Management method and system for realizing remote unmanned supervision based on LBS (Location Based Service)

By constructing an LBS-based unsupervised learning pattern recognition model, the behavioral patterns of sales personnel are identified and management suggestions are provided, solving the problem of low efficiency in sales process management in existing technologies and achieving efficient and accurate remote management.

CN119004316APending Publication Date: 2024-11-22SHENZHEN FENGYI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411066259.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Existing location-based services (LBS) sales personnel management systems lack effective supervision and management of the sales process, resulting in problems such as inefficiency, inaccurate data, and high management costs.

Method used

By acquiring the location and sales information of sales personnel in real time based on LBS, a pattern recognition model based on unsupervised learning is constructed. Using clustering algorithms, anomaly detection algorithms, dimensionality reduction algorithms, autoencoders, and generative adversarial networks, the model identifies the behavioral patterns of sales personnel and triggers an early warning mechanism. Management suggestions are then presented in conjunction with visualization tools.

Benefits of technology

It improved the efficiency and accuracy of the sales process, reduced management costs, and enabled efficient management with remote, unattended supervision.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119004316A_ABST
    Figure CN119004316A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of Internet, and provides a management method and system for realizing remote unmanned supervision based on LBS, and the method comprises the steps: obtaining the position information and sales information of a target person in real time based on LBS, the position information comprising latitude, speed and direction, and the sales information comprising order number, sales amount and customer feedback; based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimension reduction algorithm, an auto-encoder and a generative adversarial network, utilizing historical data of the position information and the sales information to construct a mode recognition model based on unsupervised learning; performing behavior pattern recognition on the real-time position information and sales information by using a pattern recognition model to judge whether the behavior of the target person is normal or abnormal, and triggering an early warning mechanism when the behavior is abnormal; displaying an identification result of the mode identification model and a management suggestion corresponding to the target person through a visual tool; according to the invention, the management problem of field sales personnel in the prior art can be solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and in particular to a management method and system for realizing remote unsupervised operation based on LBS. Background Art

[0002] At present, in enterprise management, the process management of sales personnel who need to work outside mainly relies on manual supervision and reporting, which has problems such as low efficiency, inaccurate data, and high management costs. The existing location-based service (LBS) sales management system often only focuses on the location information of sales personnel and lacks effective supervision and management of the sales process. Summary of the invention

[0003] In view of the above technical problems, the present invention provides a management method and system for realizing remote unsupervised operation based on LBS, so as to solve the management problem of field sales personnel in the prior art.

[0004] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0005] According to one aspect of the present disclosure, a method for implementing remote unsupervised management based on LBS is disclosed, the method comprising:

[0006] Based on LBS, the location information and sales information of the target person are obtained in real time. The location information includes latitude, speed and direction, and the sales information includes order quantity, sales volume, and customer feedback;

[0007] Based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimensionality reduction algorithm, an autoencoder, and a generative adversarial network, a pattern recognition model based on unsupervised learning is constructed using the historical data of the location information and the sales information;

[0008] Using the pattern recognition model to perform behavioral pattern recognition on the real-time location information and sales information to determine whether the behavior of the target person is normal or abnormal, and triggering an early warning mechanism when it is abnormal;

[0009] The recognition results of the pattern recognition model and the management suggestions corresponding to the target personnel are displayed by a visualization tool.

[0010] Furthermore, after acquiring the location information, the method further includes:

[0011] The location information is cleaned, integrated and formatted to obtain the travel path, stay time and visit frequency of the target person.

[0012] Furthermore, when constructing the pattern recognition model, it includes:

[0013] Acquire unlabeled training data, where the training data is historical data of the location information and the sales information;

[0014] Cleaning, denoising, and normalizing the training data to eliminate outliers and missing values ​​in the training data;

[0015] Extracting features from the training data, wherein the features include at least location, visit frequency, and stay time;

[0016] Select one of the clustering algorithm, anomaly detection algorithm, dimensionality reduction algorithm, autoencoder, and generative adversarial network, and define the corresponding objective function or loss function;

[0017] According to the grade of the target personnel, the sales method preference value, and the sales ability level, the initial values ​​of the model parameters are defined, the objective function or the loss function is minimized by calculating the features, and the calculation is performed iteratively so that the parameters of the pattern recognition model are optimized to obtain the pattern recognition model.

