Campus leg running platform architecture based on intelligent matching algorithm

Abstract

The invention provides a campus leg running platform architecture based on an intelligent matching algorithm, and relates to the field of campus leg running. The system comprises a front-end layer, an application layer, a data layer and an infrastructure layer. The front-end layer comprises a user side, a leg builder side and an administrator side, and a user interacts through a mobile terminal application or a Web side to complete operations such as task issuing, order receiving, payment and task state checking; the application layer comprises a task service, an intelligent matching service, a scheduling service, a map and navigation service and a payment and settlement service, an appropriate leg running person is automatically matched for each task through an algorithm, the task distribution efficiency is improved, and platform resources are optimized. Through a set intelligent algorithm formula, the position of the leg running person, the priority of the task and historical data prediction are combined for calculation and are allocated to the leg running person with the highest score for execution, so that the leg running efficiency is improved, and the time is saved.

Description

Technical Field

[0001] The present invention relates to the field of campus errand running, and specifically to a campus errand running platform architecture based on an intelligent matching algorithm. Background Art

[0002] The instant delivery industry is experiencing rapid growth and transformation. Driven by the continued growth of e-commerce and increasing consumer demand for fast and convenient delivery services, the sector is booming. Furthermore, the application of new technologies, including drone delivery, autonomous vehicles, data analytics, and artificial intelligence, is driving advancements in the industry. As competition intensifies, many companies are continuously improving service quality and delivery efficiency to meet customer needs and remain competitive. Furthermore, environmental protection and sustainable development have become key industry concerns, with many companies exploring initiatives such as using eco-friendly transportation and reducing packaging waste. However, the industry's rapid expansion has also brought challenges, including urban traffic congestion, "last mile" delivery difficulties, and safety concerns. Consequently, the instant delivery industry faces pressure to continuously innovate and develop solutions. However, the instant delivery industry remains a vibrant and promising sector, requiring continuous response and improvement to adapt to market changes and provide better services to consumers.

[0003] With the rapid development of internet technology, people's pace of life has further accelerated. Comprehensive online service platforms like online shopping and food delivery have become mainstream, and university students are no exception. At many universities, "campus errands" are increasingly becoming a new symbol of fast-paced life. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides a campus errand platform architecture based on an intelligent matching algorithm, which solves the problems of untimely errand delivery and chaotic task allocation.

[0006] (2) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a campus errand running platform architecture based on intelligent matching algorithm, including a front-end layer, an application layer, a data layer and an infrastructure layer;

[0008] The front-end layer includes the user side, the errand runner side, and the administrator side. Users interact through mobile applications or web terminals to complete operations such as task publishing, order acceptance, payment, and task status checking.

[0009] The application layer includes task services, intelligent matching services, scheduling services, map and navigation services, and payment and settlement services. It uses algorithms to automatically match appropriate runners for each task, improving task allocation efficiency and optimizing platform resources.

[0010] The data layer includes relational databases and NoSQL databases, which are responsible for the platform's data storage and management. The selection of databases and caches should be optimized according to the system's performance requirements;

[0011] The infrastructure layer includes cloud platform, message queue, containerization and microservices, log and monitoring. The infrastructure layer ensures the availability and stability of the system and adopts cloud platform, containerized deployment and other technologies.

[0012] Preferably, the user terminal can publish tasks, query task status, make payments, and evaluate errand runners. At the same time, the published tasks are divided into on-campus tasks and off-campus tasks to provide multiple options;

[0013] The errand runner end includes receiving tasks, processing tasks, executing tasks, navigating, and confirming task completion. At the same time, some campuses prohibit outsiders from entering, and these off-campus tasks are completed by errand runners in a relay manner;

[0014] The administrator side includes managing users and errand runners, reviewing tasks, and viewing platform data statistics and reports.

[0015] Preferably, the task service is responsible for the creation, status update, and history record management of user tasks. After the user publishes a task, the task service will request a suitable runner from the intelligent matching service.

[0016] Preferably, the intelligent matching service performs matching based on task requirements, errand runner conditions, and real-time data, including:

[0017] Location-based matching algorithm: Matches runners based on the task posting location and their current location, giving priority to runners who are closer.

