Distribution abnormity monitoring method and device

Abstract

The embodiment of the invention relates to a distribution abnormity monitoring method and device, and the method comprises the steps: obtaining a preset distribution path according to a distribution starting point and an end point; dividing the predetermined distribution path into N sections according to a first division rule; if the actual delivery time of the ith section exceeds the delivery time threshold value of the ith section, correcting the actual delivery time according to a second correction rule to obtain the corrected delivery time of the ith section; and if the i-th segment of corrected delivery time exceeds the i-th segment of delivery time threshold, sending out a first abnormity prompt. According to the technical scheme provided by the embodiment of the invention, the predetermined delivery path is divided into N sections, the delivery condition of each section is monitored, the position of a deliveryman does not need to be tracked in the whole process, the delivery time of each path section is controlled, the change of the delivery state can be effectively captured on a key node, and once the delivery is abnormal, the delivery condition of the deliveryman is monitored. Based on the time data of each segment, the specific path segment where the abnormality occurs can be quickly locked.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of Internet logistics distribution planning, and particularly to a method and device for monitoring distribution anomalies. Background Art

[0002] In the process of logistics distribution, real-time monitoring of the distribution situation is crucial. Whether the distribution anomalies can be monitored in a timely manner and corresponding measures can be taken promptly not only directly determines whether the entire distribution process can be completed smoothly, but also concerns whether the emergencies encountered by the delivery staff can be properly handled.

[0003] In the prior art, the tracking of the distribution situation is usually achieved by means of GPS positioning on the delivery staff's device. This method aims to obtain the location of the delivery staff in real time so as to master the distribution progress. However, this monitoring method has obvious drawbacks. On the one hand, in order to ensure that any dynamic changes in the distribution process can be captured in a timely manner, it is necessary to monitor the location of the delivery staff almost continuously. This means that a large amount of GPS data needs to be continuously collected, transmitted, and processed, which poses great challenges to both the computing power of the data processing system and the stability of network transmission. On the other hand, the actual distribution environment is complex and changeable. Traffic congestion often occurs in cities, and situations such as road construction and bad weather leading to poor road conditions are also common. Under the influence of these accidental factors, the delivery staff often has to pause temporarily. However, the existing GPS positioning monitoring mechanism, due to the lack of the ability to comprehensively judge the distribution environment and various accidental factors, is very likely to misjudge such normal pauses caused by external environments as abnormal situations based only on the appearance that the location of the delivery staff has not changed for a long time, and then issue false alarms. Such false alarms not only disrupt the normal management order of logistics distribution, but also mislead the management to make unnecessary response decisions, further wasting human and material resources. Summary of the Invention

[0004] Based on the above situation of the prior art, the purpose of the embodiments of the present invention is to provide a method and device for monitoring distribution anomalies, which can improve the efficiency and accuracy of distribution anomaly monitoring.

[0005] To achieve the above object, according to one aspect of the present invention, a method for monitoring distribution anomalies is provided, and the method includes the steps of:

[0006] Obtain a predetermined distribution path according to the starting point and the ending point of the distribution;

[0007] Divide the predetermined distribution path into N segments according to the first division rule;

[0008] If the actual distribution time of the i-th segment exceeds the distribution time threshold of the i-th segment, correct the actual distribution time according to the second correction rule to obtain the corrected distribution time of the i-th segment;

[0009] If the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, a first anomaly reminder is issued;

[0010] Where N is a natural number greater than or equal to 2, and i ≤ N.

[0011] Furthermore, the method further includes:

[0012] If the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, and i < N, predict the delivery time of the (i + 1)-th segment according to the third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment;

[0013] If the sum of the predicted delivery time of the (i + 1)-th segment and the corrected delivery time of the i-th segment exceeds the sum of the delivery time threshold of the i-th segment and the delivery time threshold of the (i + 1)-th segment, a second anomaly reminder is issued.

[0014] Furthermore, dividing the predetermined delivery path into N segments according to the first division rule includes:

[0015] Estimate the estimated delivery time and driving difficulty coefficient of each continuous road segment according to map data, real-time road conditions, and road conditions;

[0016] Initialize a path segment, add the first continuous road segment to this path segment, and obtain the first total estimated delivery time and the first total difficulty coefficient after the first continuous road segment is added to this path segment;

[0017] Traverse the subsequent continuous road segments. For each continuous road segment, obtain the second total estimated delivery time and the second total difficulty coefficient after adding it to the current path segment;

[0018] Divide the predetermined delivery path according to the comparison results of the first total estimated delivery time and the second total estimated delivery time, and the first total difficulty coefficient and the second total difficulty coefficient.

