Sales link abnormity dynamic detection method and device, terminal and storage medium
By acquiring and calculating the dynamic baseline and time deviation values of order nodes, and combining data integrity and the status of associated nodes, the problem of low accuracy in anomaly detection in existing technologies is solved, and accurate risk identification in complex business processes is achieved.
Patent Information
- Application Number
- CN202510956875.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-11
- Publication Date
- 2025-10-31
AI Technical Summary
Existing technologies consider only a single factor in anomaly detection, resulting in low accuracy and difficulty in accurately identifying true risks in complex business processes.
By obtaining the initial dynamic baseline and the dates of each node corresponding to the target order, the target dynamic baseline is determined based on the initial dynamic baseline and the dates of each node. The time deviation value is calculated, and the anomaly level is determined by combining the data integrity and the status of related nodes.
It improves the accuracy of anomaly detection, enabling precise identification of real risk situations in complex business processes, avoiding resource waste and timely handling of potential risks.
Smart Images
Figure CN120873675A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computers, and more particularly to a method, apparatus, terminal, and storage medium for dynamic detection of abnormalities in a sales chain. Background Technology
[0002] In the current business operating environment, anomaly detection in order processing remains a highly challenging problem. Existing systems only consider single-factor metrics for anomaly detection, such as monitoring the processing time of a single node. They rarely consider multiple factors comprehensively, such as node processing time, data integrity, and the status of related nodes, resulting in low accuracy of anomaly detection and difficulty in accurately identifying true risks in complex business processes.
[0003] Therefore, existing technologies still need improvement and development. Summary of the Invention
[0004] The technical problem to be solved by the present invention is to provide a method, device, terminal and storage medium for dynamic detection of anomalies in the sales chain, in order to address the above-mentioned deficiencies of the prior art. The aim is to solve the problem that the existing technology only considers a single factor indicator, resulting in low accuracy of anomaly detection and difficulty in accurately identifying the real risk situation in complex business processes.
[0005] The technical solution adopted by this invention to solve the problem is as follows:
[0006] In a first aspect, embodiments of the present invention provide a method for dynamic detection of anomalies in a sales chain, wherein the method includes:
[0007] Obtain the initial dynamic baseline and the node dates corresponding to the target order, and determine the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date;
[0008] The time deviation value is determined based on the target dynamic baseline and the date of each node;
[0009] Obtain the data integrity and associated node status corresponding to the target order, and determine the anomaly level corresponding to the target order based on the data integrity, the associated node status, and the time deviation value.
[0010] In one implementation method, the method for obtaining an initial dynamic baseline includes:
[0011] Retrieve historical order data and historical marked date ranges;
[0012] The target sliding time window is determined based on the historical order data and the historical marked date range;
[0013] The initial dynamic baseline is determined by calculating the historical order data based on the target sliding time window.
[0014] In one implementation method, determining a target sliding time window based on the historical order data and the historical marked date range includes:
[0015] Determine whether the historical order data falls within each of the historical marked date ranges;
[0016] If so, then obtain the first sliding time window corresponding to the historical marked date interval where the historical order data is located, and use the first sliding time window as the target sliding time window;
[0017] If not, then the preset second sliding time window will be used as the target sliding time window.
[0018] In one implementation method, obtaining the first sliding time window corresponding to the historical marked date interval where the historical order data is located includes:
[0019] Obtain the historical business operation rules corresponding to the historical marked date range;
[0020] The first sliding time window corresponding to the historical marked date interval is determined according to the historical business operation rules.
[0021] In one implementation method, determining the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date includes:
[0022] Obtain the marked date range and determine whether the date of each node is within the marked date range;
[0023] If so, obtain the business operation rules corresponding to the marked date interval, adjust the initial dynamic baseline corresponding to the marked date interval according to the business operation rules, and determine the target dynamic baseline;
[0024] If not, the initial dynamic baseline corresponding to each node date shall be used as the target dynamic baseline.
