An enterprise logistics intelligent operation and maintenance platform based on internet of things and mobile internet

Abstract

The application relates to the technical field of enterprise logistics operation and maintenance, and discloses an enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet, which comprises the following modules: a basic data modeling module, which is used for acquiring multi-source data and establishing a basic data system; a demand prediction generation module, which is used for calculating future demand, safety stock and ordering points and generating candidate work orders; a task urgency evaluation module, which is used for calculating unified urgency and merging tasks to generate a task list; a personnel task matching module, which is used for generating an employee task sequence and a work path; a mobile execution control module, which is used for executing tasks on a mobile terminal and generating abnormal work orders; a verification and quality inspection reconciliation module, which is used for verifying task authenticity and generating an inventory difference analysis work order; and a parameter optimization update module, which is used for updating inventory prediction parameters, urgency weights and comprehensive cost weights. The application realizes the intelligentization and high efficiency of task scheduling, execution control and abnormal traceability in enterprise logistics operation and maintenance.

Description

Technical Field

[0001] This invention belongs to the field of enterprise logistics operation and maintenance technology, specifically relating to an intelligent enterprise logistics operation and maintenance platform based on the Internet of Things and mobile Internet. Background Technology

[0002] As enterprises continue to expand, logistics management plays an increasingly important role in enterprise operations and maintenance. Traditional logistics management models typically rely on manual recording and scheduling, which not only suffers from data lag and incompleteness but also easily leads to information silos in areas such as inventory management, equipment maintenance, and task allocation, resulting in low overall operational efficiency. Especially in inventory management, the lack of dynamic forecasting of factors such as real-time inventory, historical demand fluctuations, and replenishment lead times often results in inventory backlogs or material shortages, severely impacting the continuity of enterprise operations. In equipment maintenance, traditional passive maintenance methods struggle to detect potential faults in a timely manner, increasing the risk of equipment downtime and maintenance costs. Regarding personnel scheduling, the lack of detailed consideration of personnel location, remaining workload, and skill suitability often leads to task allocation relying on manual experience, easily causing uneven resource allocation or redundant scheduling, reducing overall execution efficiency.

[0003] In recent years, the Internet of Things (IoT) and mobile internet technologies have been gradually introduced into enterprise logistics management, providing new technical means for data collection, real-time monitoring, and mobile execution. However, existing systems generally suffer from problems such as insufficient data utilization, simplistic task scheduling logic, and a lack of closed-loop control in the execution process. For example, in urgency assessment, they often rely solely on time factors, failing to comprehensively consider the task's impact level and risk factors; in path planning, they typically ignore the time window constraints for task execution, leading to some tasks not being completed on time; and in execution monitoring, the lack of anomaly handling and tracing mechanisms makes it difficult to verify task authenticity and detect inventory discrepancies in a timely manner. Summary of the Invention

[0004] This invention provides an intelligent enterprise logistics operation and maintenance platform based on the Internet of Things and mobile Internet, which solves the technical problems of inaccurate inventory forecasting, lack of preventive equipment maintenance, low task allocation efficiency, and lack of closed-loop control in the execution process in related technologies.

[0005] This invention provides an intelligent enterprise logistics operation and maintenance platform based on the Internet of Things and mobile Internet, comprising:

[0006] The basic data modeling module is used to acquire real-time inventory data, equipment status data, personnel data, and task execution records, establish a basic data system for enterprise logistics scenarios, and generate form data and personnel trajectory data based on real-time inventory data and personnel data.

[0007] The demand forecasting generation module is used to calculate future demand, safety stock, and reorder point based on historical demand data, real-time inventory data, and equipment status data, and to generate candidate work orders.

[0008] The task urgency assessment module is used to calculate a uniform urgency level based on candidate work orders and form data, and to merge tasks to generate a task list;

[0009] The personnel task matching module is used to calculate the overall cost and generate the task sequence and work path for each employee based on the task list, personnel data and personnel trajectory data.

