Logistics order collection scheduling method and device, equipment and storage medium

Abstract

The invention relates to the technical field of logistics, in particular to a logistics order collection scheduling method and device, equipment and a storage medium. The logistics order collection scheduling method comprises the following steps: obtaining real-time collection data, and carrying out the preprocessing of the real-time collection data; performing correlation analysis on the preprocessed real-time collected data by using a grey correlation degree method to obtain correlated data; inputting the associated data into a prediction model to obtain a collection prediction result; and scheduling salesman collection according to the collection prediction result to obtain a scheduling plan. The method can continuously adjust and optimize the scheduling plan of the collection of the salesman according to the real-time collection data, guarantees the timeliness and flexibility of the scheduling, is higher in collection route efficiency, and can improve the logistics collection time and customer satisfaction.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and in particular to a collection and scheduling method, device, equipment and storage medium for logistics orders. Background Art

[0002] With the booming development of the e-commerce industry, online shopping has become an indispensable part of people's daily lives. This trend has greatly promoted the rapid development of the logistics industry. Against this background, the number of logistics orders has shown a blowout growth trend, which not only tests the operational capabilities and service quality of logistics companies, but also puts higher requirements on each link in the order processing process. Especially in the key link of order collection, its efficiency and accuracy are directly related to the smoothness of the subsequent logistics process and the level of customer satisfaction.

[0003] Although the existing collection data statistics methods support the basic operation of logistics business to a certain extent, their limitations are becoming increasingly prominent in the face of such a huge order volume. On the one hand, incomplete data collection is a prominent problem. Traditional methods often rely on manual entry or simple automated scanning, which makes it difficult to achieve real-time and comprehensive acquisition of all order information, especially during peak periods, when missed orders and wrong orders often occur. On the other hand, processing speed is also an important factor restricting logistics efficiency. With the surge in order volume, the load on the data processing system increases. If the system performance is insufficient or the algorithm is inefficient, it will cause data processing delays, affecting the rapid response and timely processing of orders.

[0004] It can be seen that the existing technology still needs to be improved and enhanced. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide a method, device, equipment and storage medium for collecting and scheduling logistics orders, aiming to optimize the collection and scheduling of logistics orders, thereby improving the response speed of order collection.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The first aspect of the present invention provides a collection and scheduling method for logistics orders, comprising the following steps: obtaining real-time collection data and preprocessing the real-time collection data; performing correlation analysis on the preprocessed real-time collection data using a grey correlation method to obtain correlated data; inputting the correlated data into a prediction model to obtain a collection prediction result; and scheduling the collection of salesmen according to the collection prediction result to obtain a scheduling plan.

[0008] Optionally, in a first implementation method of the first aspect of the present invention, the real-time collection data is obtained and pre-processed, specifically including: obtaining real-time collection data, the real-time collection data including collection weight, collection address, collection quantity, and collection type, removing duplicate data from the real-time collection data to obtain first pre-processed data; filling missing values ​​of the first pre-processed data to obtain second pre-processed data; formatting the second pre-processed data to obtain formatted data, and storing the formatted data in a database or a cloud storage platform.

[0009] Optionally, in a second implementation of the first aspect of the present invention, the grey correlation method is used to perform correlation analysis on the preprocessed real-time collection data to obtain correlation data, which specifically includes: building a real-time alarm system, detecting the preprocessed real-time collection data through the real-time alarm system to exclude abnormal data; obtaining the detected real-time collection data, specifying the prediction indicator series and the influencing factor series in the real-time collection data; calculating the correlation coefficient of the prediction indicator series and the influencing factor series at each moment, and calculating the correlation based on multiple correlation coefficients.

[0010] Optionally, in a third implementation method of the first aspect of the present invention, the associated data is input into a prediction model to obtain a collection prediction result, specifically comprising: obtaining historical collection data, calculating the preprocessed historical collection data using a grey correlation method, and obtaining historical correlation data; using a BP neural network model as an initial model, and training the BP neural network model using the historical correlation data to obtain the prediction model; inputting the associated data into the prediction model to obtain a collection prediction result.

[0011] Optionally, in a fourth implementation method of the first aspect of the present invention, the BP neural network model is used as the initial model, and the historical correlation data is used to train the BP neural network model to obtain the prediction model, specifically including: setting the parameters of the BP neural network model, setting the number of input layer nodes to 10, the number of output layer nodes to 1, and the number of hidden layer nodes to 8, to obtain the initial model; dividing the historical correlation data in proportion to obtain a training set and a validation set; using ReLu as an activation function, and using the training set and the validation set to train the initial model to obtain the prediction model.