[0019] Furthermore, dividing the predetermined delivery path according to the comparison results of the first total estimated delivery time and the second total estimated delivery time, and the first total difficulty coefficient and the second total difficulty coefficient includes:

[0020] If the difference between the first total estimated delivery time and the second total estimated delivery time is greater than the delivery time difference threshold; or,

[0021] If the difference between the first total difficulty coefficient and the second total difficulty coefficient is greater than the difficulty coefficient difference threshold;

[0022] Then divide a new path segment and add the subsequent continuous road segments to the new path segment.

[0023] Further, the actual delivery time is corrected according to the second correction rule to obtain the corrected delivery time for the i-th segment, including:

[0024] Obtain a comprehensive correction coefficient according to the congestion correction coefficient, construction correction coefficient, and weather correction coefficient;

[0025] Use the comprehensive correction coefficient and the following formula to correct the actual delivery time to obtain the corrected delivery time for the i-th segment:

[0026] T c = T a * K

[0027] where T c represents the corrected delivery time, T a represents the actual delivery time, and K represents the comprehensive correction coefficient.

[0028] Further, the delivery time for the (i + 1)-th segment is predicted according to the third prediction rule to obtain the predicted delivery time for the (i + 1)-th segment, including:

[0029] Extract the road feature coefficient, congestion feature coefficient, construction feature coefficient, and weather feature coefficient for the i-th segment;

[0030] Find a similar path segment for the i-th segment in the historical data, where the similarity of the similar path segment to the i-th segment is higher than the similarity threshold, and the similarity is determined based on the road feature coefficient, congestion feature coefficient, construction feature coefficient, and weather feature coefficient;

[0031] Obtain the predicted delivery time for the (i + 1)-th segment according to the delivery time statistics of the subsequent path segment of the similar path segment.

[0032] Further, the method further includes:

[0033] The first abnormal reminder includes application program alarm and sending a delivery abnormality reminder to the customer terminal;

[0034] The second abnormal reminder includes intervening in the delivery process of the deliveryman.

[0035] Further, the method further includes:

[0036] If the second abnormal reminder is not sent within the first predetermined time period after the first abnormal reminder, the application program alarm is turned off, and an abnormal solution notice is sent to the customer terminal.

[0037] According to the second aspect of the present invention, a delivery abnormality monitoring device is provided, and the device includes:

[0038] A predetermined delivery path acquisition module, configured to obtain a predetermined delivery path according to the starting point and the ending point of the delivery;

[0039] A path segmentation module, configured to divide the predetermined delivery path into N segments according to a first division rule;

[0040] A delivery time correction module, configured to correct the actual delivery time according to a second correction rule to obtain the corrected delivery time of the i-th segment if the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment;

[0041] A first anomaly reminder module, configured to send a first anomaly reminder if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment;

[0042] Wherein, N is a natural number greater than or equal to 2, and i ≤ N.

[0043] Further, the apparatus further includes:

[0044] A delivery time prediction module, configured to predict the delivery time of the (i + 1)-th segment according to a third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment and i < N;

[0045] A second anomaly reminder module, configured to send a second anomaly reminder if the sum of the predicted delivery time of the (i + 1)-th segment and the corrected delivery time of the i-th segment exceeds the sum of the delivery time threshold of the i-th segment and the delivery time threshold of the (i + 1)-th segment.

