An order data analysis system and method based on ERP management system

Abstract

The present invention discloses an order data analysis system and method based on an ERP management system, and relates to the technical field of order data analysis. The system includes an order information collection module, an order information processing module, a similarity model construction and analysis module, a user order data analysis module and an ERP management module; the order information collection module is connected to the order information processing module; the order information processing module is connected to the similarity model construction and analysis module; the similarity model construction and analysis module is connected to the user order data analysis module; the user order data analysis module is connected to the ERP management module. The present invention can identify and screen orders with order feature marks, monitor the transportation damage of marked orders, return damaged marked orders in time at the transfer station, and solve the problem of damaged orders in time.

Description

Technical Field

[0001] The present invention relates to the technical field of order data analysis, and in particular to an order data analysis system and method based on an ERP management system. Background Art

[0002] ERP management system is the operation mode of modern enterprise management. It is a highly integrated system applied throughout the company, covering management work such as customers, projects, inventory, procurement, supply and production. It optimizes enterprise resources to maximize resource benefits, standardizes basic enterprise management work, promotes flat management of enterprises, improves the cost management awareness of enterprise employees, and enhances the market competitiveness of enterprises.

[0003] With the further popularization and application of the Internet of Things in China, online shopping has gradually become one of people's online behaviors. The new shopping experience and convenient consumption methods have replaced or replaced the traditional shopping methods on a larger scale, saving people's shopping time and shopping costs to a large extent. However, at the same time, during the express delivery process, there will be problems such as damage to goods caused by express stacking and squeezing or encountering bad weather. Only when consumers sign for the goods can they find the problem of damaged goods, which brings great inconvenience and trouble to consumers; in addition, some companies have the situation of express delivery returned midway, but due to the lack of analysis of the actual situation of users, it is impossible to further realize intelligent management. Summary of the invention

[0004] The purpose of the present invention is to provide an order data analysis system and method based on an ERP management system to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An order data analysis method based on an ERP management system, the method comprising the following steps:

[0007] Step S1: collecting order packaging information at the order sending site, detecting order characteristic marks based on the order packaging information, and generating a marked order when the order characteristic marks are detected;

[0008] Step S2: collecting an initial image of the marked order, processing the initial image of the marked order, and obtaining an initial vector of an initial grayscale image of the marked order;

[0009] Step S3: Obtain all the transfer stations of the marked order. When the marked order arrives at each transfer station, obtain the transfer images of the marked order at each transfer station respectively and process the transfer images of the marked order at each transfer station to obtain the transfer quantity of the transfer grayscale image of the marked order at each transfer station;

[0010] Step S4: construct a marked order similarity calculation model, compare the initial vector of the initial grayscale image of the marked order with the transfer vector of the transfer grayscale image of the marked order at each transfer station, and calculate the similarity of the marked order;

[0011] Step S5: setting a similarity threshold of the marked order. When the similarity of the marked order does not exceed the threshold, the marked order is determined to have transportation damage, the historical order data of the user to whom the marked order belongs is obtained, and the probability that the user to whom the marked order belongs applies for a return due to transportation damage is calculated;

[0012] Step S6: setting a probability threshold for a marked order to be returned due to transportation damage. When the probability of the user to whom the marked order belongs to apply for a return due to transportation damage exceeds the probability threshold for the marked order to be returned due to transportation damage, the marked order is determined to be returned due to transportation damage, and the transit image of the marked order at the current transit station is fed back to the merchant and the user to whom it belongs, and the marked order is returned;

[0013] Step S7: Use the ERP management system to re-mark the current transit site and the previous transit site of the marked order, and count the number of marked orders returned in transit. When the number of marked orders returned in transit at a certain transit site exceeds the threshold, the ERP management system will alarm the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information.

[0014] Further, in steps S1-S2, the order packaging information includes order transportation information and order characteristic flags;

[0015] The order transportation information includes the order number, waybill number, shipping address and delivery address;

[0016] The order characteristic marks include a fragile mark, a handle with care mark and a moisture-proof mark;

[0017] When the order feature flag is detected as a fragile flag, a marked order is generated When the order feature information is detected as a handle with care sign, a marked order is generated When the order feature information is detected as a moisture-proof mark, a marked order is generated

[0018] Collect and mark orders Initial image Marking Orders Initial image Perform grayscale processing and generate marked orders The initial grayscale image Marking orders The initial grayscale image Perform edge computing;

[0019] According to the formula, the marked orders The initial grayscale image Each pixel has:

[0020]

[0021] Where G(x,y) represents the Gaussian filter function; E x0,n (x,y) indicates the marked order The initial grayscale image The horizontal result after each pixel of E is Gaussian filtered; y0,n (x,y) indicates the marked order The initial grayscale image The vertical result of each pixel after Gaussian filtering;

[0022] Marking orders The initial grayscale image The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated:

[0023]

[0024] Among them, E x0,avg (x,y) indicates the marked order The initial grayscale image The average value of the horizontal result after Gaussian filtering of each pixel; E y0,avg (x,y) indicates the marked order The initial grayscale image The vertical average of each pixel after Gaussian filtering;

[0025] Marking Orders The initial grayscale image The initial vector is:

[0026]

[0027] in, Indicates marking order The initial grayscale image The initial vector of .