[0018] Matching task priority with errand runner ratings: If the task is urgent, consider increasing its priority and selecting the errand runner based on their ratings and past performance.

[0019] Prediction algorithm based on historical data: Use machine learning algorithms to analyze historical task data and the task completion status of runners to predict the most suitable runners;

[0020] Multi-objective optimization algorithm: Combines factors such as the runner's location, time, task type, and completion speed to conduct a comprehensive evaluation and select the best runner;

[0021] The comprehensive algorithm formula is as follows: based on the geographical location of tasks and runners, the distance (d_i) between the task location and the runner's current location is calculated; the smaller the distance, the higher the priority for selecting the runner;

[0022] [d_i=\sqrt{(x_{\text{task}}-x_{\text{courier}i})^2+(y{\text{task}}-y_{\text{courier}_i})^2}]

[0023] Here, (d_i) represents the Euclidean distance between the task posting location and the errand runner's current location;

[0024] If the task is urgent, we adjust the matching weight based on the task priority and the runner's score. For urgent tasks, we can assign a priority adjustment coefficient (w_{\text{task}}) to increase the priority of task matching.

[0025] For urgent tasks, the task priority adjustment factor (w_{\text{task}}=\alpha) (e.g. (\alpha=2));

[0026] For non-urgent tasks, the task priority adjustment coefficient (w_{\text{task}}=1);

[0027] The runner's rating (R_{\text{courier}i}) and historical performance (H{\text{courier}_i}) will be combined to form a comprehensive score:

[0028] [S_{\text{courier}i}=\beta\cdot R{\text{courier}i}+\gamma\cdot H{\text{courier}_i}]

[0029] (R_{\text{courier}_i}) is the courier score;

[0030] (H_{\text{courier}_i}) is the courier's historical performance (e.g., task completion rate);

[0031] (\beta,\gamma) are weight coefficients used to balance the runner's rating and historical performance;

[0032] Use machine learning algorithms to analyze historical task data to predict which runner is best suited for a task. This prediction can be viewed as a comprehensive matching score (P_{\text{courier}_i}), which can be calculated based on historical data and task characteristics.

[0033] P_{\text{courier}_i}=f(\text{historical data},\text{task characteristics},\text{courier characteristics})

[0034] A multi-objective optimization is performed by comprehensively considering factors such as the runner's location, time, task type, task priority, runner's rating, and historical data. The comprehensive score (M_{\text{courier}_i}) can be calculated using the following formula:

[0035] [M_{\text{courier}_i}=w{\text{task}}\cdot\frac{1}{d_i}+\lambda_1\cdot S_{\text{courier}_i}+\lambda_2\cdot P_{\text{courier}_i}]

[0036] in:

[0037] (\frac{1}{d_i}) is the inverse of the position, indicating that the closer the distance, the higher the score;

[0038] (w_{\text{task}}) is the task priority adjustment coefficient (taking into account the urgency of the task);

[0039] (S_{\text{courier}_i}) is the comprehensive score of the courier (score + historical performance);

[0040] (P_{\text{courier}_i}) is the prediction score based on historical data;

[0041] (\lambda_1) and (\lambda_2) are weight coefficients used to balance the influence of various factors;

[0042] All couriers are sorted by their comprehensive scores (M_{\text{courier}_i}) and the courier with the highest score is selected as the executor of the task.

[0043] Preferably, the dispatching service ensures the rationality of the errand runner's task when he accepts the task, and avoids the errand runner from taking too many orders or being too far away; the dispatching service can also dynamically adjust the task allocation to deal with situations where the errand runner cannot complete the task in time due to weather or traffic jams; the map and navigation service is a map service integrated with map APIs such as Amap, Baidu or Google Maps, providing accurate location tracking, route planning and navigation functions to help errand runners complete tasks more efficiently; the payment and settlement service is the module responsible for payment and settlement after the task is completed, and supports payment interfaces such as Alipay and WeChat Pay; the settlement system will also settle commissions based on the performance of the errand runner.