[0012] Optionally, in a fifth implementation of the first aspect of the present invention, the scheduling of the collection of the salesmen according to the collection prediction results to obtain a scheduling plan specifically includes: obtaining the collection prediction results, and preliminarily scheduling the collection of the salesmen according to the collection prediction results to obtain scheduling feedback data; performing abnormality inspection and preprocessing on the scheduling feedback data to obtain inspection pass data; and re-scheduling the collection of the salesmen according to the inspection pass data to obtain an optimized scheduling plan.

[0013] Optionally, in a sixth implementation of the first aspect of the present invention, the acquisition of collection prediction results and preliminary scheduling of salesmen for collection according to the collection prediction results to obtain scheduling feedback data specifically include: obtaining resource allocation conditions, the resource allocation conditions including network distribution conditions and the number of salesmen, pre-processing the resource allocation conditions to obtain resource allocation data; using a simulated annealing algorithm to calculate a collection route according to the collection prediction results and resource allocation data; and outputting the collection route to the execution end to obtain scheduling feedback data.

[0014] The second aspect of the present invention provides a collection and scheduling device for logistics orders, including: a preprocessing module, used to obtain real-time collection data and preprocess the real-time collection data; an association module, used to perform association analysis on the preprocessed real-time collection data using a grey correlation method to obtain associated data; a prediction module, used to input the associated data into a prediction model to obtain a collection prediction result; and a scheduling module, used to schedule the collection of salesmen according to the collection prediction result to obtain a scheduling plan.

[0015] Optionally, in a first implementation method of the second aspect of the present invention, the preprocessing module includes: a removal unit, used to obtain real-time collection data, the real-time collection data including collection weight, collection address, collection quantity, and collection type, and remove duplicate data from the real-time collection data to obtain first preprocessed data; a filling unit, used to fill in missing values ​​of the first preprocessed data to obtain second preprocessed data; and a formatting unit, used to format the second preprocessed data to obtain formatted data, and store the formatted data in a database or a cloud storage platform.

[0016] Optionally, in a second implementation of the second aspect of the present invention, the association module includes: a monitoring unit, used to construct a real-time alarm system, and to detect the pre-processed real-time collection data through the real-time alarm system to eliminate abnormal data; an acquisition unit, used to acquire the detected real-time collection data, and specify the prediction indicator series and the influencing factor series in the real-time collection data; a calculation unit, used to calculate the correlation coefficient of the prediction indicator series and the influencing factor series at each moment, and calculate the correlation degree based on multiple correlation coefficients.

[0017] Optionally, in a third implementation of the second aspect of the present invention, the prediction module includes: a calculation submodule, used to obtain historical collection data, and use a grey correlation method to calculate the preprocessed historical collection data to obtain historical correlation data; a construction submodule, used to use a BP neural network model as an initial model, and use the historical correlation data to train the BP neural network model to obtain the prediction model; a prediction submodule, used to input the correlated data into the prediction model to obtain a collection prediction result.

[0018] Optionally, in a fourth implementation method of the second aspect of the present invention, the prediction submodule includes: a setting unit, used to set the parameters of the BP neural network model, setting the number of input layer nodes to 10, the number of output layer nodes to 1, and the number of hidden layer nodes to 8, to obtain an initial model; a partitioning unit, used to divide the historical correlation data in proportion to obtain a training set and a validation set; a training unit, used to use ReLu as an activation function, and use the training set and the validation set to train the initial model to obtain the prediction model.

[0019] Optionally, in a fifth implementation of the second aspect of the present invention, the scheduling module includes: a scheduling sub-module, used to obtain a collection prediction result, and preliminarily schedule the collection of the salesman according to the collection prediction result to obtain scheduling feedback data; a verification sub-module, used to perform abnormality inspection and preprocessing on the scheduling feedback data to obtain inspection pass data; and an optimization sub-module, used to re-schedule the collection of the salesman according to the inspection pass data to obtain an optimized scheduling plan.

[0020] Optionally, in a sixth implementation of the second aspect of the present invention, the scheduling submodule includes: a preprocessing unit, used to obtain resource allocation information, the resource allocation information includes the distribution of outlets and the number of salesmen, and preprocess the resource allocation information to obtain resource allocation data; a planning unit, used to use a simulated annealing algorithm to calculate a collection route according to the collection prediction results and resource allocation data; an output unit, used to output the collection route to the execution end to obtain scheduling feedback data.