[0046] In summary, the embodiments of the present invention provide a delivery anomaly monitoring method and apparatus. The method includes obtaining a predetermined delivery path according to the starting point and the ending point of the delivery; dividing the predetermined delivery path into N segments according to a first division rule; correcting the actual delivery time according to a second correction rule to obtain the corrected delivery time of the i-th segment if the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment; and sending a first anomaly reminder if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment. The technical solution provided by the embodiments of the present invention monitors the delivery situation of each segment separately by dividing the predetermined delivery path into N segments, without the need to track the position of the delivery person throughout the process. By controlling the delivery time of each path segment, it can effectively capture the change of the delivery status at key nodes, and once an anomaly occurs in the delivery, based on the time data of each segment, it can quickly lock the specific path segment where the anomaly occurs. Moreover, by using a correction rule to correct the actual delivery time, various real-time road conditions are fully considered, such as short-term impacts of traffic congestion, road construction, or weather on the delivery time, which can effectively avoid false alarms. Description of the Drawings

[0047] Figure 1 is a flowchart of the delivery anomaly monitoring method provided by the embodiments of the present invention. Detailed Embodiments

[0048] To make the objectives, technical solutions and advantages of the present invention clearer and more apparent, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily confusing the concepts of the present invention.

[0049] It should be noted that unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should have the ordinary meaning as understood by those of ordinary skill in the field to which the present invention pertains. The terms "first", "second" and similar terms used in one or more embodiments of the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0050] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings. An embodiment of the present invention provides a method for monitoring distribution anomalies. Figure 1 The flowchart of the method for monitoring distribution anomalies according to the embodiment of the present invention is shown in Figure 1 As shown, the method includes the following steps:

[0051] S202. Obtain a predetermined distribution path according to the starting point and the ending point of the distribution. For each distribution task, according to the starting point and the ending point of the distribution task, obtain the corresponding map information, and adopt a specific path planning algorithm according to the map information to obtain the predetermined distribution path of the distribution task. The path planning algorithm can, for example, adopt the Dijkstra algorithm or the A* algorithm, and the obtained predetermined distribution path is the driving path for the distributor to perform the distribution.

[0052] S204. Divide the predetermined distribution path into N segments according to the first division rule. In the embodiment of the present invention, the predetermined distribution path is divided into N segments, and the distribution conditions of each segment are monitored separately. It is not necessary to track the position of the distributor throughout the process. By controlling the distribution time of each path segment, the change of the distribution state can be effectively captured at key nodes, and once an anomaly occurs in the distribution, based on the time data of each segment, the specific path segment where the anomaly occurs can be quickly locked. Specifically, it can include the following steps:

[0053] S2041. Estimate the estimated delivery time and driving difficulty coefficient of each continuous road section based on map data, real-time traffic conditions and road conditions. The length, speed limit and other information of the road can be obtained from the map data, and the estimated driving time of each continuous road section can be estimated in combination with the real-time traffic information. The driving difficulty coefficient of each continuous road section can also be evaluated based on factors such as the slope of the road, the number of curves, and the traffic volume. In the embodiments of the present application, a continuous road refers to a section of road that is geographically spatially interconnected on the delivery path, the driving direction remains roughly unchanged, the traffic rules (lane rules, traffic signal control, etc.) and the road conditions (road type, road condition characteristics, etc.) remain relatively consistent, and the means of transportation used for delivery can directly drive through without obvious interruption or the need to significantly change the driving state.

[0054] S2042, initialize a path segment, add the first continuous road segment to the path segment, and obtain the first estimated delivery time sum and the first difficulty coefficient sum after the first continuous road segment is added to the path segment. For example, if the initialized path segment is 0, the estimated delivery time and driving difficulty coefficient of the first continuous road segment are calculated, which is the first estimated delivery time sum T of the current path segment. sum The sum of the first difficulty coefficient D sum .

[0055] S2043, traverse the subsequent continuous roads, and for each continuous road segment, obtain the sum of the second estimated delivery time and the sum of the second difficulty coefficient after adding to the current path segment. Continuing with the example above, add the second continuous road segment following the first continuous road segment to the first continuous road segment to obtain "the first continuous road segment + the second continuous road segment" (i.e., the current path segment), and calculate the sum of the estimated delivery time and the sum of the difficulty coefficient of "the first continuous road segment + the second continuous road segment", which is the second estimated delivery time sum T newSum and the sum of the second difficulty coefficient D newSum .

[0056] S2044: Divide the scheduled delivery routes according to the comparison result between the sum of the first estimated delivery time and the sum of the second estimated delivery time, and the sum of the first difficulty coefficient and the sum of the second difficulty coefficient:

[0057] If the difference between the sum of the first estimated delivery time and the sum of the second estimated delivery time is greater than the delivery time difference threshold; or,

[0058] If the difference between the sum of the first difficulty coefficient and the sum of the second difficulty coefficient is greater than the difficulty coefficient difference threshold;

[0059] A new path segment is then divided, and the subsequent continuous roads are added to the new path segment.