[0018] Furthermore, after obtaining the pattern recognition model, a validation set or a test set is used to evaluate it.

[0019] Furthermore, the method further comprises:

[0020] Combining the sales information and the location information, a method of evaluating the work performance of the target personnel is performed based on regression analysis and time series analysis.

[0021] According to another aspect of the present disclosure, a management system for implementing remote unsupervised operation based on LBS is disclosed, the system comprising:

[0022] A data collection module is used to obtain the location information and sales information of the target person in real time based on LBS, wherein the location information includes latitude, speed and direction, and the sales information includes order quantity, sales volume, and customer feedback;

[0023] A data analysis module, for constructing a pattern recognition model based on unsupervised learning using the historical data of the location information and the sales information based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimensionality reduction algorithm, an autoencoder, and a generative adversarial network, and using the pattern recognition model to perform behavioral pattern recognition on the real-time location information and the sales information to determine whether the behavior of the target person is normal or abnormal, and triggering an early warning mechanism when it is abnormal;

[0024] The management module is used to display the recognition results of the pattern recognition model and the management suggestions corresponding to the target personnel by means of a visualization tool.

[0025] The technical solution disclosed in this disclosure has the following beneficial effects:

[0026] By collecting sales staff's location information and sales data in real time and using algorithm logic for remote unmanned supervision and management, the problems existing in existing technologies are solved, the efficiency and accuracy of the sales process are improved, and management costs are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of a remote unsupervised management method based on LBS in an embodiment of this specification;

[0028] Figure 2 This is a structural block diagram of a management system that implements remote unsupervised operation based on LBS in an embodiment of this specification. DETAILED DESCRIPTION

[0029] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0030] In addition, the accompanying drawings are only schematic illustrations of the present disclosure. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0031] like Figure 1 As shown, the embodiment of this specification provides a remote unsupervised management method based on LBS, the execution subject of the method can be a computer or a server, and the method includes steps S101-104:

[0032] In step S101, based on LBS, the location information and sales information of the target person are acquired in real time, the location information including latitude, speed and direction, and the sales information including order quantity, sales amount, and customer feedback.

[0033] Specifically, the target personnel can be sales personnel, who can use LBS technology, i.e., location-based services, such as applications installed on mobile devices, to collect the sales personnel's location information and behavior data in real time, and obtain the sales personnel's dimensions, speed, direction, number of orders, sales, and customer feedback, among which, latitude: the latitude of the sales personnel's current geographical location, speed: the speed at which the sales personnel moves, which can reflect their movement status, direction: the direction in which the sales personnel moves, which helps to understand their path of action, number of orders: the total number of orders completed by the sales personnel within a certain period of time, sales: the total sales of the sales personnel within a certain period of time, customer feedback: the customer's evaluation of the sales personnel's services or products, which reflects the service quality. This information provides basic data for subsequent data analysis and pattern recognition.

[0034] In step S102, based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimensionality reduction algorithm, an autoencoder, and a generative adversarial network, a pattern recognition model based on unsupervised learning is constructed using the historical data of the location information and the sales information.

[0035] Specifically, based on clustering algorithms, sales staff behaviors are divided into different groups or "clusters" to discover behavioral patterns; based on anomaly detection algorithms, behavioral patterns that do not conform to the norm are identified, indicating problems or areas that need attention; based on dimensionality reduction algorithms, such as PCA (principal component analysis), the dimensions of the data are reduced while retaining the most important information; based on autoencoders, a type of neural network, used to learn effective encoding of data, that is, for feature learning; based on generative adversarial networks (GANs), a network composed of generators and discriminators, which can be used to generate new data samples or improve the representation of data. Through these algorithms, the system can learn the behavioral patterns of target personnel and lay the foundation for subsequent real-time analysis to identify possible abnormal behaviors of target personnel.

[0036] In step S103, the pattern recognition model is used to perform behavioral pattern recognition on the real-time location information and the sales information to determine whether the behavior of the target person is normal or abnormal, and trigger an early warning mechanism when it is abnormal.