[0021] A third aspect of the present invention provides a collection and scheduling device for logistics orders, comprising a memory and at least one processor, wherein the memory stores computer-readable instructions; the at least one processor calls the computer-readable instructions in the memory to execute each step of the collection and scheduling method for logistics orders as described above.

[0022] A fourth aspect of the present invention provides a computer-readable storage medium having computer-readable instructions stored thereon. When the computer-readable instructions are executed by a processor, the various steps of the method for collecting and scheduling logistics orders as described above are implemented.

[0023] Beneficial effect: The present invention provides a collection and scheduling method for logistics orders, which first obtains real-time collection data in the logistics system and preprocesses the real-time collection data; then uses the grey correlation method to perform correlation analysis on the preprocessed real-time collection data, and then inputs the analyzed data into a pre-established and trained prediction model for prediction to obtain the collection data prediction result; finally, the salesperson is scheduled to collect the collection according to the collection data prediction result, which can significantly improve the data statistical accuracy and processing efficiency in the logistics order collection process, and can quickly schedule the collection in real time, thereby improving the overall logistics service level. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A first flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0025] Figure 2 A second flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0026] Figure 3 A third flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0027] Figure 4 A fourth flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0028] Figure 5 A fifth flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0029] Figure 6 A sixth flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0030] Figure 7 A seventh flow chart of the method for collecting and scheduling logistics orders provided in an embodiment of the present invention;

[0031] Figure 8 A schematic diagram of the structure of a collection and dispatching device for logistics orders provided in an embodiment of the present invention;

[0032] Fig. 9 Another structural schematic diagram of the collection and scheduling device for logistics orders provided by an embodiment of the present invention;

[0033] Fig.10A schematic diagram of the structure of a collection and scheduling device for logistics orders provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The present invention provides a collection scheduling method, device, equipment and storage medium for logistics orders. The present invention first obtains real-time collection data, pre-processes the real-time collection data, obtains data that is easy to identify and use, and then uses the gray correlation method to perform correlation analysis on the data to obtain correlation data, and then inputs the above data into the prediction model to obtain reliable collection prediction results, and finally schedules the collection of salesmen according to the collection prediction results, and forms a scheduling plan. The present invention pre-processes and performs correlation analysis on real-time collection data, effectively improving the reliability and accuracy of the data, and then uses the model to predict, further improving the accuracy of the collection prediction results. Scheduling the collection of salesmen based on the collection prediction results can significantly improve the processing efficiency of the logistics order collection process, thereby improving the overall logistics service level.

[0035] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 The first embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0037] S101. Obtain real-time collection data and pre-process the real-time collection data;

[0038] Specifically, when obtaining real-time collection data, IoT devices and sensors can be deployed at logistics sites. IoT devices have autonomous learning and decision-making capabilities, can automatically process data and take corresponding actions according to preset rules and algorithms, and can automatically collect, exchange and process data.

[0039] In order to make the data format more uniform, reduce the preprocessing steps and save system memory, the data collection system can be designed in advance when acquiring data to ensure real-time transmission of data and that the transmitted data can be edited in a specific way.

[0040] The quality of real-time collection data can be improved by preprocessing it. For example, real-time collection data may contain erroneous, duplicate or invalid information. Preprocessing can identify and remove such information to ensure that the data used in subsequent analysis is accurate and reliable. Real-time collection data from different sources may have different formats and units. Preprocessing can unify the formats and units of these data to facilitate subsequent data integration and analysis.

[0041] During execution, data preprocessing can be performed by developing data cleaning scripts to target different types of data. Preprocessed data can improve the analysis efficiency of the system. Preprocessing can reduce the size of the data set through data compression, dimensionality reduction and other technologies, thereby reducing the amount of calculation for subsequent data analysis and improving analysis efficiency. In addition, it can also enable data to enter the analysis stage faster to meet real-time requirements.

[0042] S102. Use the grey correlation method to perform correlation analysis on the pre-processed real-time collection data to obtain the correlated data;

[0043] Grey correlation analysis has relatively loose requirements on data and is suitable for situations where sample data is small or data quality is poor. In the scenario of real-time data collection, due to the large amount of data and high real-time requirements, data quality may be affected by many factors. The grey correlation method can better handle these problems, making the analysis results more reliable.

[0044] In addition, grey correlation analysis is very suitable for dynamic process analysis. During the development of the system, the correlation between various factors may change over time. Through grey correlation analysis, these changes can be tracked in real time, providing data support for dynamic adjustment of logistics strategies.