[0025] In one implementation method, the initial dynamic baseline is determined by calculating the historical order data based on the target sliding time window, including:
[0026] Calculate the average processing time and standard deviation of processing time for each node of the historical order data within the target sliding time window;
[0027] The initial dynamic baseline is determined based on the average processing time and the standard deviation of the processing time.
[0028] In one implementation method, determining the anomaly level of the target order based on the data integrity, the status of associated nodes, and the time deviation value includes:
[0029] Obtain the weights corresponding to the data integrity, the status of the associated nodes, and the time deviation value;
[0030] An anomaly assessment score is determined based on the data completeness, the status of the associated nodes, the time deviation value, and each of the weights.
[0031] Obtain the abnormal score range, and determine the abnormal level based on the abnormal assessment score and the abnormal score range.
[0032] Secondly, embodiments of the present invention also provide a sales link anomaly dynamic detection device, wherein the sales link anomaly dynamic detection device includes:
[0033] The target dynamic baseline determination module is used to obtain the initial dynamic baseline and the node dates corresponding to the target order, and determine the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date.
[0034] The time deviation value calculation module is used to determine the time deviation value based on the target dynamic baseline and the date of each node.
[0035] The anomaly level determination module is used to obtain the data integrity and associated node status corresponding to the target order, and determine the anomaly level corresponding to the target order based on the data integrity, the associated node status, and the time deviation value.
[0036] Thirdly, embodiments of the present invention also provide a terminal, the terminal including a memory and one or more processors; the memory stores one or more programs; the programs include instructions for executing the sales link anomaly dynamic detection method as described above; the processor is used to execute the programs.
[0037] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a plurality of instructions, wherein the instructions are adapted to be loaded and executed by a processor to implement any of the above-described sales link anomaly dynamic detection methods.
[0038] The beneficial effects of this invention are as follows: This invention obtains the initial dynamic baseline and the dates of each node corresponding to the target order; determines the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and the dates of each node; determines the time deviation value based on the target dynamic baseline and the dates of each node; obtains the data integrity and associated node status corresponding to the target order; and determines the anomaly level corresponding to the target order based on the data integrity, associated node status, and time deviation value. Because this invention calculates the time deviation value based on the target dynamic baseline corresponding to the node date, and determines the anomaly level of the target order based on the time deviation value, data integrity, and associated node status, it effectively solves the problem that existing technologies only consider a single factor indicator, resulting in low accuracy in anomaly detection and difficulty in accurately identifying the true risk situation in complex business processes. Attached Figure Description
[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0040] Figure 1 This is a flowchart illustrating the sales link anomaly dynamic detection method provided in this embodiment of the invention.
[0041] Figure 2 This is a schematic flowchart of a specific embodiment of the sales link anomaly dynamic detection method provided in this invention.
[0042] Figure 3 This is a schematic diagram of the internal modules of the sales link anomaly dynamic detection device provided in an embodiment of the present invention.
[0043] Figure 4 This is a schematic diagram of the terminal provided in the embodiment of the present invention. Detailed Implementation
[0044] This invention discloses a method, apparatus, terminal, and storage medium for dynamic detection of anomalies in the sales chain. To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention.
[0045] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.
[0046] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.
[0047] In the current business operating environment, anomaly detection in order processing remains a highly challenging problem. Existing systems only consider single-factor metrics for anomaly detection, such as the processing time of a single node. They rarely consider multiple factors comprehensively, such as node processing time, data integrity, and the status of related nodes, resulting in low accuracy of anomaly detection and difficulty in accurately identifying true risks in complex business processes.
[0048] To address the aforementioned shortcomings of existing technologies, this invention provides a method for dynamic detection of anomalies in the sales chain. The method involves obtaining an initial dynamic baseline and the dates of each node corresponding to the target order; determining the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and the dates of each node; determining the time deviation value based on the target dynamic baseline and the dates of each node; obtaining the data integrity and associated node status corresponding to the target order; and determining the anomaly level of the target order based on the data integrity, associated node status, and time deviation value. Because this invention calculates the time deviation value based on the target dynamic baseline corresponding to the node dates, and determines the anomaly level of the target order based on the time deviation value, data integrity, and associated node status, it effectively solves the problem that existing technologies only consider a single factor indicator, resulting in low accuracy in anomaly detection and difficulty in accurately identifying true risk situations in complex business processes.