[0028] In the above technical scheme, firstly, orders containing characteristic order marks such as fragile mark, handle with care mark and moisture-proof mark are screened and marked as orders that are prone to damage, so as to determine the detection objects of the system; secondly, the initial image of the marked order is collected, and the initial image of the marked order is gray-processed and edge calculated in order to extract the contour of the image and ensure the accuracy of the image; thirdly, edge calculation through the Canny operator can obtain clearer image contour edges, which provides a basis for calculating the similarity between the initial image and the transfer image; finally, the pixel coordinates of the grayscale image can better reflect the changes of each pixel in the image.

[0029] Further, in step S3,

[0030] Get Mark Order All transfer stations B={b1,b2,...,b z}; Corresponding mark order Transit images at various transit points Grayscale processing is performed on the transfer images of the marked orders at each transfer station to generate marked orders Grayscale images of the transfer at each transfer station Marking orders The transferred grayscale image F is edge computed and the marked order is The horizontal and vertical results of each pixel of the transferred grayscale image F after Gaussian filtering are averaged to obtain the marked order The transit quantity of the transit grayscale image at each transit station

[0031] Furthermore, in step S4, constructing a marked order similarity calculation model includes:

[0032] Get Mark Order The initial grayscale image The initial vector Get Mark Order The transit quantity of the transit grayscale image at each transit station

[0033] Calculate Mark Order Similarity:

[0034]

[0035] in, Indicates marking order Similarity at the jth transfer station; Indicates marking order The initial grayscale image The initial vector of Indicates marking order Grayscale image of the transfer at the jth transfer station The center steering volume.

[0036] In the above technical scheme, the cosine similarity algorithm is used to compare the similarity of two vectors. The closer the cosine value is to 1, the closer the angle is to 0 degrees, that is, the more similar the two vectors are. Using the cosine similarity algorithm to compare the initial vector of the initial grayscale image of the marked order and the intermediate grayscale image of the marked order can clearly show the significant changes in the grayscale image of the marked order.

[0037] Further, in steps S5-S6, set the marked order A i The similarity threshold of , denoted as cos(θ0);

[0038] When marking an order When the similarity exceeds the threshold, the order is marked. No shipping damage occurred;

[0039] When marking an order When the similarity does not exceed the threshold, the order is marked. In case of shipping damage, get marked order User C k Historical order data;

[0040] The acquisition of marked order User C g The historical order data includes:

[0041] Get Mark Order User C h The number of orders that incur shipping damage, denoted as U h ;

[0042] Get Mark Order User C h The number of orders that were returned due to shipping damage, denoted as V h ;

[0043] Calculate Mark Order User C h Probability of applying for return due to shipping damage:

[0044]

[0045] Among them, p h Indicates marking order User C h The probability of applying for return due to shipping damage;

[0046] Set the probability threshold for applying for return due to shipping damage for marked orders, denoted as p0; User C h When the probability of applying for a return due to shipping damage exceeds the probability threshold for marking an order for a return due to shipping damage, the order will be marked The transit image at the current transit station is fed back to the merchant and the user, and the order is marked return.

[0047] In the above technical solution, if an order is damaged during transportation, the consumer will most likely choose to apply for a return. When the historical order data of the user to whom the marked order belongs is queried, and the probability of the user applying for a return due to transportation damage is calculated, if the user has a high probability of applying for a return due to transportation damage, the system will be able to make a judgment in a timely manner and return the damaged order, so as to achieve the effect of efficiently processing damaged orders and provide users with a better shopping experience.

[0048] An order data analysis system based on an ERP management system, the system comprising an order information collection module, an order information processing module, a similarity model building and analysis module, a user order data analysis module and an ERP management module;

[0049] The order information collection module is used to collect order packaging information, detect order feature marks, generate marked orders, collect the initial image of the marked order, mark all the transit sites of the order and the transit images of the marked order at each transit site; the order information processing module processes the initial image of the marked order and the transit images of the marked order at each transit site to obtain the initial vector of the initial grayscale image of the marked order and the diversion amount of the transit grayscale image of the marked order at each transit site; the similarity model construction and analysis module is used to construct a marked order similarity calculation model, compare the initial vector of the initial grayscale image of the marked order with the diversion amount of the transit grayscale image of the marked order at each transit site, and calculate the similarity of the marked order; the user order data analysis module is used to set the similarity threshold of the marked order, and when the similarity of the marked order does not exceed the threshold, it is judged that the marked order is marked. If a marked order generates transportation damage, the historical order data of the user to whom the marked order belongs is obtained, and the probability that the user to whom the marked order belongs applies for return due to transportation damage is calculated; the ERP management module is used to set a probability threshold for the marked order to apply for return due to transportation damage. When the probability that the user to whom the marked order belongs applies for return due to transportation damage exceeds the probability threshold for the marked order to apply for return due to transportation damage, it is determined that the marked order is applied for return due to transportation damage, and the transit image of the marked order at the current transit station is fed back to the merchant and the user to whom it belongs, and the marked order is returned. The ERP management system is used to perform secondary marking on the current transit station and the previous transit station of the marked order, and the number of marked orders returned by transit is counted. When the number of marked orders returned by transit at a certain transit station exceeds the threshold, the ERP management system issues an alarm to the current transit station and the previous transit station of the marked order, and traces the transportation route and transportation personnel information;

[0050] The output end of the order information collection module is connected to the input end of the order information processing module; the output end of the order information processing module is connected to the input end of the similarity model construction and analysis module; the output end of the similarity model construction and analysis module is connected to the input end of the user order data analysis module; the output end of the user order data analysis module is connected to the input end of the ERP management module.