[0045] S103. Inputting the associated data into the prediction model to obtain a collection prediction result;

[0046] By inputting the data after correlation analysis, the prediction model can comprehensively consider the impact of multiple factors on the collection results, including human resources (such as the number of sorting personnel and the number of collection salesmen), material resources (such as the number of collection equipment, the number of transportation tools, and the capacity of station warehouses), etc. This comprehensive consideration makes the prediction results more accurate and can better reflect the actual situation. In addition, the use of correlated data helps to identify and eliminate noise and interference factors in the data, thereby reducing errors in the prediction process.

[0047] S104. Schedule the salespersons to collect goods according to the collection forecast results to obtain a scheduling plan.

[0048] Through the collection prediction results, we can accurately understand the future collection needs in various regions and time periods. The dispatching system can reasonably assign collection tasks to salesmen based on these prediction results to avoid uneven task distribution or waste of resources. In the dispatching plan, the optimal collection route can be planned for the salesmen based on factors such as the geographical location of the collection point, the collection time, and the transportation capacity. This can not only reduce the time and cost of the salesmen on the road, but also improve the efficiency of collection. According to the collection prediction results, the salesmen's collection can be dispatched to form a reasonable dispatching plan, which can ensure that the salesmen arrive at the collection point within the time expected by the customer and reduce the customer's waiting time. This helps to improve customer satisfaction and loyalty.

[0049] See also Figure 2 The second embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0050] S201 obtains real-time collection data, the real-time collection data includes collection weight, collection address, collection quantity, collection type, removes duplicate data of the real-time collection data to obtain the first pre-processed data;

[0051] Duplicate data can distort analysis results because they erroneously increase the weight of certain data points. After removing duplicate data, the basis for data analysis is more accurate, thereby improving the reliability of analysis results. In addition, duplicate data takes up unnecessary storage space in the database, increasing the cost of data storage and maintenance. Removing these redundant data can optimize database performance and improve data storage efficiency. In the subsequent data analysis process, duplicate data will increase the amount of calculation and reduce the speed of analysis. After removing duplicate data, the analysis process is more concise and efficient, and results can be obtained faster.

[0052] S202. Filling missing values ​​of the first preprocessed data to obtain second preprocessed data;

[0053] Missing values ​​often bias the results of data analysis because they may cause the weights of certain features or samples in the model to be incorrectly amplified or reduced. By filling missing values, this bias can be reduced and the reliability of the analysis results can be improved. In subsequent machine learning or statistical modeling, filling missing values ​​in the early stage can enable the model to make full use of all available data during the training process, thereby improving the model's prediction accuracy and generalization ability.

[0054] S203. Format the second pre-processed data to obtain formatted data, and store the formatted data in a database or a cloud storage platform.

[0055] Formatting can ensure the consistency of data formats, types, and units, reduce differences and conflicts between data, and improve the overall quality of data. In addition, formatted data is usually more compact and efficient, which can reduce the demand for storage space. Especially for large-scale data sets, optimizing storage efficiency can significantly reduce storage costs. By storing formatted data in a database or cloud storage platform, storage efficiency can be optimized, data security can be enhanced, and data sharing and collaboration can be supported. These benefits help enterprises better manage and utilize data resources and improve business efficiency and competitiveness.

[0056] The collection and scheduling method for logistics orders provided by the present invention effectively improves the data quality of real-time collection data by preprocessing the data and performing correlation analysis, and provides effective data support for dynamically adjusting logistics strategies. By inputting the data, the prediction model can calculate more accurate collection prediction results, and the scheduling plan further obtained in turn will be more reasonable.

[0057] See also Figure 3 A third embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0058] S301. Build a real-time alarm system, and detect the pre-processed real-time collection data through the real-time alarm system to exclude abnormal data;

[0059] Traditional data processing often requires a lot of manual review and proofreading, which is not only inefficient but also prone to errors. The real-time alarm system can automatically complete this process, greatly reducing the need for manual intervention. The real-time alarm system can detect data as it is generated, and promptly detect and eliminate abnormal data, thereby preventing these abnormal data from interfering with subsequent analysis and decision-making.

[0060] S302. Obtain the detected real-time collection data, and specify the prediction index series and influencing factor series in the real-time collection data;

[0061] When specifying the prediction indicator series, the data series that best reflects the system behavior characteristics is usually selected; when specifying the influencing factor series, the factor series that affects the system behavior is selected.

[0062] S303. Calculate the correlation coefficients between the prediction index series and the influencing factor series at each moment, and calculate the correlation degree based on multiple correlation coefficients.