[0037] Specifically, normal behavior is when salespeople perform tasks according to established workflows and plans, and abnormal behavior patterns are when salespeople’s behavior deviates from the normal pattern and requires further investigation or intervention. If abnormal behavior is identified, an early warning mechanism is automatically triggered to notify management for further review or action.

[0038] In step S104, the recognition results of the pattern recognition model and the management suggestions corresponding to the target personnel are displayed by a visualization tool.

[0039] Specifically, visualization tools can be charts, heat maps, time series charts, etc. to show the salesperson's behavior patterns and sales performance. Management suggestions can be customized suggestions based on the identification results, such as optimizing visit routes, adjusting sales strategies, etc.

[0040] In the above implementation mode, through this method, managers can more intuitively understand the working status of sales personnel, discover problems in a timely manner, and formulate corresponding management measures, thereby improving the efficiency and effectiveness of the entire sales team.

[0041] The key to the entire approach is real-time, accuracy, and personalized management recommendations, which are achieved through carefully designed steps of data collection, model building, and result presentation. This approach can effectively reduce the need for manual supervision, reduce management costs, and improve the overall performance of the sales team.

[0042] In one embodiment, after acquiring the location information, the method further includes:

[0043] The location information is cleaned, integrated and formatted to obtain the travel path, stay time and visit frequency of the target person.

[0044] Among them, cleaning, integrating and formatting location information is an important step in data analysis, especially in location-based service (LBS) management systems. The purpose of data cleaning is to ensure the quality and accuracy of data. The main steps include:

[0045] Noise Removal: Identify and remove outliers or erroneous records in the data, such as impossible speed values ​​or unreasonable position jumps.

[0046] Filling missing values: For missing location information, interpolation methods or data from the previous time point can be used to estimate it.

[0047] Format uniformity: Make sure all data follows the same format and units, such as converting all latitude and longitude coordinates to the same coordinate system.

[0048] Data integration is the process of combining data from different sources into a unified data set. The steps include:

[0049] Data fusion: Combine the salesperson’s location information with sales data (number of orders, sales, etc.) to form a comprehensive data set.

[0050] Time synchronization: Ensure that location information and sales data are consistent in time, for example, associating each location record with sales activity at the same point in time.

[0051] Data formatting is the process of converting data into a format suitable for analysis and processing, including:

[0052] Normalization: Converting data to a uniform metric, for example, converting speed from kilometers per hour to meters per second.

[0053] Structuring: Organizing data into tabular form to facilitate further analysis and processing.

[0054] Path reconstruction: Based on the salesperson’s location information, reconstruct their movement path, including the starting point, end point, and route passed.

[0055] Route Optimization: Analyze the efficiency of a route and identify possible optimization points, such as reducing detours or avoiding traffic congestion areas.

[0056] Timestamp analysis: Use timestamps in location information to determine the salesperson's arrival and departure times at various stopover points.

[0057] Dwell time statistics: Calculate the time salespeople spend at each customer location or region to assess visit efficiency.

[0058] Visit frequency statistics specifically include visit records and frequency analysis:

[0059] Visit records: Record the number and frequency of sales staff's visits to each customer.

[0060] Frequency Analysis: Analyze the visiting patterns of salespeople to identify customers or areas that are visited frequently, as well as low-frequency areas that may require increased attention.

[0061] Behavioral pattern recognition includes pattern recognition and behavioral scoring:

[0062] Pattern recognition: Using cleaned, integrated and formatted data, algorithms are used to analyze sales staff’s behavioral patterns, such as normal visits, frequent stops or abnormal movements.

[0063] Behavioral scoring: Based on the results of behavioral pattern analysis, sales staff’s behaviors are scored or classified for easier management and optimization.

[0064] Through the above steps, valuable information can be effectively extracted from the original location information, such as travel path, dwell time and visit frequency, which is crucial for remote unsupervised sales process management. Through in-depth analysis of these data, the sales team's work efficiency can be improved, resource allocation can be optimized, and potential problems can be identified and resolved in a timely manner.