[0051] Further, the order information collection module includes an order packaging information collection unit, an order characteristic mark detection unit, a marked order generation unit, a marked order initial image collection unit, a marked order transit site acquisition unit and a marked order transit image collection unit;

[0052] The order packaging information collection unit is used to collect order packaging information at the order sending site;

[0053] The order characteristic mark detection unit is used to detect the order characteristic mark based on the order packaging information;

[0054] The marked order generating unit is used to generate a marked order when an order characteristic flag is detected;

[0055] The marked order initial image acquisition unit is used to acquire an initial image of the marked order;

[0056] The marked order transit site acquisition unit is used to acquire all the transit sites of the marked order;

[0057] The marked order transit image acquisition unit is used to acquire the transit images of the marked order at each transit station;

[0058] The output end of the order packaging information collection unit is connected to the input end of the order characteristic mark detection unit; the output end of the order characteristic mark detection unit is connected to the input end of the marked order generation unit; the output end of the marked order generation unit is connected to the input end of the marked order initial image collection unit; the output end of the marked order initial image collection unit is connected to the input end of the marked order transfer site acquisition unit; the output end of the marked order transfer site acquisition unit is connected to the input end of the marked order transfer image collection unit; the output end of the marked order transfer image collection unit is connected to the input end of the order information processing module.

[0059] Further, the order information processing module includes a marked order initial image processing unit and a marked order transit image processing unit;

[0060] The marked order initial image processing unit is used to process the marked order initial image to obtain an initial vector of the marked order initial grayscale image;

[0061] The marked order transit image processing unit is used to process the transit images of the marked order at each transit station to obtain the transit quantity of the transit grayscale image of the marked order at each transit station;

[0062] The output end of the marked order initial image processing unit is connected to the input end of the marked order transit image processing unit; the output end of the marked order transit image processing unit is connected to the input end of the similarity model construction and analysis module.

[0063] Further, the similarity model construction and analysis module includes a similarity model construction unit and a similarity model analysis unit;

[0064] The similarity model building unit is used to build a marked order similarity calculation model;

[0065] The similarity model analysis unit is used to compare the initial vector of the initial grayscale image of the marked order with the intermediate transfer amount of the transfer grayscale image of the marked order at each transfer station to calculate the similarity of the marked order;

[0066] The output end of the similarity model building unit is connected to the similarity model analysis unit; the output end of the similarity model analysis unit is connected to the output end of the user order data analysis module.

[0067] Furthermore, the user order data analysis module includes a marked order similarity judgment unit, a user history order acquisition unit and a user transport damage return probability calculation unit;

[0068] The marked order similarity judgment unit is used to set a similarity threshold of the marked order, and when the similarity of the marked order does not exceed the threshold, it is determined that the marked order has caused transportation loss;

[0069] The user historical order acquisition unit is used to acquire historical order data of the user to which the marked order belongs after determining that the marked order has caused transportation damage;

[0070] The user damaged return probability calculation unit is used to calculate the probability that the user of the marked order applies for a return due to damage based on the historical order data of the user of the marked order;

[0071] The output end of the marked order similarity judgment unit is connected to the input end of the user history order acquisition unit; the output end of the user history order acquisition unit is connected to the input end of the user damage return probability calculation unit; the output end of the user damage return probability calculation unit is connected to the input end of the ERP management module;

[0072] The ERP management module includes a user damage return judgment unit, a first output unit and a second output unit;

[0073] The user damaged return judgment unit is used to set a probability threshold for applying for a return due to damage to the marked order. When the probability of the user to whom the marked order belongs applying for a return due to damage exceeds the probability threshold for applying for a return due to damage to the marked order, it is determined that the marked order is applied for a return due to damage.

[0074] The first output unit is used to feed back the transit image of the marked order at the current transit site to the merchant and the corresponding user, and return the marked order;

[0075] The second output unit is used to use the ERP management system to re-mark the current transit site and the previous transit site of the marked order, and count the number of marked orders returned by transit. When the number of marked orders returned by a certain transit site exceeds a threshold, the ERP management system will alarm the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information;

[0076] The output end of the user damaged return judgment unit is connected to the input end of the first output unit; the output end of the first output unit is connected to the input end of the second output unit.

[0077] Compared with the prior art, the beneficial effects achieved by the present invention are: the present invention can detect and screen orders with order characteristic marks, monitor the transportation damage of the marked orders at various transit stations, detect and feedback the marked orders that are damaged during transportation, and further analyze the users to whom the marked orders belong, so as to help users in need to return the damaged marked orders in time, improve the efficiency of transportation and solve the problem of consumers spending extra time and energy on handling damaged orders. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0079] Figure 1 It is a structural schematic diagram of an order data analysis system based on an ERP management system of the present invention;

[0080] Figure 2 It is a flow chart of an order data analysis method based on an ERP management system of the present invention. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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.