[0063] Grey correlation method is a method used to analyze the correlation between various elements in a system. Its basic idea is to use the correlation to determine whether the relationship between various elements is close by the similarity between the curves of the prediction index series and other influencing factor series. Among them, the correlation is a quantitative expression of the correlation between the comparison series and the reference series, and usually takes the average value of the correlation coefficient.

[0064] During execution, you can build a real-time data processing pipeline, such as using Apache Kafka or Spark Streaming to receive real-time data streams, and then further process the data. Apache Kafka has high performance, high reliability and scalability, and has a huge ecosystem, including various connectors, client libraries and tools, which makes it easy to integrate with other systems and tools. Spark Streaming can handle large-scale data streams and ensure high throughput; it also has strong fault tolerance: Based on the fault tolerance mechanism of RDD, Spark Streaming can ensure reliable processing and transmission of data.

[0065] See also Figure 4 The fourth embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0066] S401. Obtain historical collection data, calculate the pre-processed historical collection data using the grey correlation method, and obtain historical correlation data;

[0067] Specifically, the historical collection data is of the same type as the aforementioned real-time collection data, and during preprocessing, the same processing steps are taken as for the real-time collection data to ensure consistency between the historical collection data and the real-time collection data obtained later.

[0068] S402. Using the BP neural network model as the initial model, the BP neural network model is trained using the historical association data to obtain the prediction model;

[0069] S403. Input the associated data into a prediction model to obtain a collection prediction result.

[0070] BP neural network is a multi-layer feedforward neural network model based on error back propagation, which consists of three layers of input, hidden and output network structures. Its basic idea is to continuously correct the multi-layer connection weights and thresholds through the back propagation algorithm to minimize the sum of square errors between the actual output value and the expected output value of the network. The BP neural network model is clear and suitable for data types with multiple feature inputs, such as historical collection data obtained from the logistics system.

[0071] During training, the BP neural network model can be adjusted and optimized according to different problems and data sets. By adjusting parameters such as network structure, activation function, and learning rate, the performance of the model can be further improved, making it more suitable for application scenarios of collection prediction results.

[0072] In addition, BP neural network can explain the decision-making process of the model through weights and bias values. This helps to understand the importance of different features to the model, so as to select and rank features and provide strong support for subsequent model optimization.

[0073] See also Figure 5 The fifth embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0074] S501. Set the parameters of the BP neural network model, set the number of input layer nodes to 10, the number of output layer nodes to 1, and the number of hidden layer nodes to 8, and obtain an initial model;

[0075] S502. Divide the historical association data in proportion to obtain a training set and a validation set;

[0076] S503. Use ReLu as the activation function, use the training set and the validation set to train the initial model, and obtain the prediction model.

[0077] By using Python software, the BP neural network model in the Sklearn model library can be called for training. After continuous debugging, ReLu is selected as the activation function, and its training accuracy is the best, which is 0.001. Since the logistics collection demand is affected by multiple factors, the input layer nodes are set to 10 and the output layer nodes are set to 1. After debugging, the training effect is best when the number of hidden layer nodes is 8. The data set is generally divided in a ratio of 7:3. This method selects the associated data as the training set and validation set. After repeated training of the model, when the number of training times reaches 100, the error is almost reduced to the minimum.

[0078] See also Figure 6 The sixth embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0079] S601. Obtain the collection forecast results, and make preliminary dispatch of the salesperson to collect the collection according to the collection forecast results to obtain dispatch feedback data;

[0080] Accurate data feedback can timely understand deficiencies and errors in the scheduling process, so that timely measures can be taken to improve and optimize, helping to improve the response speed of logistics services and obtain better scheduling arrangements.

[0081] S602. Perform abnormal inspection and preprocessing on the scheduling feedback data to obtain inspection passing data;

[0082] The obtained dispatch feedback data may contain errors or unreasonable values. These unreasonable data will affect the optimization of the dispatch plan. If not processed, they may even cause greater errors in the dispatch plan. For example, data entry errors, abnormal readings caused by equipment failures, etc. Specifically, when humans provide feedback data, they may make a mistake and enter an extra number, which will cause the value to increase by one digit, which will have a greater impact on subsequent dispatch adjustments. By performing an abnormality check on the data, these problem data can be discovered in a timely manner to avoid interfering with the optimization of the dispatch plan.

[0083] In addition, during the feedback process, various external factors may interfere with the scheduling feedback data, causing it to contain noise. For example, when data is fed back through different devices, noise is inevitably generated. Anomaly detection and preprocessing can help remove these noises and make the data purer.