[0010] The mobile execution control module is used to execute tasks on mobile devices according to task sequences, complete check-in, execute according to standard operating procedures and update existing inventory in real time, record anomalies and generate anomaly work orders, and inherit area identifiers, personnel trajectory data and on-site data;

[0011] The verification and quality inspection reconciliation module is used to verify the authenticity of tasks based on task execution records and perform quality inspection reconciliation. It compares the actual inventory with the existing inventory and generates inventory difference analysis work orders, and associates regional identifiers, material identifiers and personnel trajectory data.

[0012] The parameter optimization and update module is used to update the inventory forecast parameters, urgency weights, and overall cost weights based on completed task data, historical execution data, and inventory forecast data.

[0013] Furthermore, based on historical demand data, real-time inventory data, and equipment status data, future demand, safety stock, and reorder point are calculated to generate candidate work orders, including:

[0014] Step 11: Calculate future demand using exponential smoothing based on historical demand data, and update the daily average demand and demand standard deviation by setting a rolling window, and truncate outliers in the window.

[0015] Step 12: Obtain the safety stock by multiplying the service level quantile by the square root of the demand standard deviation and the replenishment lead time; and determine the reorder point by adding the safety stock to the product of the average daily demand and the replenishment lead time.

[0016] Step 13: When the sum of existing inventory and in-transit inventory minus allocated inventory is less than or equal to the reorder point, a replenishment candidate work order is generated. The economic order quantity is compared with the result of the reorder point plus safety stock minus existing inventory. The larger of the two values ​​is taken as the suggested replenishment quantity. The suggested replenishment quantity is rounded up to an integer multiple of the packaging specification.

[0017] Step 14: Based on the running time, number of operations and alarm frequency in the equipment status data, determine the equipment that meets the corresponding threshold and generate a candidate work order for preventive maintenance;

[0018] Step 15: Merge replenishment candidate work orders and preventive maintenance candidate work orders into a unified candidate work order set. Merge replenishment candidate work orders with the same area identifier, the same time window, and the same material identifier, and use the sum of the suggested replenishment quantities before merging as the suggested replenishment quantity after merging.

[0019] Furthermore, the generation of a unified urgency level includes:

[0020] Step 21: Standardize the candidate work orders and form data into a set of task records. Each task record includes a region identifier, work interest point, service level time limit, impact level, risk item, estimated working hours, and time window. Among them, the task record for replenishment tasks also includes existing inventory, safety stock, and reorder point fields.

[0021] Step 22: For non-replenishment tasks, the ratio of the difference between the service level time limit and the consumed time to the service level time limit is used as the remaining service level ratio; the impact level is normalized, and the impact level and risk item are weighted and summed using preset weight parameters to obtain the unified urgency of non-replenishment tasks.

[0022] Step 23: For replenishment tasks, the ratio of the difference between the existing inventory and the reorder point to the safety stock is determined as the inventory health; the difference between one and the inventory health and then divided by two is used as the unified urgency of the replenishment task.

[0023] Furthermore, the generation of risk items in step 22 includes: taking the ratio of the number of historical abnormal tasks to the total number of historical tasks as the historical abnormality rate; taking the ratio of the number of device alarms to the preset alarm threshold as the device risk; and combining the risk level of the area where the task is located, weighted summing and normalizing to obtain the risk item.

[0024] Furthermore, the tasks are merged to generate a task list, including:

[0025] Step 31: Identify tasks in the task record set that have the same area identifier and the same estimated working hours within the same time window that are less than or equal to a preset working hour threshold, and the distance between the task interest points is less than or equal to a preset proximity threshold.

[0026] Step 32: Merge the identified tasks to form a combined list, and set the uniform urgency of the combined list to the maximum uniform urgency of the merged tasks.

[0027] Step 33: Combine the tasks with the unified urgency calculation and the merged list to generate a task list with unified urgency, and bind a task identifier, unified urgency, region identifier, task interest point, time window and task type to each record.