[0082] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0083] An order data analysis method based on an ERP management system, the method comprising the following steps:

[0084] Step S1: collecting order packaging information at the order sending site, detecting order characteristic marks based on the order packaging information, and generating a marked order when the order characteristic marks are detected;

[0085] Step S2: collecting an initial image of the marked order, processing the initial image of the marked order, and obtaining an initial vector of an initial grayscale image of the marked order;

[0086] Step S3: Obtain all the transfer stations of the marked order. When the marked order arrives at each transfer station, obtain the transfer images of the marked order at each transfer station respectively and process the transfer images of the marked order at each transfer station to obtain the transfer quantity of the transfer grayscale image of the marked order at each transfer station;

[0087] Step S4: construct a marked order similarity calculation model, compare the initial vector of the initial grayscale image of the marked order with the transfer vector of the transfer grayscale image of the marked order at each transfer station, and calculate the similarity of the marked order;

[0088] Step S5: setting a similarity threshold of the marked order. When the similarity of the marked order does not exceed the threshold, the marked order is determined to have transportation damage, the historical order data of the user to whom the marked order belongs is obtained, and the probability that the user to whom the marked order belongs applies for a return due to transportation damage is calculated;

[0089] Step S6: setting a probability threshold for a marked order to be returned due to transportation damage. When the probability of the user to whom the marked order belongs to apply for a return due to transportation damage exceeds the probability threshold for the marked order to be returned due to transportation damage, the marked order is determined to be returned due to transportation damage, and the transit image of the marked order at the current transit station is fed back to the merchant and the user to whom it belongs, and the marked order is returned;

[0090] Step S7: Use the ERP management system to re-mark the current transit site and the previous transit site of the marked order, and count the number of marked orders returned in transit. When the number of marked orders returned in transit at a certain transit site exceeds the threshold, the ERP management system will alarm the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information.

[0091] Further, in steps S1-S2, the order packaging information includes order transportation information and order characteristic flags;

[0092] The order transportation information includes the order number, waybill number, shipping address and delivery address;

[0093] The order characteristic marks include a fragile mark, a handle with care mark and a moisture-proof mark;

[0094] When the order feature flag is detected as a fragile flag, a marked order is generated When the order feature information is detected as a handle with care sign, a marked order is generated When the order feature information is detected as a moisture-proof mark, a marked order is generated

[0095] Collect and mark orders Initial image Marking orders Initial image Perform grayscale processing and generate marked orders The initial grayscale image Marking orders The initial grayscale image Perform edge computing;

[0096] According to the formula, the marked orders The initial grayscale image Each pixel has:

[0097]

[0098] Where G(x,y) represents the Gaussian filter function; E x0,n (x,y) indicates the marked order The initial grayscale image The horizontal result after each pixel is Gaussian filtered; E y0,n (x,y) indicates the marked order The initial grayscale image The vertical result of each pixel after Gaussian filtering;

[0099] Marking orders The initial grayscale image The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated:

[0100]

[0101] Among them, E x0,avg (x,y) indicates the marked order The initial grayscale image The average value of the horizontal result after Gaussian filtering of each pixel; E y0,avg (x,y) indicates the marked order The initial grayscale image The vertical average of each pixel after Gaussian filtering;

[0102] Marking Orders The initial grayscale image The initial vector is:

[0103]

[0104] in, Indicates marking order The initial grayscale image The initial vector of .

[0105] Further, in step S3,

[0106] Get Mark Order All transfer stations B={b1,b2,...,b z}; Corresponding mark order Transit images at various transit points Grayscale processing is performed on the transfer images of the marked orders at each transfer station to generate marked orders Grayscale images of the transfer at each transfer station Marking orders The transferred grayscale image F is edge computed and the marked order is The horizontal and vertical results of each pixel of the transferred grayscale image F after Gaussian filtering are averaged to obtain the marked order The transit quantity of the transit grayscale image at each transit station

[0107] Furthermore, in step S4, constructing a marked order similarity calculation model includes:

[0108] Get Mark Order The initial grayscale image The initial vector Get Mark Order The transit quantity of the transit grayscale image at each transit station

[0109] Calculate Mark Order Similarity:

[0110]

[0111] in, Indicates marking order Similarity at the jth transfer station; Indicates marking order The initial grayscale image The initial vector of Indicates marking order Grayscale image of the transfer at the jth transfer station The center steering volume.

[0112] Further, in steps S5-S6, set the marked order Ai The similarity threshold of , denoted as cos(θ0);

[0113] When marking an order When the similarity exceeds the threshold, the order is marked. No shipping damage occurred;

[0114] When marking an order When the similarity does not exceed the threshold, the order is marked. In case of shipping damage, get marked order User C k Historical order data;

[0115] The acquisition of marked order User C g The historical order data includes:

[0116] Get Mark Order User C h The number of orders that incur shipping damage, denoted as U h ;

[0117] Get Mark Order User C h The number of orders that apply for return due to shipping damage, denoted as V h ;

[0118] Calculate Mark Order User C h Probability of applying for return due to shipping damage:

[0119]

[0120] Among them, p h Indicates marking order User C h The probability of applying for return due to shipping damage;

[0121] Set the probability threshold for applying for return due to shipping damage for marked orders, denoted as p0; User C h When the probability of applying for a return due to shipping damage exceeds the probability threshold for marking an order for a return due to shipping damage, the order will be marked The transit image at the current transit station is fed back to the merchant and the user, and the order is marked return.