[0084] S603. Re-dispatch the salespersons' collection according to the inspection pass data to obtain an optimized dispatch plan.

[0085] During the initial dispatch, there may be a problem that the dispatch result is inconsistent with the actual executable situation. At this time, the salesperson can feedback the corresponding problem through the system and form dispatch feedback data, such as feedback correction of the number of salespersons on duty at the outlets, correction feedback of the service scope of the outlets, etc. The dispatch plan can dynamically adjust the configuration of salespersons and the dispatch of vehicles according to the changes in collection demand. This helps to optimize resource allocation and improve operational efficiency. The dispatch feedback data can better optimize the dispatch plan and make it more reasonable and in line with the actual situation.

[0086] See also Figure 7 The seventh embodiment of the method for collecting and dispatching logistics orders in the embodiment of the present invention includes:

[0087] S701. Obtain resource allocation, including the distribution of outlets and the number of salesmen, pre-process the resource allocation, and obtain resource allocation data;

[0088] In addition to the distribution of outlets and the number of salesmen, resource allocation can also include transportation tools (such as the number of trucks, truck capacity, number of tricycles, etc.), storage facilities (such as warehouse size, number of shelves, etc.), sorting personnel (such as the number of sorting personnel, shift arrangements), and the number of necessary equipment for collection personnel (such as handheld terminals, scanners, packaging tools, etc.).

[0089] Since the resource allocation situation includes data from multiple sources, the formats and editing methods of these data are diverse. Therefore, before using these data, they need to be preprocessed to unify the data format and remove outliers and duplicate data.

[0090] S702. Using a simulated annealing algorithm, calculate the collection route based on the collection prediction results and resource allocation data;

[0091] Simulated Annealing (SA) is a heuristic global optimization algorithm that simulates the annealing process of solid materials. In physics, annealing refers to the process of slowly cooling a solid after heating it to a certain temperature, which can bring the material to a low energy state and reduce internal defects. Similarly, the simulated annealing algorithm also attempts to find a global optimal solution or a solution close to the global optimal solution. The simulated annealing algorithm can be applied to path optimization problems, such as the logistics order collection path optimization problem.

[0092] The simulated annealing algorithm is insensitive to the choice of initial solution and can find a solution close to the global optimal solution under different initial conditions. It can well deal with the rapid changes in real-time collection data and the impact of changes in multiple factors on collection scheduling. This robustness makes the simulated annealing algorithm more reliable in logistics collection scheduling applications.

[0093] S703. Output the collection route to the execution end to obtain scheduling feedback data.

[0094] When the collection route is calculated, it is sent to each logistics station for execution. After receiving the corresponding collection route arrangement, the logistics station can feedback from the system if there are unreasonable collection routes or erroneous data based on actual conditions, such as the delivery address is beyond the service scope of the logistics station, or the responsible salesperson is absent from work.

[0095] The above describes the method for dispatching the collection of logistics orders in the embodiment of the present invention. The following describes the device for dispatching the collection of logistics orders in the embodiment of the present invention. Figure 8 In one embodiment of the present invention, a collection and dispatching device for logistics orders includes:

[0096] The preprocessing module 10 is used to obtain real-time collection data and preprocess the real-time collection data;

[0097] The association module 20 is used to perform association analysis on the pre-processed real-time collection data using a grey association method to obtain associated data;

[0098] A prediction module 30, used for inputting the associated data into a prediction model to obtain a collection prediction result;

[0099] The scheduling module 40 is used to schedule the salesperson to collect goods according to the collection prediction result to obtain a scheduling plan.

[0100] See also Fig. 9 In one embodiment of the present invention, a collection and dispatching device for logistics orders includes:

[0101] The preprocessing module 10 is used to obtain real-time collection data and preprocess the real-time collection data;

[0102] The association module 20 is used to perform association analysis on the pre-processed real-time collection data using a grey association method to obtain associated data;

[0103] A prediction module 30, used for inputting the associated data into a prediction model to obtain a collection prediction result;

[0104] The scheduling module 40 is used to schedule the salesperson to collect goods according to the collection prediction result to obtain a scheduling plan;

[0105] In this embodiment, the preprocessing module 10 includes:

[0106] A removal unit 11 is used to obtain real-time collection data, the real-time collection data including collection weight, collection address, collection quantity, and collection type, and remove duplicate data from the real-time collection data to obtain first pre-processed data;

[0107] A filling unit 12, configured to fill missing values ​​of the first preprocessed data to obtain second preprocessed data;