[0065] In one embodiment, when constructing the pattern recognition model, it includes: obtaining unlabeled training data, wherein the training data is historical data of the location information and the sales information; cleaning, denoising, and normalizing the training data to eliminate outliers and missing values ​​in the training data; extracting features in the training data, wherein the features include at least location, visit frequency, and residence time; selecting one of a clustering algorithm, anomaly detection algorithm, dimensionality reduction algorithm, autoencoder, and generative adversarial network, and defining a corresponding objective function or loss function; defining the initial values ​​of the model parameters according to the level of the target personnel, the sales method preference value, and the sales ability level, minimizing the objective function or the loss function by calculating the features, and iteratively calculating so that the parameters of the pattern recognition model are optimized to obtain the pattern recognition model.

[0066] Among them, historical data is collected, including the location information and sales information of the salesperson, which are unlabeled, that is, there is no predefined category or result. After preprocessing the historical data according to the above implementation method, feature extraction is performed, which includes: location features, that is, extracting the geographical location information of the salesperson, such as latitude and longitude, moving path, etc.; visit frequency, that is, calculating the number or frequency of salespersons visiting customers; stay time, that is, recording the average stay time of salespersons at each customer location. Then select the algorithm and define the objective function. According to business needs and data characteristics, select a suitable unsupervised learning algorithm, such as clustering, anomaly detection, dimensionality reduction, etc., and then define the objective function or loss function for the selected algorithm, which will guide the training process of the model. For example, in the clustering algorithm, the silhouette coefficient or Davis-Bolding index is used as the optimization target. According to the salesperson's level, sales method preference value, sales ability level and other information, the initial parameter values ​​are defined for the model. These parameters include the number of cluster centers, the hidden layer size of the autoencoder, etc. The main goal of model training is to minimize the objective function or loss function by calculating features, and then use gradient descent or other optimization algorithms for iterative calculations, continuously adjust model parameters to optimize performance, monitor the performance of the model during training, and stop training when the preset convergence condition or number of iterations is reached. At the end of training, use the validation set or test set to evaluate the model, select appropriate evaluation indicators, such as the silhouette coefficient of the clustering algorithm, and adjust the model parameters or replace the algorithm according to the evaluation results to improve the accuracy and generalization ability of the model. Applying the trained pattern recognition model to actual tasks can classify and identify the behavior patterns of sales personnel.

[0067] In one embodiment, the method further includes: combining the sales information and the location information to evaluate the work performance of the target personnel based on regression analysis or time series analysis.

[0068] Combining sales information and location information to evaluate the target person's job performance is a multi-step process, including:

[0069] Merge the salesperson's location information with the sales information to form a comprehensive data set, ensuring that the two are synchronized in time; extract useful features from the location information, such as visit frequency, average length of stay, customer visit patterns, etc.; extract or construct features from the sales information, such as sales, order quantity, customer satisfaction, etc. Select an appropriate regression model, such as linear regression, logistic regression, or a more complex regression model to evaluate the relationship between sales performance and features; use regression analysis to estimate model parameters and determine the degree of influence of each feature on sales performance.

[0070] Then use time series analysis to analyze the trend of sales performance over time, identify seasonal patterns or long-term trends, and evaluate whether there are cyclical fluctuations in sales performance, such as weekly, monthly or quarterly changes. Apply ARIMA, seasonal decomposition time series forecasting models, etc. to predict future sales performance.

[0071] Define indicators for evaluating sales staff's work performance, such as sales growth rate, customer visit efficiency, customer satisfaction, etc., and calculate performance scores for each sales staff based on the results of regression analysis and time series analysis.

[0072] At the same time, standard deviation analysis is performed to identify significant deviations from the average sales performance, and thresholds are set to distinguish between normal performance and abnormal performance. Finally, charts and graphs are produced to display the performance evaluation results of sales personnel, such as bar charts, line charts, etc., and a dashboard is built to integrate all key performance indicators so that managers can quickly understand team performance.

[0073] In addition, performance evaluation can also be used for strategy formulation. The evaluation results can be used to help formulate sales strategies, such as adjusting sales targets, optimizing visit plans, etc., and based on the performance evaluation results, reasonably allocate sales resources, such as manpower and materials.