[0122] An order data analysis system based on an ERP management system, the system comprising an order information collection module, an order information processing module, a similarity model building and analysis module, a user order data analysis module and an ERP management module;

[0123] The order information collection module is used to collect order packaging information, detect order feature marks, generate marked orders, collect the initial image of the marked order, mark all the transit sites of the order and the transit images of the marked order at each transit site; the order information processing module processes the initial image of the marked order and the transit images of the marked order at each transit site to obtain the initial vector of the initial grayscale image of the marked order and the diversion amount of the transit grayscale image of the marked order at each transit site; the similarity model construction and analysis module is used to construct a marked order similarity calculation model, compare the initial vector of the initial grayscale image of the marked order with the diversion amount of the transit grayscale image of the marked order at each transit site, and calculate the similarity of the marked order; the user order data analysis module is used to set the similarity threshold of the marked order, and when the similarity of the marked order does not exceed the threshold, it is judged that the marked order is marked. If a marked order generates transportation damage, the historical order data of the user to whom the marked order belongs is obtained, and the probability that the user to whom the marked order belongs applies for return due to transportation damage is calculated; the ERP management module is used to set a probability threshold for the marked order to apply for return due to transportation damage. When the probability that the user to whom the marked order belongs applies for return due to transportation damage exceeds the probability threshold for the marked order to apply for return due to transportation damage, it is determined that the marked order is applied for return due to transportation damage, and the transit image of the marked order at the current transit station is fed back to the merchant and the user to whom it belongs, and the marked order is returned. The ERP management system is used to perform secondary marking on the current transit station and the previous transit station of the marked order, and the number of marked orders returned by transit is counted. When the number of marked orders returned by transit at a certain transit station exceeds the threshold, the ERP management system issues an alarm to the current transit station and the previous transit station of the marked order, and traces the transportation route and transportation personnel information;

[0124] The output end of the order information collection module is connected to the input end of the order information processing module; the output end of the order information processing module is connected to the input end of the similarity model construction and analysis module; the output end of the similarity model construction and analysis module is connected to the input end of the user order data analysis module; the output end of the user order data analysis module is connected to the input end of the ERP management module.

[0125] Further, the order information collection module includes an order packaging information collection unit, an order characteristic mark detection unit, a marked order generation unit, a marked order initial image collection unit, a marked order transit site acquisition unit and a marked order transit image collection unit;

[0126] The order packaging information collection unit is used to collect order packaging information at the order sending site;

[0127] The order characteristic mark detection unit is used to detect the order characteristic mark based on the order packaging information;

[0128] The marked order generating unit is used to generate a marked order when an order characteristic flag is detected;

[0129] The marked order initial image acquisition unit is used to acquire an initial image of the marked order;

[0130] The marked order transit site acquisition unit is used to acquire all the transit sites of the marked order;

[0131] The marked order transit image acquisition unit is used to acquire the transit images of the marked order at each transit station;

[0132] The output end of the order packaging information collection unit is connected to the input end of the order characteristic mark detection unit; the output end of the order characteristic mark detection unit is connected to the input end of the marked order generation unit; the output end of the marked order generation unit is connected to the input end of the marked order initial image collection unit; the output end of the marked order initial image collection unit is connected to the input end of the marked order transfer site acquisition unit; the output end of the marked order transfer site acquisition unit is connected to the input end of the marked order transfer image collection unit; the output end of the marked order transfer image collection unit is connected to the input end of the order information processing module.

[0133] Further, the order information processing module includes a marked order initial image processing unit and a marked order transit image processing unit;

[0134] The marked order initial image processing unit is used to process the marked order initial image to obtain an initial vector of the marked order initial grayscale image;

[0135] The marked order transit image processing unit is used to process the transit images of the marked order at each transit station to obtain the transit quantity of the transit grayscale image of the marked order at each transit station;

[0136] The output end of the marked order initial image processing unit is connected to the input end of the marked order transit image processing unit; the output end of the marked order transit image processing unit is connected to the input end of the similarity model construction and analysis module.

[0137] Further, the similarity model construction and analysis module includes a similarity model construction unit and a similarity model analysis unit;

[0138] The similarity model building unit is used to build a marked order similarity calculation model;

[0139] The similarity model analysis unit is used to compare the initial vector of the initial grayscale image of the marked order with the intermediate transfer amount of the transfer grayscale image of the marked order at each transfer station to calculate the similarity of the marked order;

[0140] The output end of the similarity model building unit is connected to the similarity model analysis unit; the output end of the similarity model analysis unit is connected to the output end of the user order data analysis module.

[0141] Furthermore, the user order data analysis module includes a marked order similarity judgment unit, a user history order acquisition unit and a user transport damage return probability calculation unit;

[0142] The marked order similarity judgment unit is used to set a similarity threshold of the marked order, and when the similarity of the marked order does not exceed the threshold, it is determined that the marked order has caused transportation loss;

[0143] The user historical order acquisition unit is used to acquire historical order data of the user to which the marked order belongs after determining that the marked order has caused transportation damage;

[0144] The user damaged return probability calculation unit is used to calculate the probability that the user of the marked order applies for a return due to damage based on the historical order data of the user to whom the marked order belongs;

[0145] The output end of the marked order similarity judgment unit is connected to the input end of the user history order acquisition unit; the output end of the user history order acquisition unit is connected to the input end of the user damage return probability calculation unit; the output end of the user damage return probability calculation unit is connected to the input end of the ERP management module;

[0146] The ERP management module includes a user damage return judgment unit, a first output unit and a second output unit;

[0147] The user damaged return judgment unit is used to set a probability threshold for applying for a return due to damage to the marked order. When the probability of the user to whom the marked order belongs applying for a return due to damage exceeds the probability threshold for applying for a return due to damage to the marked order, it is determined that the marked order is applied for a return due to damage.