[0044] Preferably, the data layer includes:

[0045] Relational database: stores user information, errand runner information, task records, payment information, etc.

[0046] NoSQL database: used to store real-time information about tasks, dynamic data about errand runners, etc.

[0047] Caching: Use caching technologies such as Redis to store frequently queried data, such as task status, runner location, etc.

[0048] Preferably, the infrastructure layer includes:

[0049] Cloud platform: provides elastic computing, storage and other resources to ensure high availability of the platform;

[0050] Message queues: such as Kafka or RabbitMQ, to ensure reliable delivery of tasks and notifications;

[0051] Containerization and microservices: Implementing microservice architecture through Docker and Kubernetes improves system scalability and operation and maintenance efficiency;

[0052] Logging and monitoring: Use logging and monitoring tools to monitor system status and troubleshoot problems.

[0053] (3) Beneficial effects

[0054] The present invention provides a campus errand platform architecture based on an intelligent matching algorithm. It has the following beneficial effects:

[0055] 1. The intelligent algorithm formula combines the runner's position, task priority and historical data prediction to calculate and assign the task to the runner with the highest score to improve the efficiency of running errands and save time. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1This is a schematic diagram of the overall architecture of the campus errand running platform architecture based on the intelligent matching algorithm proposed in the present invention. DETAILED DESCRIPTION

[0057] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0058] Example 1:

[0059] like Figure 1 As shown, the embodiment of the present invention provides a campus errand platform architecture based on an intelligent matching algorithm, including a front-end layer, an application layer, a data layer, and an infrastructure layer;

[0060] The front-end layer includes the user side, the errand runner side, and the administrator side. Users interact through mobile applications or web terminals to complete operations such as task publishing, order acceptance, payment, and task status checking.

[0061] The application layer includes task services, intelligent matching services, scheduling services, map and navigation services, and payment and settlement services. It uses algorithms to automatically match appropriate runners for each task, improving task allocation efficiency and optimizing platform resources.

[0062] The data layer includes relational databases and NoSQL databases, which are responsible for the platform's data storage and management. The selection of databases and caches should be optimized according to the system's performance requirements;

[0063] The infrastructure layer includes cloud platform, message queue, containerization and microservices, log and monitoring. The infrastructure layer ensures the availability and stability of the system and adopts cloud platform, containerized deployment and other technologies.

[0064] The user terminal can publish tasks, query task status, make payments, and evaluate errand runners. At the same time, the published tasks are divided into on-campus tasks and off-campus tasks, providing multiple options;

[0065] The errand runner end includes receiving tasks, processing tasks, executing tasks, navigating, and confirming task completion. At the same time, some campuses prohibit outsiders from entering, and these off-campus tasks are completed by errand runners in a relay manner;

[0066] The administrator side includes managing users and errand runners, reviewing tasks, and viewing platform data statistics and reports.

[0067] The task service is responsible for the creation, status update, and history management of user tasks. After the user publishes a task, the task service will request a suitable runner from the intelligent matching service.

[0068] The intelligent matching service matches people based on task requirements, runner conditions, and real-time data, including:

[0069] Location-based matching algorithm: Matches runners based on the task posting location and their current location, giving priority to runners who are closer.

[0070] Matching task priority with errand runner ratings: If the task is urgent, consider increasing its priority and selecting the errand runner based on their ratings and past performance.

[0071] Prediction algorithm based on historical data: Use machine learning algorithms to analyze historical task data and the task completion status of runners to predict the most suitable runners;

[0072] Multi-objective optimization algorithm: Combines factors such as the runner's location, time, task type, and completion speed to conduct a comprehensive evaluation and select the best runner;

[0073] The comprehensive algorithm formula is as follows: based on the geographical location of tasks and runners, the distance (d_i) between the task location and the runner's current location is calculated; the smaller the distance, the higher the priority for selecting the runner;

[0074] [d_i=\sqrt{(x_{\text{task}}-x_{\text{courier}i})^2+(y{\text{task}}-y_{\text{courier}_i})^2}]

[0075] Here, (d_i) represents the Euclidean distance between the task posting location and the errand runner's current location;

[0076] If the task is urgent, we adjust the matching weight based on the task priority and the runner's score. For urgent tasks, we can assign a priority adjustment coefficient (w_{\text{task}}) to increase the priority of task matching.