[0049] Exemplary method:
[0050] like Figure 1 As shown, the method includes:
[0051] Step S100: Obtain the initial dynamic baseline and the node dates corresponding to the target order, and determine the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date.
[0052] In simple terms, an order involves several stages. If the processing time of any stage is too long, the order may be considered abnormal. For example, a shopping order might include stages such as merchant order acceptance, order sorting and packaging, logistics allocation, and logistics pickup. If the logistics pickup stage takes too long, an anomaly may have occurred (e.g., the package was not properly picked up, or the logistics company received the package but failed to scan and enter it into the logistics system). The initial dynamic baseline includes the processing time baselines for each stage of an order for each date throughout the year. To accurately determine whether the processing time of each stage of a target order is abnormal, a target dynamic baseline is determined based on the dates of each stage of the target order and the initial dynamic baseline. The processing time of each stage is then assessed based on this target dynamic baseline.
[0053] In one implementation, the method for obtaining the initial dynamic baseline includes:
[0054] Step S101: Obtain historical order data and historical marked date range;
[0055] Step S102: Determine the target sliding time window based on the historical order data and the historical marked date range;
[0056] Step S103: Calculate the historical order data based on the target sliding time window to determine the initial dynamic baseline.
[0057] Specifically, historical order data is collected from various systems, including historical data related to order processing. This includes order creation time, start and end times of each node in the processing flow, and relevant business parameter data. This data should cover different time periods, including normal working days, holidays, and promotional periods. After acquiring historical order data, it undergoes preprocessing, such as data cleaning to remove outliers and erroneous data, and data normalization to ensure data format consistency.
[0058] The initial dynamic baseline is obtained through pre-analysis of the processing time of each node in historical orders. Before generating the initial dynamic baseline based on historical orders, the order volume corresponding to each date in a year can be determined based on historical order data throughout the year. The date-order volume curve for that year is then determined based on the order volume corresponding to each date. For example, based on historical order data from 2024, the order volume corresponding to different dates in 2024 is obtained, and thus the date-order volume curve for 2024 is generated based on the order volume corresponding to each date. The slope of the date-order volume curve determines the historical marked date intervals. For example, near the date of a certain holiday, the order volume increases, so the slope of the date-order volume curve increases. After the holiday, the order volume will return to its original level, and the slope will be negative during this recovery process. Each historical marked date interval can be labeled according to the corresponding event (e.g., if the historical marked date interval is before or after holiday A in the year, then the historical marked date interval can be set as holiday A). In addition to the above methods, historical marked date intervals can also be manually set and labeled based on experience.
[0059] When the order volume and marketing strategies of merchants differ within a historical marked date range, the normal processing time of each node in the historical order also varies. To avoid the target sliding time window failing to cover the processing time of nodes in the order, this embodiment determines the target sliding time window corresponding to the historical order data based on the historical order data and each historical marked date range. Specifically, this includes: Step S1021, determining whether the historical order data is located within each of the historical marked date ranges; Step S1022, if yes, obtaining the first sliding time window corresponding to the historical marked date range where the historical order data is located, and using the first sliding time window as the target sliding time window; Step S1023, if no, using a preset second sliding time window as the target sliding time window. By automatically adjusting the sliding time window used to calculate the initial dynamic baseline for special time periods such as holidays and promotional periods, it is ensured that the initial dynamic baseline can adapt to the business operation rules during special periods.
[0060] Once the target sliding time window is determined, historical order data is calculated based on the target sliding time window to obtain the initial dynamic baseline. Since the corresponding holidays and major promotional events (such as June or November each year) are roughly the same each year, the initial dynamic baseline basically conforms to the changes in the processing time of each node of the order each year, and can be used to determine the processing time of subsequent order nodes.