[0148] The first output unit is used to feed back the transit image of the marked order at the current transit site to the merchant and the corresponding user, and return the marked order;

[0149] The second output unit is used to use the ERP management system to re-mark the current transit site and the previous transit site of the marked order, and count the number of marked orders returned by transit. When the number of marked orders returned by a certain transit site exceeds a threshold, the ERP management system will alarm the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information;

[0150] The output end of the user damaged return judgment unit is connected to the input end of the first output unit; the output end of the first output unit is connected to the input end of the second output unit.

[0151] In this embodiment:

[0152] Collect order transportation information such as order number, waybill number, shipping address and delivery address;

[0153] Collect order feature marks such as fragile mark, handle with care mark and moisture-proof mark;

[0154] The order feature flag is detected as a fragile flag, and a marked order is generated

[0155] Collect and mark orders Initial image Marking orders Initial image Perform grayscale processing and generate marked orders The initial grayscale image Marking orders The initial grayscale image Perform edge computing;

[0156] According to the formula, the marked orders The initial grayscale image Each pixel has:

[0157]

[0158] Where G(x,y) represents the Gaussian filter function; E x0,n (x,y) indicates the marked order The initial grayscale image The horizontal result after each pixel of E is Gaussian filtered; y0,n (x,y) indicates the marked order The initial grayscale image The vertical result of each pixel after Gaussian filtering;

[0159] Marking orders The initial grayscale image The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated:

[0160]

[0161] Among them, E x0,avg (x,y) indicates the marked order The initial grayscale image The average value of the horizontal result after Gaussian filtering of each pixel; E y0,avg (x,y) indicates the marked order The initial grayscale image The vertical average of each pixel after Gaussian filtering;

[0162] At this point, mark the order The initial grayscale image The number of pixels n = 50;

[0163] Mark the order The initial grayscale image The initial vector is:

[0164]

[0165] in, Indicates marking order The initial grayscale image The initial vector of .

[0166] Get Mark Order All transit stations B={b1,b2,...,b5}; corresponding marked orders Transit images at various transit points Grayscale processing is performed on the transfer images of the marked orders at each transfer station to generate marked orders Grayscale images of the transfer at each transfer station Marking orders The transferred grayscale image F is edge computed and the marked order is The horizontal and vertical results of each pixel of the transferred grayscale image F after Gaussian filtering are averaged to obtain the marked order The transit quantity of the transit grayscale image at each transit station

[0167] Furthermore, in step S4, constructing a marked order similarity calculation model includes:

[0168] Get Indicates marking order The initial grayscale image The initial vector Get Mark Order The transfer volume of the transfer grayscale image at the third transfer station

[0169] Calculate Mark Order Similarity:

[0170]

[0171] in, Indicates marking order Similarity at the third transfer station; Indicates marking order The initial grayscale image The initial vector of Indicates marking order Grayscale image of the transfer at the third transfer station The center steering volume.

[0172] Set Mark Order A i The similarity threshold is cos(θ0)=0.95;

[0173] Because the marking order When the similarity does not exceed the threshold, the order is marked. In case of shipping damage, get marked order Historical order data of user C4;

[0174] The acquisition of marked order The historical order data of user C4 includes:

[0175] Get Mark Order The number of orders with shipping damage for user C4 is U4=10;

[0176] Get Mark Order The number of orders for which user C4 applied for return due to shipping damage is V4 = 9;

[0177] Calculate Mark Order The probability that user C4 applies for a return due to shipping damage:

[0178]

[0179] Set the probability threshold for marking an order for return due to shipping damage p0 = 0.8;

[0180] Because the marking order The probability of user C5 applying for a return due to shipping damage exceeds the probability threshold for marking an order for a return due to shipping damage, so the order is marked The transit image at the current transit station is fed back to the merchant and the user, and the order is marked return.

[0181] The ERP management system is used to re-mark the current transit site and the previous transit site of the marked order, and the number of marked orders returned during transit is counted. When the number of marked orders returned during transit at a certain transit site exceeds the threshold, the ERP management system will issue an alarm to the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information.