[0077] For urgent tasks, the task priority adjustment factor (w_{\text{task}}=\alpha) (e.g. (\alpha=2));

[0078] For non-urgent tasks, the task priority adjustment coefficient (w_{\text{task}}=1);

[0079] The runner's rating (R_{\text{courier}i}) and historical performance (H{\text{courier}_i}) will be combined to form a comprehensive score:

[0080] [S_{\text{courier}i}=\beta\cdot R{\text{courier}i}+\gamma\cdot H{\text{courier}_i}]

[0081] (R_{\text{courier}_i}) is the courier score;

[0082] (H_{\text{courier}_i}) is the courier's historical performance (e.g., task completion rate);

[0083] (\beta,\gamma) are weight coefficients used to balance the runner's rating and historical performance;

[0084] Use machine learning algorithms to analyze historical task data to predict which runner is best suited for a task. This prediction can be viewed as a comprehensive matching score (P_{\text{courier}_i}), which can be calculated based on historical data and task characteristics.

[0085] P_{\text{courier}_i}=f(\text{historical data},\text{task characteristics},\text{courier characteristics})

[0086] A multi-objective optimization is performed by comprehensively considering factors such as the runner's location, time, task type, task priority, runner's rating, and historical data. The comprehensive score (M_{\text{courier}_i}) can be calculated using the following formula:

[0087] [M_{\text{courier}_i}=w{\text{task}}\cdot\frac{1}{d_i}+\lambda_1\cdot S_{\text{courier}_i}+\lambda_2\cdot P_{\text{courier}_i}]

[0088] in:

[0089] (\frac{1}{d_i}) is the inverse of the position, indicating that the closer the distance, the higher the score;

[0090] (w_{\text{task}}) is the task priority adjustment coefficient (taking into account the urgency of the task);

[0091] (S_{\text{courier}_i}) is the comprehensive score of the courier (score + historical performance);

[0092] (P_{\text{courier}_i}) is the prediction score based on historical data;

[0093] (\lambda_1) and (\lambda_2) are weight coefficients used to balance the influence of various factors;

[0094] All couriers are sorted by their comprehensive scores (M_{\text{courier}_i}) and the courier with the highest score is selected as the executor of the task.

[0095] The dispatching service ensures the rationality of the errand runner's tasks when he accepts them, and prevents the errand runner from taking too many orders or being located too far away; the dispatching service can also dynamically adjust task allocation to handle situations where the errand runner cannot complete the task in time due to weather or traffic jams; the map and navigation service is a map service integrated with map APIs such as Amap, Baidu or Google Maps, providing accurate location tracking, route planning and navigation functions to help errand runners complete tasks more efficiently; the payment and settlement service is a module responsible for payment and settlement after the task is completed, and supports payment interfaces such as Alipay and WeChat Pay; the settlement system will also settle commissions based on the performance of the errand runner.

[0096] 1. Calculation of errand runner commission

[0097] Commission calculations are typically based on the difficulty of the task, the runner's performance, and the timeliness of task completion. Given a base commission (C_{base}), based on the task complexity (T_c), timeliness (T_t), and the runner's evaluation (R), commission can be calculated using the following formula: [C_{total}=C_{base}\times(1+\alpha T_c+\beta T_t+\gamma R)], where (\alpha,\beta,\gamma) are adjustment factors, (T_c) is the complexity of the task (e.g., long or multi-site tasks), (T_t) is the timeliness of the task (whether the task was completed on time), and (R) is the runner's evaluation score (based on ratings or historical performance).

[0098] 2. Payment settlement formula

[0099] During payment settlement, assuming that the runner incurs a total fee (F) after completing the task, and the platform charges a certain percentage of service fee (p_{service}), the commission the runner should receive is: [P_{payout}=F\times(1-p_{service})], where (P_{payout}) is the runner's final payment amount, (F) is the total fee after the runner completes the task, and (p_{service}) is the platform service fee ratio.