[0060] For example, the sum of the first estimated delivery time and the sum of the second estimated delivery time satisfy the formula:

[0061] ∣T newSum -T sum ∣>ΔT

[0062] Alternatively, the sum of the first difficulty coefficients and the sum of the second difficulty coefficients satisfy the formula:

[0063] ∣D newSum -D sum ∣>ΔD

[0064] Wherein, ΔT represents the threshold value of the delivery time difference, and ΔD represents the threshold value of the difficulty coefficient difference. In this embodiment of the present invention, for the method of segmenting the predetermined delivery path, factors such as road length, road speed limit, and actual road conditions are comprehensively considered to estimate the expected driving time and driving difficulty of each segment of the path. Segments with similar expected driving time and difficulty are divided into the same segment, which can make the distribution task volume of each path segment relatively balanced, facilitate the reasonable arrangement of delivery time and resources, and also facilitate real-time positioning in case of abnormal situations.

[0065] S206. If the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, correct the actual delivery time according to the second correction rule to obtain the corrected delivery time of the i-th segment. This step fully considers various real-time road condition factors, such as the short-term impact of traffic congestion, road construction, or weather on the delivery time, and avoids false alarm situations. The correction can be specifically carried out according to the following steps:

[0066] S2061. Obtain the comprehensive correction coefficient according to the congestion correction coefficient, the construction correction coefficient, and the weather correction coefficient. The congestion correction coefficient can be calculated according to the real-time congestion index. For example, when the congestion index is C (0 ≤ C ≤ 1), when C = 0, it means the road is unobstructed, and the congestion correction coefficient K C = 1; when C > 0, the congestion correction coefficient K is calculated through the following formula C :

[0067] K C = 1 / (1 + αC)

[0068] Wherein, α represents the coefficient obtained by fitting historical data and is used to adjust the impact degree of congestion on the delivery time. The α values of different types of roads are different. For example, α = 2 for urban arterial roads and α = 3 for secondary arterial roads, etc.

[0069] If there is road construction, the construction correction coefficient can be used to correct the actual delivery time, and the construction correction coefficient can be calculated according to the construction road length, the construction type, and the construction road type:

[0070]

[0071] Wherein, K tRepresents the construction correction coefficient, L total Represents the total length of the i-th path segment, L represents the length of the construction road, β represents the basic influence coefficient of the construction type. Different construction types have different impacts on traffic. For example, pavement maintenance construction has a relatively small impact, while bridge construction has a relatively large impact, so the value of β is different; γ represents the road type adjustment coefficient. Different road types have different tolerances for construction. For example, highway construction has a relatively large impact on traffic, and the value of γ may be 0.9; small road construction has a relatively small impact, and the value of γ may be 0.6.

[0072] Weather correction coefficient K w Can be determined according to different weather conditions and road types. For example, when it is sunny, the weather correction coefficient is 1; in heavy rain weather, for urban arterial roads, the weather correction coefficient is 0.8, and for mountain roads, the weather correction coefficient is 0.6, etc. The weather correction coefficient can be determined based on historical data statistics and experience, reflecting the impact of different weather on the delivery time of different roads.

[0073] Taking into account the above congestion correction coefficient, construction correction coefficient and weather correction coefficient, the comprehensive correction coefficient K can be calculated by the geometric mean method:

[0074]

[0075] S2062. Use the comprehensive correction coefficient and the following formula to correct the actual delivery time to obtain the corrected delivery time of the i-th segment:

[0076] T c =T a *K

[0077] Where, T c Represents the corrected delivery time, T a Represents the actual delivery time, and K represents the comprehensive correction coefficient.

[0078] S208. If the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, send a first abnormal reminder. In the above steps, N is a natural number greater than or equal to 2, and i ≤ N. After correcting the corrected delivery time of the i-th segment, if the corrected delivery time of the i-th segment still exceeds the delivery time threshold of the i-th segment, it means that an unexpected situation that cannot be recovered in a short time may occur during the delivery process, so a first abnormal reminder is sent. The first abnormal reminder can be given in the form of an application program alarm and a delivery exception reminder to the customer terminal to remind the control center and the user that an abnormality may occur during the delivery process. At this time, the control center should closely monitor the subsequent process of this delivery to determine whether a more serious abnormal situation occurs.