[0028] Furthermore, based on the task list, personnel data, and personnel trajectory data, the overall cost is calculated, and a task sequence and work path for each employee are generated, including:

[0029] Step 41: Use the Euclidean distance between the current position and the position of the task interest point in the personnel trajectory data as the spatial distance, normalize the estimated working hours, spatial distance and remaining working capacity in the personnel data, and perform a weighted summation to obtain the comprehensive cost.

[0030] Step 42: Calculate the ratio of the intersection of the skill set in the personnel data and the skill set required for the task to the number of skills required for the task to obtain the skill matching degree, and filter to obtain a set of candidate tasks; calculate the comprehensive cost of the candidate tasks, and sort them in ascending order of comprehensive cost to obtain the task sequence of the employees;

[0031] Step 43: Based on the employee's task sequence and current position in the personnel trajectory data, a heuristic path search algorithm is used to generate the work path. When there are multiple feasible paths, the one with the shortest total travel distance is selected as the work path.

[0032] Furthermore, the inventory before the update is added together with the inventory change caused by the task to obtain the updated inventory. When the inventory change exceeds the preset change threshold, the operation steps fail to execute, or the data is missing, the system records the anomaly and automatically generates an abnormal work order.

[0033] Furthermore, the authenticity of the task execution is determined by comparing the executor, execution time, and execution location in the task execution record with the corresponding information in the task plan table; and it is checked whether the task execution record contains the required data in the standard operating procedure.

[0034] Furthermore, extract the existing inventory from the physical inventory data and the inventory data, and calculate the difference between the two. When the absolute value of the difference exceeds the preset difference threshold, generate an inventory difference analysis work order, and record the task identifier, region identifier, material identifier, difference amount and timestamp in the inventory difference analysis work order.

[0035] Furthermore, based on completed task data, historical execution data, and inventory forecast data, the inventory forecast parameters, urgency weight, and overall cost weight are updated, including:

[0036] Step 51: Extract the latest demand and execution cycle based on the completed task data and historical execution data, and adjust the inventory forecast parameters in combination with inventory forecast data. The inventory forecast parameters include future demand, safety stock and reorder point; wherein, future demand and safety stock are updated using an exponential smoothing method.

[0037] Step 52: Calculate the ratio of actual task time to planned task time to obtain the time consumption deviation; calculate the proportion of stockouts to the total number of tasks to obtain the stockout deviation; construct a deviation function based on the weighted average of the time consumption deviation and the stockout deviation; normalize the deviation function by adjusting the adjustment coefficient and correct the urgency weight.

[0038] Step 53: When the difference between the comprehensive cost and the preset target cost exceeds the preset difference, the comprehensive cost weight is corrected using the gradient descent method.

[0039] The beneficial effects of this invention are as follows: This invention achieves integrated data management for enterprise logistics operations by unifying the modeling of inventory, equipment, personnel, and task data. Specifically, the demand forecasting module dynamically calculates future demand, safety stock, and reorder points based on historical demand and inventory status, thereby avoiding inventory backlog or shortages; the task urgency assessment and merging mechanism quantitatively sorts different types of tasks within a unified framework, improving the rationality and comparability of scheduling; the personnel-task matching module performs comprehensive cost calculations based on spatial distance, working hours, and skill matching, and introduces time window constraints in path planning to ensure efficient and timely task allocation; the mobile execution control module updates inventory in real time and generates abnormal work orders during task execution, achieving traceability of task execution; the verification, quality inspection, and reconciliation module ensures execution compliance and data accuracy through task authenticity verification and inventory discrepancy analysis; and the parameter optimization and update module forms a closed-loop learning mechanism, continuously correcting forecast parameters and weights to improve the platform's adaptability and overall operational efficiency. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of a module of an enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet according to the present invention. Detailed Implementation

[0041] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0042] like Figure 1 As shown, an intelligent enterprise logistics operation and maintenance platform based on the Internet of Things and mobile Internet includes:

[0043] Basic data modeling module 1 is used to acquire real-time inventory data, equipment status data, personnel data and task execution records, establish a basic data system for enterprise logistics scenarios, and generate form data and personnel trajectory data based on real-time inventory data and personnel data.