[0100] The data layer includes:

[0101] Relational database: stores user information, errand runner information, task records, payment information, etc.

[0102] NoSQL database: used to store real-time information about tasks, dynamic data about errand runners, etc.

[0103] Caching: Use caching technologies such as Redis to store frequently queried data, such as task status, runner location, etc.

[0104] The infrastructure layer does include:

[0105] Cloud platform: provides elastic computing, storage and other resources to ensure high availability of the platform;

[0106] Message queues: such as Kafka or RabbitMQ, to ensure reliable delivery of tasks and notifications;

[0107] Containerization and microservices: Implementing microservice architecture through Docker and Kubernetes improves system scalability and operation and maintenance efficiency;

[0108] Logging and monitoring: Use logging and monitoring tools to monitor system status and troubleshoot problems.

[0109] Example 2:

[0110] This embodiment is based on the first embodiment:

[0111] 1. Collection and integration of supply information: The service team first collects product information (including text, photos, etc.) from various supply sources inside and outside the school and publishes it on the website for students to choose.

[0112] 2. Student confirms order: Students confirm orders and can purchase for themselves or others.

[0113] 3. Data processing: The service center will regularly collect statistics on various information and send the order information of each dormitory building to the relevant person in charge in advance.

[0114] 4. Pickup, diversion and delivery: The service center will designate personnel to pick up the goods, gather them at the station, and hand them over to the person in charge of each dormitory for diversion and delivery.

[0115] 5. Verify account information: The financial manager will calculate the financial information on a daily basis and prepare weekly statistics every Sunday evening.

[0116] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A campus errand platform architecture based on intelligent matching algorithm, characterized by: include: Front-end layer, application layer, data layer and infrastructure layer; The front-end layer includes the user side, the errand runner side, and the administrator side. Users interact through mobile applications or web terminals to complete operations such as task publishing, order acceptance, payment, and task status checking. The application layer includes task services, intelligent matching services, scheduling services, map and navigation services, and payment and settlement services. It uses algorithms to automatically match appropriate runners for each task, improving task allocation efficiency and optimizing platform resources. The data layer includes relational databases and NoSQL databases, which are responsible for the platform's data storage and management. The selection of databases and caches should be optimized according to the system's performance requirements; The infrastructure layer includes cloud platform, message queue, containerization and microservices, log and monitoring. The infrastructure layer ensures the availability and stability of the system and adopts cloud platform, containerized deployment and other technologies.

2. The campus errand running platform architecture based on the intelligent matching algorithm according to claim 1 is characterized by: The user terminal can publish tasks, query task status, make payments, and evaluate errand runners. At the same time, the published tasks are divided into on-campus tasks and off-campus tasks, providing multiple options; The errand runner end includes receiving tasks, processing tasks, executing tasks, navigating, and confirming task completion. At the same time, some campuses prohibit outsiders from entering, and these off-campus tasks are completed by errand runners in a relay manner; The administrator side includes managing users and errand runners, reviewing tasks, and viewing platform data statistics and reports.

3. The campus errand running platform architecture based on the intelligent matching algorithm according to claim 1 is characterized by: The task service is responsible for the creation, status update, and history management of user tasks. After the user publishes a task, the task service will request a suitable runner from the intelligent matching service.

