uploading the identification data to a server backend; wherein the server backend stores a vehicle body information database, and the vehicle body information database includes complete information of the workpiece corresponding to the identification data, the corresponding skid body number, and the storage time; S104: Compare and match the identification data with the vehicle body information database to obtain complete information of the workpiece.

2. A three-dimensional warehouse inventory method according to claim 1, characterized in that: The complete information of the workpiece includes the order number of the vehicle body, the vehicle model, the manufacturing route, the color of the vehicle body mid-coat, the color of the vehicle body topcoat, and whether the vehicle body is a color-matching vehicle.

3. A three-dimensional warehouse inventory system, applied to the three-dimensional warehouse inventory method according to any one of claims 1 to 2, characterized in that: include: Communication module, workpiece detection module, data acquisition module, data comparison and processing module, alarm module, storage location determination logic module and motion strategy control module; The communication module is used for communication between each module and the server backend; The workpiece detection module is used to identify whether there is a workpiece in the storage location; The data acquisition module is used to acquire identification data of the workpiece; The data comparison processing module is used to compare and match the identification data with the vehicle body information database. If the comparison is successful, the complete information of the workpiece is successfully obtained and the matching result is output. If the comparison fails, the signal is transmitted to the alarm module, and the alarm module issues an alarm. The storage location determination logic module controls the mobile vehicle to the next storage location for determination after the workpiece detection module identifies that there is no workpiece in the storage location or the data processing module outputs a matching result; The motion strategy control module is used to control the running state of the mobile car.

4. The three-dimensional library disk storage system according to claim 3 is characterized in that: It also includes multiple mobile carts, which are respectively located at the entrance of each layer of the RGV-type vertical warehouse. The workpiece detection module is an ultrasonic sensor or a diffuse reflection sensor installed on the mobile cart, which determines whether there is a workpiece in the storage location through ultrasonic or diffuse reflection recognition; the data acquisition module is a camera installed on the mobile cart, which is used for visual inventory.

5. A dynamic parking space allocation method, applied to the RGV-type three-dimensional parking garage with visual parking function as claimed in any one of claims 1 to 4, characterized in that: The following steps are involved: S201: Establish RGV type three-dimensional warehouse information system; S202: constructing a decision cost matrix based on the stereo library information system; S203: Establish cost expectation based on decision cost matrix based on Bayesian decision theory; S204: Construct a minimum risk decision cost model based on cost expectations; S205: Execute the minimum risk decision cost model to obtain the minimum risk decision result; S206: Execute dynamic parking space allocation according to the minimum risk decision result.

6. A dynamic parking space allocation method according to claim 5, characterized in that: The stereo library information system is represented as: S=(U,At,{V a |a∈At},{I a |a∈At}); Among them, U is the domain of the system; At is the attribute set of the system; V a is the conditional attribute set of the system; a is the decision attribute set of the system.

7. A dynamic parking space allocation method according to claim 6, characterized in that: The domain U includes a state set Ω and an action set Δ, where the state set Ω = {C P ,C N }, action set Δ={A P ,A B ,A N }, the decision cost matrix of the state set and action set is constructed as follows: Among them, C P Indicates that the target storage location during the warehousing or outbound process can meet the demand; C N Indicates that the target storage location during the warehousing or outbound process cannot meet the demand; A P Indicates normal execution of storage and outgoing actions; A B Indicates a fuzzy boundary; A N Indicates that the storage entry and exit actions are refused.

8. A dynamic parking space allocation method according to claim 7, characterized in that: The cost expectation is expressed as: Among them, R(A|x) represents the expected cost of object x taking different actions A; Pr(C P |x) represents the probability that the target storage location can meet the demand during the process of object x entering or leaving the warehouse; Pr(C N |x) represents the probability that the target storage location cannot meet the demand during the storage or outbound process of object x; the object x represents the storage location of the three-dimensional warehouse and the inventory at the storage location.

9. A dynamic parking space allocation method according to claim 8, characterized in that: The steps to calculate the probability of object x meeting different states C include: Step 1: Construct the fuzzy decision matrix D, D = (r ij ) m×n ,r ij =wΘy ij ; Where D is the fuzzy decision matrix; r is the fuzzy decision matrix element, i is the serial number of object x, j is the serial number of attribute item, m is the number of matrix rows, n is the number of matrix columns; w is the weight coefficient of attribute item; y is the fuzzy index; Step 2: according to the distance between different objects and the fuzzy positive ideal and the fuzzy negative ideal; Among them, the distance between different objects and the fuzzy positive ideal Distances of different objects from the fuzzy negative ideal In the formula, is the distance between different objects and the fuzzy positive ideal, is the distance between different objects and the fuzzy negative ideal, are fuzzy positive ideals of different indicators, It is a fuzzy negative ideal for different indicators; Step 3: According to the distance between different objects and the ideal blur and the distances of different objects from the fuzzy negative ideal Calculate the membership degree μ; the membership degree Among them, the probability Pr(C P |x)=μ; The probability Pr(C N |x)=1-μ.

10. A dynamic parking space allocation method according to claim 9, characterized in that: The minimum risk decision cost model is expressed as:

Images