[0182] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0183] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An order data analysis method based on ERP management system, characterized in that: The method comprises the following steps: Step S1: collecting order packaging information at the order sending site, detecting order characteristic marks based on the order packaging information, and generating a marked order when the order characteristic marks are detected; Step S2: collecting an initial image of the marked order, processing the initial image of the marked order, and obtaining an initial vector of an initial grayscale image of the marked order; Step S3: Obtain all the transfer stations of the marked order. When the marked order arrives at each transfer station, obtain the transfer images of the marked order at each transfer station respectively and process the transfer images of the marked order at each transfer station to obtain the transfer quantity of the transfer grayscale image of the marked order at each transfer station; Step S4: construct a marked order similarity calculation model, compare the initial vector of the initial grayscale image of the marked order with the transfer vector of the transfer grayscale image of the marked order at each transfer station, and calculate the similarity of the marked order; Step S5: setting a similarity threshold of the marked order. When the similarity of the marked order does not exceed the threshold, the marked order is determined to have transportation damage, the historical order data of the user to whom the marked order belongs is obtained, and the probability that the user to whom the marked order belongs applies for a return due to transportation damage is calculated; Step S6: setting a probability threshold for a marked order to be returned due to transportation damage. When the probability of the user to whom the marked order belongs to apply for a return due to transportation damage exceeds the probability threshold for the marked order to be returned due to transportation damage, the marked order is determined to be returned due to transportation damage, and the transit image of the marked order at the current transit station is fed back to the merchant and the user to whom it belongs, and the marked order is returned; Step S7: Use the ERP management system to re-mark the current transit site and the previous transit site of the marked order, and count the number of marked orders returned by transit. When the number of marked orders returned by a certain transit site exceeds the threshold, the ERP management system will alert the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information; In steps S1-S2, the order packaging information includes order transportation information and order characteristic flags; The order transportation information includes the order number, waybill number, shipping address and delivery address; The order characteristic marks include a fragile mark, a handle with care mark and a moisture-proof mark; When the order feature flag is detected as a fragile flag, a marked order is generated When the order feature information is detected as a handle with care sign, a marked order is generated When the order feature information is detected as a moisture-proof mark, a marked order is generated Collect and mark orders Initial image Marking Orders Initial image Perform grayscale processing and generate marked orders The initial grayscale image Marking Orders The initial grayscale image Perform edge computing; According to the formula, the marked orders The initial grayscale image Each pixel has: Where G(x,y) represents the Gaussian filter function; E x0,n (x,y) indicates the marked order The initial grayscale image The horizontal result after each pixel of E is Gaussian filtered; y0,n (x,y) indicates the marked order The initial grayscale image The vertical result of each pixel after Gaussian filtering; Marking orders The initial grayscale image The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated: Among them, E x0,avg (x,y) indicates the marked order The initial grayscale image The average value of the horizontal result after Gaussian filtering of each pixel; E y0,avg (x,y) indicates the marked order The initial grayscale image The vertical average of each pixel after Gaussian filtering; Marking Orders The initial grayscale image The initial vector is: in, Indicates marking order The initial grayscale image The initial vector of In step S3, Get Mark Order All transfer stations B={b1,b2,...,b z }; Corresponding mark order Transit images at various transit points Grayscale processing is performed on the transfer images of the marked orders at each transfer station to generate marked orders Grayscale images of the transfer at each transfer station Marking orders The transferred grayscale image F is edge computed and the marked order is The horizontal and vertical results of each pixel of the transferred grayscale image F after Gaussian filtering are averaged to obtain the marked order The transit quantity of the transit grayscale image at each transit station In step S4, constructing a marked order similarity calculation model includes: Get Mark Order The initial grayscale image The initial vector Get Mark Order The transit quantity of the transit grayscale image at each transit station Calculate Mark Order Similarity: in, Indicates marking order Similarity at the jth transfer station; Indicates marking order The initial grayscale image The initial vector of Indicates marking order Grayscale image of the transfer at the jth transfer station The center steering volume.

2. The order data analysis method based on the ERP management system according to claim 1, characterized in that: In steps S5-S6, set the marked order A i The similarity threshold of , denoted as cos(θ0); When marking an order When the similarity exceeds the threshold, the order is marked. No shipping damage occurred; When marking an order When the similarity does not exceed the threshold, the order is marked. In case of shipping damage, get marked order User C k Historical order data; The acquisition of marked order User C g The historical order data includes: Get Mark Order User C h The number of orders that incur shipping damage, denoted as U h ; Get Mark Order User C h The number of orders that were returned due to shipping damage, denoted as V h ; Calculate Mark Order User C h Probability of applying for return due to shipping damage: Among them, p h Indicates marking order User C h The probability of applying for return due to shipping damage; Set the probability threshold for applying for return due to shipping damage for marked orders, denoted as p0; User C h When the probability of applying for a return due to shipping damage exceeds the probability threshold for marking an order for a return due to shipping damage, the order will be marked The transit image at the current transit station is fed back to the merchant and the user, and the order is marked return.

3. An order data analysis system based on an ERP management system using an order data analysis method based on an ERP management system according to any one of claims 1 to 2, characterized in that: The system includes an order information collection module, an order information processing module, a similarity model building and analysis module, a user order data analysis module and an ERP management module; The order information collection module is used to collect order packaging information, detect order feature marks, generate marked orders, collect initial images of marked orders, all transit sites of marked orders and transit images of marked orders at various transit sites; The order information processing module processes the initial image of the marked order and the transfer images of the marked order at each transfer station to obtain the initial vector of the initial grayscale image of the marked order and the transfer vector of the transfer grayscale image of the marked order at each transfer station; The similarity model building and analysis module is used to build a marked order similarity calculation model, compare the initial vector of the initial grayscale image of the marked order with the transfer volume of the transfer grayscale image of the marked order at each transfer station, and calculate the similarity of the marked order; The user order data analysis module is used to set a similarity threshold for marked orders. When the similarity of marked orders does not exceed the threshold, it is determined that the marked order has caused transportation damage, the historical order data of the user to whom the marked order belongs is obtained, and the probability that the user to whom the marked order belongs applies for return due to transportation damage is calculated; the ERP management module is used to set a probability threshold for marked orders to apply for return due to transportation damage. When the probability that the user to whom the marked order belongs applies for return due to transportation damage exceeds the probability threshold for marked orders to apply for return due to transportation damage, it is determined that the marked order applies for return due to transportation damage, the transit image of the marked order at the current transit site is fed back to the merchant and the user to whom it belongs, and the marked order is returned. The ERP management system is used to perform secondary marking on the current transit site and the previous transit site of the marked order, and the number of marked orders returned by transit is counted. When the number of marked orders returned by transit at a certain transit site exceeds the threshold, the ERP management system issues an alarm to the current transit site and the previous transit site of the marked order, and traces the transportation route and transportation personnel information; The output end of the order information collection module is connected to the input end of the order information processing module; the output end of the order information processing module is connected to the input end of the similarity model construction and analysis module; The output end of the similarity model construction and analysis module is connected to the input end of the user order data analysis module; the output end of the user order data analysis module is connected to the input end of the ERP management module.