[0108] A formatting unit 13, configured to format the second preprocessed data to obtain formatted data, and store the formatted data in a database or a cloud storage platform;

[0109] In this embodiment, the association module 20 includes:

[0110] The monitoring unit 21 is used to construct a real-time alarm system, through which the pre-processed real-time collection data is detected to exclude abnormal data;

[0111] An acquisition unit 22 is used to acquire the detected real-time collection data and specify the prediction index series and the influencing factor series in the real-time collection data;

[0112] A calculation unit 23 is used to calculate the correlation coefficients of the prediction index series and the influencing factor series at each moment, and calculate the correlation degree according to the multiple correlation coefficients;

[0113] In this embodiment, the prediction module 30 includes:

[0114] The calculation submodule 31 is used to obtain historical collection data, and use the grey correlation method to calculate the pre-processed historical collection data to obtain historical correlation data;

[0115] A construction submodule 32 is used to use the BP neural network model as an initial model and train the BP neural network model using the historical association data to obtain the prediction model;

[0116] A prediction submodule 33, used for inputting the associated data into a prediction model to obtain a collection prediction result;

[0117] In this embodiment, the prediction submodule 33 includes:

[0118] A setting unit 331 is used to set the parameters of the BP neural network model, setting the number of input layer nodes to 10, the number of output layer nodes to 1, and the number of hidden layer nodes to 8, to obtain an initial model;

[0119] A partitioning unit 332, used for partitioning the historical association data in proportion to obtain a training set and a validation set;

[0120] A training unit 333 is used to use ReLu as an activation function, and train the initial model using the training set and the validation set to obtain the prediction model;

[0121] In this embodiment, the scheduling module 40 includes:

[0122] The scheduling submodule 41 is used to obtain the collection prediction results, and perform preliminary scheduling of the salesperson's collection according to the collection prediction results to obtain scheduling feedback data;

[0123] The inspection submodule 42 is used to perform abnormal inspection and preprocessing on the scheduling feedback data to obtain inspection passing data;

[0124] An optimization submodule 43 is used to re-schedule the salesmen to collect goods according to the inspection pass data to obtain an optimized scheduling plan;

[0125] In this embodiment, the scheduling submodule 41 includes:

[0126] A preprocessing unit 411 is used to obtain resource allocation information, including network point distribution information and the number of salesmen, and preprocess the resource allocation information to obtain resource allocation data;

[0127] A planning unit 412, configured to calculate a collection route according to the collection prediction result and resource allocation data by using a simulated annealing algorithm;

[0128] The output unit 413 is used to output the collection route to the execution end to obtain scheduling feedback data.

[0129] The collection and dispatching device for logistics orders provided by the present invention obtains real-time collection data, pre-processes the real-time collection data, and then uses the grey correlation method to perform correlation analysis on the pre-processed real-time collection data to obtain the correlated data, which can effectively evaluate the correlation between different factors in the real-time collection data, and then analyze the correlated data through a trained prediction model to obtain a reliable collection prediction result, and finally formulate a dispatching plan based on the collection prediction result, so that the dispatching plan is more reliable. In addition, the present invention can continuously adjust and optimize the dispatching plan according to the real-time collection data to ensure that the dispatching is timely and flexible.

[0130] The above is a detailed description of the collection and scheduling device for logistics orders in an embodiment of the present invention from the perspective of modular functional entities. The following is a detailed description of the collection and scheduling device for logistics orders in an embodiment of the present invention from the perspective of hardware processing.

[0131] Fig.10 A schematic diagram of the structure of a collection and scheduling device for a logistics order provided in an embodiment of the present invention, wherein the collection and scheduling device 900 for a logistics order may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 910 (for example, one or more processors) and a memory 920, and one or more storage media 930 (for example, one or more mass storage devices) storing application programs 933 or data 932. Among them, the memory 920 and the storage medium 930 may be short-term storage or permanent storage. The program stored in the storage medium 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the collection and scheduling device 900 for a logistics order. Furthermore, the processor 910 may be configured to communicate with the storage medium 930, and execute a series of instruction operations in the storage medium 930 on the collection and scheduling device 900 for a logistics order, so as to implement the steps of the collection and scheduling method for a logistics order provided in the above-mentioned method embodiments.

[0132] The logistics order collection and dispatching device 900 may also include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input and output interfaces 960, and / or one or more operating systems 931, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Fig.10 The structure of the collection and scheduling equipment for logistics orders shown does not constitute a limitation on the collection and scheduling equipment for logistics orders, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0133] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of a method for collecting and scheduling logistics orders.