[0061] In one implementation, obtaining the first sliding time window corresponding to the historical marked date interval where the historical order data is located includes:
[0062] S10221. Obtain the historical business operation rules corresponding to the historical marked date range;
[0063] S10222. Determine the first sliding time window corresponding to the historical marked date interval according to the historical business operation rules.
[0064] In short, considering that merchants adopt different business operation rules for different holidays or different product promotion periods, the processing time of each node in the order is different (for example, some product promotion periods start pre-sales one month in advance). Therefore, the first sliding time window is set according to the business operation rules corresponding to the historical marked date interval, so that it can fully cover the processing time of the nodes in the order, which is convenient for subsequent calculation of the initial dynamic baseline.
[0065] In one implementation, the initial dynamic baseline is determined by calculating the historical order data based on the target sliding time window, including:
[0066] Step S1031: Calculate the average processing time and standard deviation of processing time for each node of the historical order data within the target sliding time window;
[0067] Step S1032: Determine the initial dynamic baseline based on the average processing time and the standard deviation of the processing time.
[0068] Specifically, historical order data is sorted by time, and a target sliding time window is used to calculate the baseline processing time for each node by sliding across the historical order data sequence. For example, a target sliding time window of one week is set. The average processing time for the "Payment → OMS Order Acceptance" node within the target sliding time window is calculated. The formula for calculating the average processing time can be... Where x i This represents the processing time of the i-th order within the time window, where n is the number of orders within the target sliding time window. Simultaneously, the standard deviation of processing times within the time window is calculated. The initial dynamic baseline (normal threshold) corresponding to the target sliding time window is then set to the average value plus or minus a certain multiple of the standard deviation. For example, the initial dynamic baseline is set to...
[0069]
[0070] In one implementation, determining the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date includes:
[0071] Step S104: Obtain the marked date range and determine whether the date of each node is within the marked date range;
[0072] Step S105: If yes, obtain the business operation rules corresponding to the marked date interval, adjust the initial dynamic baseline corresponding to the marked date interval according to the business operation rules, and determine the target dynamic baseline;
[0073] Step S106: If not, use the initial dynamic baseline corresponding to each node date as the target dynamic baseline.
[0074] Specifically, the marked date range is the date range corresponding to the year in which the target order is located. The marked date range can be obtained by the user entering it on the terminal or by determining the dates of various holidays in the calendar or the dates of major promotional periods in previous years. There is a mapping relationship between the marked date range and historical marked date ranges, such as the date of Holiday A in 2025 and Holiday A in 2025. Based on the mapping relationship between the marked date range and historical marked date ranges, a mapping relationship can be established between the initial dynamic baseline and the dates of the year in which the target order is located.
[0075] After obtaining the marked date range, it is determined whether the dates of each node of the target order fall within the marked date range. If the dates of each node of the target order fall within the marked date range, the business operation rules corresponding to the marked date range are obtained. The initial dynamic baseline corresponding to the marked date range is adjusted according to the business operation rules to determine the target dynamic baseline. Since the business operation rules corresponding to each holiday in a year differ from those corresponding to the same holidays in historical years, historical business operation rules for the marked date range can be obtained. Based on the business operation rules and historical operation rules, the dynamic baseline adjustment value is determined. The initial dynamic period corresponding to the marked date range is adjusted according to the dynamic baseline adjustment value to determine the target dynamic baseline. If the dates of each node of the target order do not fall within the marked date range, the initial dynamic baseline corresponding to each node date is used as the target dynamic baseline.
[0076] like Figure 1 As shown, the method includes:
[0077] Step S200: Determine the time deviation value based on the target dynamic baseline and the date of each node.
[0078] In simple terms, once the target dynamic baseline corresponding to the target order is determined, the processing time of each node can be determined based on the date of each node. The difference between the target dynamic baseline corresponding to each node in the target order and the processing time of each node can be calculated to obtain the time deviation value corresponding to each node.
[0079] like Figure 1 As shown, the method includes:
[0080] Step S300: Obtain the data integrity and associated node status corresponding to the target order, and determine the anomaly level corresponding to the target order based on the data integrity, the associated node status, and the time deviation value.