[0079] According to some optional embodiments, the method may further include the following steps:

[0080] S210. If the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment and i < N, predict the delivery time of the (i + 1)-th segment according to the third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment. If the current path segment is not the last path segment of the predetermined delivery path, find a similar delivery scenario from the historical data according to the conditions of the current path segment and predict the delivery time of the next path segment of the current path segment. The prediction can be carried out according to the following steps:

[0081] S2101. Extract the road characteristic coefficient, congestion characteristic coefficient, construction characteristic coefficient, and weather characteristic coefficient of the i-th segment. The road characteristic coefficient can represent the road type, and the road type can be divided into, for example, expressways, urban arterial roads, secondary arterial roads, branch roads, rural roads, etc. The one-hot encoding method can be used to convert it into a vector representation. For example, assuming there are 5 road types, the expressway is represented as [1, 0, 0, 0, 0], the urban arterial road is represented as [0, 1, 0, 0, 0], and so on. The congestion characteristic coefficient is used to represent the congestion situation of the road, and the congestion characteristic coefficient is usually between 0 and 1, where 0 means smooth and 1 means severe congestion. For example, if the congestion index of the current i-th segment of the road is 0.6, it means there is a certain degree of congestion. The construction characteristic coefficient can be calculated using the following formula:

[0082]

[0083] The meanings of the parameters in this formula are the same as those in the formula for calculating the construction correction coefficient in the above text. The weather characteristic coefficient can represent the weather type, and the weather type includes sunny, cloudy, light rain, moderate rain, heavy rain, rainstorm, snowy, foggy, etc. The weather characteristic coefficient can also be represented in a similar one-hot encoding manner. For example, sunny is represented as [1, 0, 0, 0, 0, 0, 0, 0], light rain is represented as [0, 1, 0, 0, 0, 0, 0, 0], etc.

[0084] S2102. Search for a similar path segment of the i-th segment in the historical data. The similarity of this similar path segment to the i-th segment is higher than the similarity threshold, and the similarity is determined based on the above-mentioned road feature coefficient, congestion feature coefficient, construction feature coefficient, and weather feature coefficient. In the historical data, for each path segment, relevant data can be expressed in advance as needed. For example, it can be marked whether a delivery anomaly has occurred (i.e., the actual delivery time exceeds the delivery time threshold). For the path segments where delivery anomalies have occurred, record the delivery time conditions of the subsequent path segments at the same time, especially those cases where the delivery time of the subsequent path segments of the abnormal path segments can make up for the time loss caused by the anomaly (i.e., the sum of the corrected delivery time of the abnormal path segment and the actual delivery time of the subsequent path segments does not exceed the overall delivery time threshold) and those that cannot make up for it. In this step, search for path segments with similar features in the historical data, and determine the matching degree by calculating the similarity between the features. For example, the similarity between the features can be calculated using existing similarity calculation methods according to the characteristics of the feature coefficients, such as the Jaccard similarity, Euclidean distance, etc. Screen out the data of the path segments and their subsequent path segments in the historical data whose similarity to the i-th segment exceeds the similarity threshold (e.g., 0.8). These screened historical path segments and their subsequent path segments form a scenario set similar to the i-th segment scenario.

[0085] S2103. Obtain the predicted delivery time of the (i + 1)-th segment according to the delivery time statistics of the subsequent path segments of this similar path segment. In the screened similar scenario set, count the delivery time distribution of the subsequent path segments (i.e., corresponding to the (i + 1)-th segment). Statistical quantities such as the mean and median of the delivery times of these subsequent path segments can be calculated and used as the estimated value of the predicted delivery time of the (i + 1)-th segment. For example, if there are 50 matching subsequent path segments in the similar scenario set and the mean of their delivery times is 35 minutes, then 35 minutes is used as the predicted delivery time of the (i + 1)-th segment.

[0086] The above prediction steps provided by this embodiment of the present invention can accurately match similar scenarios from historical data when encountering the current delivery scenario. Compared with simply relying on a theoretical model, it can be more in line with the actual delivery situation of similar road segments and provide a more reliable basis for judging the compensation situation of time loss.