[0074] Based on the same content, such as Figure 2 As shown, a management system based on LBS to realize remote unsupervised operation, the system includes:

[0075] The data collection module 201 is used to obtain the location information and sales information of the target person in real time based on LBS. The location information includes latitude, speed and direction, and the sales information includes order quantity, sales volume, and customer feedback.

[0076] The data analysis module 202 is used to construct a pattern recognition model based on unsupervised learning based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimensionality reduction algorithm, an autoencoder, and a generative adversarial network, using the historical data of the location information and the sales information, and use the pattern recognition model to perform behavioral pattern recognition on the real-time location information and the sales information to determine whether the behavior of the target person is normal or abnormal, and trigger an early warning mechanism when it is abnormal.

[0077] The management module 203 is used to display the recognition results of the pattern recognition model and the management suggestions corresponding to the target personnel by using a visualization tool.

[0078] In the above implementation, by collecting the location information and sales data of sales personnel in real time and using algorithm logic for remote unmanned supervision and management, the problems existing in the prior art are solved, the efficiency and accuracy of the sales process are improved, and management costs are reduced.

[0079] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A remote unsupervised management method based on LBS, It is characterized in that The method comprises: Based on LBS, the location information and sales information of the target person are obtained in real time. The location information includes latitude, speed and direction, and the sales information includes order quantity, sales volume, and customer feedback; Based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimensionality reduction algorithm, an autoencoder, and a generative adversarial network, a pattern recognition model based on unsupervised learning is constructed using the historical data of the location information and the sales information; Using the pattern recognition model to perform behavioral pattern recognition on the real-time location information and sales information to determine whether the behavior of the target person is normal or abnormal, and triggering an early warning mechanism when it is abnormal; The recognition results of the pattern recognition model and the management suggestions corresponding to the target personnel are displayed by a visualization tool.

2. According to claim 1, a remote unsupervised management method based on LBS is implemented. It is characterized in that After acquiring the location information, the method further includes: The location information is cleaned, integrated and formatted to obtain the travel path, stay time and visit frequency of the target person.

3. According to claim 1, a remote unsupervised management method based on LBS is implemented. It is characterized in that When constructing the pattern recognition model, it includes: Acquire unlabeled training data, where the training data is historical data of the location information and the sales information; Cleaning, denoising, and normalizing the training data to eliminate outliers and missing values ​​in the training data; Extracting features from the training data, wherein the features include at least location, visit frequency, and stay time; Select one of the clustering algorithm, anomaly detection algorithm, dimensionality reduction algorithm, autoencoder, and generative adversarial network, and define the corresponding objective function or loss function; According to the grade of the target personnel, the sales method preference value, and the sales ability level, the initial values ​​of the model parameters are defined, the objective function or the loss function is minimized by calculating the features, and the calculation is performed iteratively so that the parameters of the pattern recognition model are optimized to obtain the pattern recognition model.

4. According to claim 3, a remote unsupervised management method based on LBS is implemented. It is characterized in that After obtaining the pattern recognition model, it is evaluated using a validation set or a test set.

5. According to claim 1, a remote unsupervised management method based on LBS is implemented. It is characterized in that The method further comprises: Combining the sales information and the location information, a method of evaluating the work performance of the target personnel is performed based on regression analysis and time series analysis.

6. A remote unsupervised management system based on LBS, It is characterized in that The system comprises: A data collection module is used to obtain the location information and sales information of the target person in real time based on LBS, wherein the location information includes latitude, speed and direction, and the sales information includes order quantity, sales volume, and customer feedback; A data analysis module, for constructing a pattern recognition model based on unsupervised learning using the historical data of the location information and the sales information based on one or more of a clustering algorithm, an anomaly detection algorithm, a dimensionality reduction algorithm, an autoencoder, and a generative adversarial network, and using the pattern recognition model to perform behavioral pattern recognition on the real-time location information and the sales information to determine whether the behavior of the target person is normal or abnormal, and triggering an early warning mechanism when it is abnormal; The management module is used to display the recognition results of the pattern recognition model and the management suggestions corresponding to the target personnel by means of a visualization tool.