4. The order data analysis system based on the ERP management system according to claim 3 is characterized in that: The order information collection module includes an order packaging information collection unit, an order characteristic mark detection unit, a marked order generation unit, a marked order initial image collection unit, a marked order transit site acquisition unit and a marked order transit image collection unit; The order packaging information collection unit is used to collect order packaging information at the order sending site; The order characteristic mark detection unit is used to detect the order characteristic mark based on the order packaging information; The marked order generating unit is used to generate a marked order when an order characteristic flag is detected; The marked order initial image acquisition unit is used to acquire an initial image of the marked order; The marked order transit site acquisition unit is used to acquire all the transit sites of the marked order; The marked order transit image acquisition unit is used to acquire the transit images of the marked order at each transit station; The output end of the order packaging information collection unit is connected to the input end of the order characteristic mark detection unit; the output end of the order characteristic mark detection unit is connected to the input end of the marked order generation unit; the output end of the marked order generation unit is connected to the input end of the marked order initial image collection unit; the output end of the marked order initial image collection unit is connected to the input end of the marked order transfer site acquisition unit; the output end of the marked order transfer site acquisition unit is connected to the input end of the marked order transfer image collection unit; the output end of the marked order transfer image collection unit is connected to the input end of the order information processing module.

5. The order data analysis system based on the ERP management system according to claim 4 is characterized in that: The order information processing module includes a marked order initial image processing unit and a marked order transit image processing unit; The marked order initial image processing unit is used to process the marked order initial image to obtain an initial vector of the marked order initial grayscale image; The marked order transit image processing unit is used to process the transit images of the marked order at each transit station to obtain the transit quantity of the transit grayscale image of the marked order at each transit station; The output end of the marked order initial image processing unit is connected to the input end of the marked order transit image processing unit; the output end of the marked order transit image processing unit is connected to the input end of the similarity model construction and analysis module.

6. The order data analysis system based on the ERP management system according to claim 5 is characterized by: The similarity model construction and analysis module includes a similarity model construction unit and a similarity model analysis unit; The similarity model building unit is used to build a marked order similarity calculation model; The similarity model analysis unit is used to compare the initial vector of the initial grayscale image of the marked order with the intermediate transfer amount of the transfer grayscale image of the marked order at each transfer station to calculate the similarity of the marked order; The output end of the similarity model building unit is connected to the similarity model analysis unit; the output end of the similarity model analysis unit is connected to the output end of the user order data analysis module.

7. The order data analysis system based on the ERP management system according to claim 6 is characterized by: The user order data analysis module includes a marked order similarity judgment unit, a user history order acquisition unit and a user transport damage return probability calculation unit; The marked order similarity judgment unit is used to set a similarity threshold of the marked order, and when the similarity of the marked order does not exceed the threshold, it is determined that the marked order has caused transportation loss; The user historical order acquisition unit is used to acquire historical order data of the user to which the marked order belongs after determining that the marked order has caused transportation damage; The user damaged return probability calculation unit is used to calculate the probability that the user of the marked order applies for a return due to damage based on the historical order data of the user of the marked order; The output end of the marked order similarity judgment unit is connected to the input end of the user history order acquisition unit; the output end of the user history order acquisition unit is connected to the input end of the user damage return probability calculation unit; the output end of the user damage return probability calculation unit is connected to the input end of the ERP management module; The ERP management module includes a user damage return judgment unit, a first output unit and a second output unit; The user damaged return judgment unit is used to set a probability threshold for applying for a return due to damage to the marked order. When the probability of the user to whom the marked order belongs applying for a return due to damage exceeds the probability threshold for applying for a return due to damage to the marked order, it is determined that the marked order is applied for a return due to damage. The first output unit is used to feed back the transit image of the marked order at the current transit site to the merchant and the corresponding user, and return the marked order; The second output unit is used to use the ERP management system to re-mark the current transit site and the previous transit site of the marked order, and count the number of marked orders returned by transit. When the number of marked orders returned by a certain transit site exceeds a threshold, the ERP management system will alarm the current transit site and the previous transit site of the marked order, and trace the transportation route and transportation personnel information; The output end of the user damaged return judgment unit is connected to the input end of the first output unit; the output end of the first output unit is connected to the input end of the second output unit.

Images