[0081] Specifically, in addition to processing time at each point in time, other factors that can indicate whether an order is abnormal include data integrity and the status of associated nodes. Data integrity measures whether order information is complete, while the status of associated nodes reflects whether the status of nodes linked to the order is abnormal. For example, during the payment process, monitoring the inventory status may be used to determine if it is sufficient. By using multiple factors to determine order abnormalities, the true risk situation can be accurately identified in complex business processes, avoiding resource waste caused by handling false alarms and the inability to timely and effectively identify and handle potential risks, thus affecting the stability and efficiency of overall business operations.
[0082] In one implementation, the anomaly level of the target order is determined based on the data integrity, the status of associated nodes, and the time deviation value, including:
[0083] Step S301: Obtain the weights corresponding to the data integrity, the status of the associated nodes, and the time deviation value;
[0084] Step S302: Determine the anomaly assessment score based on the data completeness, the status of the associated nodes, the time deviation value, and each of the weights;
[0085] Step S303: Obtain the abnormal score range, and determine the abnormal level based on the abnormal assessment score and the abnormal score range.
[0086] Specifically, such as Figure 3 As shown, based on the importance of each factor—data integrity, related node status, and time deviation—to order processing risk, different weights are assigned to each factor through expert experience and historical data analysis. For example, if the processing time of a certain key node has a significant impact on the overall order risk, it can be assigned a higher weight. Assume the weight of time deviation is w1, the weight of data integrity is w2, and the weight of related node status is w3, and w1 + w2 + w3 = 1.
[0087] Calculate an evaluation score for each factor in the target order. For time deviation, assign a score S1 based on the degree of deviation. For data completeness, deduct a certain score S2 if the data is incomplete. For related node status, calculate a score S3 based on the impact of a problem on the current order if a related node has a problem (e.g., insufficient inventory). Based on the scores and weights corresponding to time deviation, data completeness, and related node status, calculate the anomaly assessment score: S = w1S1 + w2S2 + w3S3, where S1, S2, and S3 are the scores for time deviation, data completeness, and related node status, respectively.
[0088] Different anomaly level ranges are set based on anomaly assessment scores. For example, when S is in the range of 0-30, it is a low-level anomaly; when S is in the range of 31-60, it is a medium-level anomaly; and when S is above 60, it is a high-level anomaly. Different anomaly levels will trigger different response mechanisms. For example, low-level anomalies may only require monitoring, while high-level anomalies require immediate intervention.
[0089] Based on the above embodiments, the present invention also provides a sales chain anomaly dynamic detection device, such as... Figure 3 As shown, the device includes:
[0090] The target dynamic baseline determination module is used to obtain the initial dynamic baseline and the node dates corresponding to the target order, and determine the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date.
[0091] The time deviation value calculation module is used to determine the time deviation value based on the target dynamic baseline and the date of each node.
[0092] The anomaly level determination module is used to obtain the data integrity and associated node status corresponding to the target order, and determine the anomaly level corresponding to the target order based on the data integrity, the associated node status, and the time deviation value.
[0093] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 4 As shown, the terminal includes a processor, memory, network interface, and display screen connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a dynamic detection method for sales link anomalies. The display screen can be an LCD screen or an e-ink screen.
[0094] Those skilled in the art will understand that Figure 4 The schematic diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0095] In one implementation, the terminal's memory stores one or more programs, and these programs are configured to be executed by one or more processors, and the programs contain instructions for performing a sales link anomaly dynamic detection method.
[0096] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.
[0097] In summary, this invention discloses a method, apparatus, terminal, and storage medium for dynamic detection of anomalies in the sales chain. The method involves obtaining an initial dynamic baseline and the dates of each node corresponding to the target order; determining the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and the dates of each node; determining the time deviation value based on the target dynamic baseline and the dates of each node; obtaining the data integrity and associated node status corresponding to the target order; and determining the anomaly level of the target order based on the data integrity, associated node status, and time deviation value. Because this invention calculates the time deviation value based on the target dynamic baseline corresponding to the node dates, and determines the anomaly level of the target order based on the time deviation value, data integrity, and associated node status, it effectively solves the problem of existing technologies that only consider a single factor indicator, resulting in low accuracy in anomaly detection and difficulty in accurately identifying true risk situations in complex business processes.