[0087] S212. If the sum of the predicted delivery time of the (i + 1)-th segment and the corrected delivery time of the i-th segment exceeds the sum of the delivery time threshold of the i-th segment and the delivery time threshold of the (i + 1)-th segment, send a second anomaly reminder. The second anomaly reminder notifies the control center to intervene in the delivery process of the delivery person. In this case, it means that according to the prediction result, the delivery requirements still cannot be met in the subsequent road segments, and a relatively serious anomaly may occur, and the control center needs to access the delivery process.

[0088] According to some optional embodiments, the method further includes the step:

[0089] S214. If no second anomaly alert is issued within the first predetermined time period after the first anomaly alert, close the application alarm and send an anomaly resolution notice to the customer terminal. If no second anomaly alert is issued within the first predetermined time period after the first anomaly alert, it indicates that the anomaly situation is predicted to be resolved within the predetermined time or has been resolved during this period. Then, close the application alarm and send an anomaly resolution notice to the customer terminal.

[0090] An embodiment of the present invention further provides a distribution anomaly monitoring device, which includes:

[0091] A predetermined delivery path acquisition module for acquiring a predetermined delivery path according to the starting point and the ending point of the delivery;

[0092] A path segmentation module for segmenting the predetermined delivery path into N segments according to the first segmentation rule;

[0093] A delivery time correction module for, if the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, correcting the actual delivery time according to the second correction rule to obtain the corrected delivery time of the i-th segment;

[0094] A first anomaly alert module for, if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, issuing a first anomaly alert;

[0095] Wherein, N is a natural number greater than or equal to 2, and i ≤ N.

[0096] According to some optional embodiments, the device further includes:

[0097] A delivery time prediction module for, if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment and i < N, predicting the delivery time of the (i + 1)-th segment according to the third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment;

[0098] A second anomaly alert module for, if the sum of the predicted delivery time of the (i + 1)-th segment and the corrected delivery time of the i-th segment exceeds the sum of the delivery time threshold of the i-th segment and the delivery time threshold of the (i + 1)-th segment, issuing a second anomaly alert.

[0099] The specific implementation of the functions of each module in the distribution anomaly monitoring device in the embodiment of the present invention is the same as each step in the distribution anomaly monitoring method provided in the above embodiment of the present invention, and the repeated description thereof will be omitted here.

[0100] In summary, the embodiments of the present invention relate to a method and device for monitoring abnormal delivery. The method includes obtaining a predetermined delivery route according to the starting point and the ending point of the delivery; dividing the predetermined delivery route into N segments according to a first division rule; if the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, correcting the actual delivery time according to a second correction rule to obtain the corrected delivery time of the i-th segment; if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, sending a first abnormal reminder. The technical solution provided by the embodiments of the present invention monitors the delivery situation of each segment separately by dividing the predetermined delivery route into N segments, without the need to track the location of the delivery person throughout the process. By controlling the delivery time of each route segment, it can effectively capture changes in the delivery status at key nodes, and once an abnormal delivery occurs, based on the time data of each segment, the specific route segment where the abnormality occurs can be quickly locked. Moreover, the actual delivery time is corrected using a correction rule, fully considering various real-time road conditions, such as the short-term impact of traffic congestion, road construction, or weather on the delivery time, which can effectively avoid false alarms.

[0101] It should be understood that the discussion of any above embodiment is only exemplary and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples; under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of one or more embodiments of the present invention as described above, which are not provided in detail for the sake of brevity. The above specific embodiments of the present invention are only used for exemplary illustration or explanation of the principle of the present invention and do not constitute a limitation to the present invention. Therefore, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention shall be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims or the equivalent forms of such scope and boundaries.

Claims

1. A method for monitoring abnormal distribution, characterized in that: The method includes the steps of: Obtaining a predetermined delivery route according to the starting point and the ending point of the delivery; Dividing the predetermined delivery route into N segments according to a first division rule; If the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, correcting the actual delivery time according to a second correction rule to obtain the corrected delivery time of the i-th segment; If the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, sending a first anomaly reminder; where N is a natural number greater than or equal to 2, and i ≤ N.