4. The campus errand running platform architecture based on the intelligent matching algorithm according to claim 1 is characterized by: The intelligent matching service matches people based on task requirements, runner conditions, and real-time data, including: Location-based matching algorithm: Matches runners based on the task posting location and their current location, giving priority to runners who are closer. Matching task priority with errand runner ratings: If the task is urgent, consider increasing its priority and selecting the errand runner based on their ratings and past performance. Prediction algorithm based on historical data: Use machine learning algorithms to analyze historical task data and the task completion status of runners to predict the most suitable runners; Multi-objective optimization algorithm: Combines factors such as the runner's location, time, task type, and completion speed to conduct a comprehensive evaluation and select the best runner; The comprehensive algorithm formula is as follows: based on the geographical location of tasks and runners, the distance (d_i) between the task location and the runner's current location is calculated; the smaller the distance, the higher the priority for selecting the runner; [d_i=\sqrt{(x_{\text{task}}-x_{\text{courier}i})^2+(y{\text{task}}-y_{\text{courier}_i})^2}] Here, (d_i) represents the Euclidean distance between the task posting location and the errand runner's current location; If the task is urgent, we adjust the matching weight based on the task priority and the runner's score. For urgent tasks, we can assign a priority adjustment coefficient (w_{\text{task}}) to increase the priority of task matching. For urgent tasks, the task priority adjustment factor (w_{\text{task}}=\alpha) (e.g. (\alpha=2)); For non-urgent tasks, the task priority adjustment coefficient (w_{\text{task}}=1); The runner's rating (R_{\text{courier}i}) and historical performance (H{\text{courier}_i}) will be combined to form a comprehensive score: [S_{\text{courier}i}=\beta\cdot R{\text{courier}i}+\gamma\cdot H{\text{courier}_i}] (R_{\text{courier}_i}) is the courier score; (H_{\text{courier}_i}) is the courier's historical performance (e.g., task completion rate); (\beta,\gamma) are weight coefficients used to balance the runner's rating and historical performance; Use machine learning algorithms to analyze historical task data to predict which runner is best suited for a task. This prediction can be viewed as a comprehensive matching score (P_{\text{courier}_i}), which can be calculated based on historical data and task characteristics. P_{\text{courier}_i}=f(\text{historical data},\text{task characteristics},\text{courier characteristics}) A multi-objective optimization is performed by comprehensively considering factors such as the runner's location, time, task type, task priority, runner's rating, and historical data. The comprehensive score (M_{\text{courier}_i}) can be calculated using the following formula: [M_{\text{courier}_i}=w{\text{task}}\cdot\frac{1}{d_i}+\lambda_1\cdot S_{\text{courier}_i}+\lambda_2\cdot P_{\text{courier}_i}] in: (\frac{1}{d_i}) is the inverse of the position, indicating that the closer the distance, the higher the score; (w_{\text{task}}) is the task priority adjustment coefficient (taking into account the urgency of the task); (S_{\text{courier}_i}) is the comprehensive score of the courier (score + historical performance); (P_{\text{courier}_i}) is the prediction score based on historical data; (\lambda_1) and (\lambda_2) are weight coefficients used to balance the influence of various factors; All couriers are sorted by their comprehensive scores (M_{\text{courier}_i}) and the courier with the highest score is selected as the executor of the task.

5. The campus errand running platform architecture based on intelligent matching algorithm according to claim 1 is characterized by: The dispatching service ensures the rationality of the errand runner's tasks when he accepts them, and prevents the errand runner from taking too many orders or being located too far away; the dispatching service can also dynamically adjust task allocation to handle situations where the errand runner cannot complete the task in time due to weather or traffic jams; the map and navigation service is a map service integrated with map APIs such as Amap, Baidu or Google Maps, providing accurate location tracking, route planning and navigation functions to help errand runners complete tasks more efficiently; the payment and settlement service is a module responsible for payment and settlement after the task is completed, and supports payment interfaces such as Alipay and WeChat Pay; the settlement system will also settle commissions based on the performance of the errand runner.

6. The campus errand running platform architecture based on intelligent matching algorithm according to claim 1 is characterized by: The data layer includes: Relational database: stores user information, errand runner information, task records, payment information, etc. NoSQL database: used to store real-time information about tasks, dynamic data about errand runners, etc. Caching: Use caching technologies such as Redis to store frequently queried data, such as task status, runner location, etc.

7. The campus errand running platform architecture based on intelligent matching algorithm according to claim 1 is characterized by: The infrastructure layer does include: Cloud platform: provides elastic computing, storage and other resources to ensure high availability of the platform; Message queues: such as Kafka or RabbitMQ, to ensure reliable delivery of tasks and notifications; Containerization and microservices: Implementing microservice architecture through Docker and Kubernetes improves system scalability and operation and maintenance efficiency; Logging and monitoring: Use logging and monitoring tools to monitor system status and troubleshoot problems.

Images