[0098] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.
Claims
1. A method for dynamic detection of anomalies in a sales chain, characterized in that, The method includes: Obtain the initial dynamic baseline and the node dates corresponding to the target order, and determine the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date; The time deviation value is determined based on the target dynamic baseline and the date of each node; Obtain the data integrity and associated node status corresponding to the target order, and determine the anomaly level corresponding to the target order based on the data integrity, the associated node status, and the time deviation value.
2. The sales link anomaly dynamic detection method according to claim 1, characterized in that, Methods for obtaining an initial dynamic baseline include: Retrieve historical order data and historical marked date ranges; The target sliding time window is determined based on the historical order data and the historical marked date range; The initial dynamic baseline is determined by calculating the historical order data based on the target sliding time window.
3. The sales link anomaly dynamic detection method according to claim 2, characterized in that, Determining the target sliding time window based on the historical order data and the historical marked date range includes: Determine whether the historical order data falls within each of the historical marked date ranges; If so, then obtain the first sliding time window corresponding to the historical marked date interval where the historical order data is located, and use the first sliding time window as the target sliding time window; If not, then the preset second sliding time window will be used as the target sliding time window.
4. The sales link anomaly dynamic detection method according to claim 3, characterized in that, Obtaining the first sliding time window corresponding to the historical marked date interval where the historical order data is located includes: Obtain the historical business operation rules corresponding to the historical marked date range; The first sliding time window corresponding to the historical marked date interval is determined according to the historical business operation rules.
5. The sales link anomaly dynamic detection method according to claim 4, characterized in that, Determining the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date includes: Obtain the marked date range and determine whether the date of each node is within the marked date range; If so, obtain the business operation rules corresponding to the marked date interval, adjust the initial dynamic baseline corresponding to the marked date interval according to the business operation rules, and determine the target dynamic baseline; If not, the initial dynamic baseline corresponding to each node date shall be used as the target dynamic baseline.
6. The sales link anomaly dynamic detection method according to claim 2, characterized in that, The initial dynamic baseline is determined by calculating the historical order data based on the target sliding time window, including: Calculate the average processing time and standard deviation of processing time for each node of the historical order data within the target sliding time window; The initial dynamic baseline is determined based on the average processing time and the standard deviation of the processing time.
7. The sales link anomaly dynamic detection method according to claim 1, characterized in that, The anomaly level of the target order is determined based on the data completeness, the status of associated nodes, and the time deviation value, including: Obtain the weights corresponding to the data integrity, the status of the associated nodes, and the time deviation value; An anomaly assessment score is determined based on the data completeness, the status of the associated nodes, the time deviation value, and each of the weights. Obtain the abnormal score range, and determine the abnormal level based on the abnormal assessment score and the abnormal score range.
8. A dynamic detection device for abnormalities in a sales chain, characterized in that, The device includes: The target dynamic baseline determination module is used to obtain the initial dynamic baseline and the node dates corresponding to the target order, and determine the target dynamic baseline corresponding to each node date based on the initial dynamic baseline and each node date. The time deviation value calculation module is used to determine the time deviation value based on the target dynamic baseline and the date of each node. The anomaly level determination module is used to obtain the data integrity and associated node status corresponding to the target order, and determine the anomaly level corresponding to the target order based on the data integrity, the associated node status, and the time deviation value.
9. A terminal, characterized in that, The terminal includes a memory and one or more processors; the memory stores one or more programs; the programs contain instructions for executing the sales link anomaly dynamic detection method as described in any one of claims 1-7; the processors are used to execute the programs.
10. A computer-readable storage medium storing a plurality of instructions thereon, characterized in that, The instructions are loaded and executed by the processor to implement the steps of the sales link anomaly dynamic detection method according to any one of claims 1-7.