[0134] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the equipment or device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0135] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0136] It is understandable that those skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of the present invention, and all these changes or substitutions should fall within the protection scope of the claims attached to the present invention.

Claims

1. A method for collecting and dispatching logistics orders, characterized in that: The steps include: Obtain real-time collection data and pre-process the real-time collection data; Use the grey correlation method to perform correlation analysis on the pre-processed real-time collection data to obtain the correlated data; Inputting the associated data into a prediction model to obtain a collection prediction result; The salesmen are scheduled to collect the goods according to the collection forecast results to obtain a scheduling plan.

2. The method for collecting and dispatching logistics orders according to claim 1, characterized in that: The obtaining of real-time collection data and preprocessing of the real-time collection data specifically include: Acquire real-time collection data, the real-time collection data including collection weight, collection address, collection quantity, and collection type, and remove duplicate data from the real-time collection data to obtain first pre-processed data; Filling missing values ​​of the first preprocessed data to obtain second preprocessed data; The second preprocessed data is formatted to obtain formatted data, and the formatted data is stored in a database or a cloud storage platform.

3. The method for collecting and dispatching logistics orders according to claim 1, characterized in that: The grey correlation method is used to perform correlation analysis on the pre-processed real-time collection data to obtain correlation data, specifically including: Build a real-time alarm system to detect the pre-processed real-time collection data to eliminate abnormal data; Obtain the detected real-time collection data, and specify the prediction index series and influencing factor series in the real-time collection data; Calculate the correlation coefficients between the prediction indicator series and the influencing factor series at each moment, and calculate the correlation degree based on multiple correlation coefficients.

4. The method for collecting and dispatching logistics orders according to claim 1, characterized in that: The step of inputting the associated data into a prediction model to obtain a collection prediction result specifically includes: Obtain historical collection data, use the grey correlation method to calculate the pre-processed historical collection data, and obtain historical correlation data; Taking the BP neural network model as the initial model, the BP neural network model is trained using the historical association data to obtain the prediction model; The associated data is input into a prediction model to obtain a collection prediction result.

5. The method for collecting and dispatching logistics orders according to claim 4 is characterized in that: The BP neural network model is used as the initial model, and the historical association data is used to train the BP neural network model to obtain the prediction model, specifically including: The parameters of the BP neural network model are set, the number of input layer nodes is set to 10, the number of output layer nodes is set to 1, and the number of hidden layer nodes is set to 8, and the initial model is obtained; Divide the historical correlation data into proportions to obtain a training set and a validation set; ReLu is used as the activation function, and the initial model is trained using the training set and the validation set to obtain the prediction model.

6. The method for collecting and dispatching logistics orders according to claim 1, characterized in that: The method of scheduling the salesperson's collection according to the collection prediction result to obtain a scheduling plan specifically includes: Obtain the collection forecast results, and make preliminary dispatch of salesmen to collect the goods according to the collection forecast results to obtain dispatch feedback data; Performing abnormality inspection and preprocessing on the scheduling feedback data to obtain inspection passing data; The salesmen are rescheduled to collect goods based on the inspection data to obtain an optimized dispatch plan.

7. The method for collecting and dispatching logistics orders according to claim 6, characterized in that: The obtaining of the collection prediction result and the preliminary scheduling of the salesperson's collection according to the collection prediction result to obtain the scheduling feedback data specifically include: Obtaining resource allocation information, including network point distribution information and the number of salespersons, and preprocessing the resource allocation information to obtain resource allocation data; Using a simulated annealing algorithm, the collection route is calculated according to the collection prediction results and resource allocation data; The collection route is output to the execution end to obtain scheduling feedback data.

8. A collection and dispatching device for logistics orders, characterized in that: include: A preprocessing module is used to obtain real-time collection data and preprocess the real-time collection data; The correlation module is used to perform correlation analysis on the pre-processed real-time collection data using the grey correlation method to obtain correlated data; A prediction module, used for inputting the associated data into a prediction model to obtain a collection prediction result; The scheduling module is used to schedule the salesperson's collection according to the collection forecast results to obtain a scheduling plan.

9. A collection and dispatching device for logistics orders, characterized in that: comprising a memory and at least one processor, wherein the memory has computer-readable instructions stored therein; The at least one processor calls the computer-readable instructions in the memory to execute the various steps of the collection and scheduling method for logistics orders as described in any one of claims 1-7.

10. A computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by the processor, the various steps of the method for collecting and scheduling logistics orders as described in any one of claims 1 to 7 are implemented.

Images