2. The method according to claim 1, characterized in that: The method further includes: If the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment and i < N, predicting the delivery time of the (i + 1)-th segment according to a third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment; If the sum of the predicted delivery time of the (i + 1)-th segment and the corrected delivery time of the i-th segment exceeds the sum of the delivery time threshold of the i-th segment and the delivery time threshold of the (i + 1)-th segment, sending a second anomaly reminder.

3. The method according to claim 2, characterized in that Dividing the predetermined delivery route into N segments according to the first division rule includes: Estimating the expected delivery time and the driving difficulty coefficient of each continuous road segment according to map data, real-time road conditions, and road conditions; Initializing a route segment, adding the first continuous road segment to this route segment, and obtaining the first total expected delivery time and the first total difficulty coefficient after the first continuous road segment is added to this route segment; Traversing the subsequent continuous road segments. For each continuous road segment, obtaining the second total expected delivery time and the second total difficulty coefficient after adding it to the current route segment; Dividing the predetermined delivery route according to the comparison results of the first total expected delivery time and the second total expected delivery time, and the first total difficulty coefficient and the second total difficulty coefficient.

4. The method according to claim 3, characterized in that Dividing the predetermined delivery route according to the comparison results of the first total expected delivery time and the second total expected delivery time, and the first total difficulty coefficient and the second total difficulty coefficient includes: If the difference between the first total expected delivery time and the second total expected delivery time is greater than the delivery time difference threshold; or, If the difference between the first total difficulty coefficient and the second total difficulty coefficient is greater than the difficulty coefficient difference threshold; Then dividing a new route segment and adding the subsequent continuous road segments to the new route segment.

5. The method according to claim 2, characterized in that: Correcting the actual delivery time according to the second correction rule to obtain the corrected delivery time of the i-th segment includes: Obtaining a comprehensive correction coefficient according to the congestion correction coefficient, the construction correction coefficient, and the weather correction coefficient; Using the comprehensive correction coefficient and the following formula to correct the actual delivery time to obtain the corrected delivery time of the i-th segment: T c =T a *K Among them, T c Indicates the corrected delivery time, T a represents the actual delivery time, and K represents the comprehensive correction coefficient.

6. The method according to claim 2, characterized in that Predicting the delivery time of the (i + 1)-th segment according to the third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment includes: Extracting the road feature coefficient, the congestion feature coefficient, the construction feature coefficient, and the weather feature coefficient of the i-th segment; Searching for a similar route segment of the i-th segment in the historical data, where the similarity of the similar route segment to the i-th segment is higher than the similarity threshold, and the similarity is determined based on the road feature coefficient, the congestion feature coefficient, the construction feature coefficient, and the weather feature coefficient; Based on the statistical distribution time of the subsequent path segment of the similar path segment, the predicted distribution time of the (i + 1)-th segment is obtained.

7. The method according to any one of claims 2 to 6, characterized in that: The method further includes: The first abnormal reminder includes application program alarm and sending a distribution abnormality reminder to the customer terminal; The second abnormal reminder includes intervening in the distribution process of the deliveryman.

8. The method according to claim 7, characterized in that The method further includes: If the second abnormal reminder is not sent within the first predetermined time period after the first abnormal reminder, the application program alarm is turned off, and an abnormality resolution notice is sent to the customer terminal.

9. A distribution abnormality monitoring device, characterized in that: The device includes: A predetermined delivery path acquisition module, configured to acquire a predetermined delivery path according to the starting point and the ending point of the delivery; A path segmentation module, configured to divide the predetermined delivery path into N segments according to a first segmentation rule; A delivery time correction module, configured to, if the actual delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment, correct the actual delivery time according to a second correction rule to obtain the corrected delivery time of the i-th segment; A first abnormal reminder module, configured to send a first abnormal reminder if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment; where N is a natural number greater than or equal to 2, and i ≤ N.

10. The device according to claim 9, characterized in that The device further includes: A delivery time prediction module, configured to, if the corrected delivery time of the i-th segment exceeds the delivery time threshold of the i-th segment and i < N, predict the delivery time of the (i + 1)-th segment according to a third prediction rule to obtain the predicted delivery time of the (i + 1)-th segment; A second abnormal reminder module, configured to send a second abnormal reminder if the sum of the predicted delivery time of the (i + 1)-th segment and the corrected delivery time of the i-th segment exceeds the sum of the delivery time threshold of the i-th segment and the delivery time threshold of the (i + 1)-th segment.

Images