[0044] The demand forecasting generation module 2 is used to calculate future demand, safety stock, and reorder point based on historical demand data, real-time inventory data, and equipment status data, and to generate candidate work orders.

[0045] Task urgency assessment module 3 is used to calculate a uniform urgency based on candidate work orders and form data, and to merge tasks to generate a task list;

[0046] The personnel task matching module 4 is used to calculate the comprehensive cost and generate the task sequence and work path for each employee based on the task list, personnel data and personnel trajectory data.

[0047] Mobile execution control module 5 is used to execute tasks on the mobile terminal according to the task sequence, complete check-in, execute according to standard operating procedures and update the existing inventory in real time, record anomalies and generate anomaly work orders, and inherit area identifiers, personnel trajectory data and on-site data;

[0048] The verification and quality inspection reconciliation module 6 is used to verify the authenticity of tasks based on task execution records and perform quality inspection reconciliation, compare the actual inventory with the existing inventory and generate inventory difference analysis work orders, and associate area identifiers, material identifiers and personnel trajectory data.

[0049] The parameter optimization and update module 7 is used to update the inventory forecast parameters, urgency weights, and comprehensive cost weights based on completed task data, historical execution data, and inventory forecast data.

[0050] In one embodiment of the present invention, the system first acquires multi-source data related to enterprise logistics operations. Real-time inventory data refers to information on the quantity, location, and status of materials in stock, collected in real time through a warehouse management system or IoT sensors. This data reflects the dynamic status of current material reserves. Equipment status data refers to equipment operating parameters collected by sensors, controllers, or equipment management systems, including runtime, number of operations, alarm frequency, and energy consumption information, used to characterize the health status and usage of the equipment. Personnel data refers to basic information and operational capabilities of personnel related to task execution, including personnel identification, region, skill set, and remaining workload, used to reflect personnel resource allocation. Task execution records refer to task completion information generated on mobile terminals or the execution system, including task identifier, executor identifier, task start and end times, execution location, and operation step data, used to record the actual execution process of the task.

[0051] Based on the aforementioned multi-source data, the system establishes a fundamental data system for enterprise logistics scenarios. This system links inventory, equipment, personnel, and task information through a unified data structure and identification method, enabling centralized management and processing of enterprise logistics elements and providing support for subsequent task scheduling and operational decisions. Furthermore, the system generates form data and personnel trajectory data based on real-time inventory and personnel data. Form data refers to a task requirement list triggered by real-time inventory data, containing information such as the materials to be replenished, their corresponding quantities, and their location, used to drive the generation of subsequent replenishment tasks. Personnel trajectory data refers to personnel movement path data obtained through the positioning module, combined with personnel identifiers and timestamps to form trajectory sequences, used to reflect the dynamic location and movement patterns of personnel in enterprise logistics scenarios.

[0052] In one embodiment of the present invention, based on historical demand data, real-time inventory data, and equipment status data, future demand, safety stock, and reorder point are calculated to generate candidate work orders, including:

[0053] Step 11: Calculate future demand using exponential smoothing based on historical demand data. This involves applying a decay coefficient to the historical demand sequence to enhance the weight of recent data, thereby improving the real-time performance and accuracy of the forecast. Furthermore, by setting a rolling window to update the daily average demand and demand standard deviation, outliers in the window are truncated to avoid interference from abnormal data with the forecast results.

[0054] Step 12: Obtain the safety stock by multiplying the service level quantile by the square root of the demand standard deviation and the replenishment lead time; and determine the reorder point by adding the safety stock to the product of the average daily demand and the replenishment lead time. The service level quantile reflects the acceptability of stockout risk, the demand standard deviation quantifies demand volatility, and the replenishment lead time represents the time required from submitting a replenishment request to the actual arrival of the goods. The reorder point indicates when replenishment should be triggered when the existing inventory falls to this value to ensure supply chain continuity.

[0055] Step 13: When the sum of existing inventory and in-transit inventory minus allocated inventory is less than or equal to the reorder point, a replenishment candidate work order is generated. The economic order quantity is compared with the result of the reorder point plus safety stock minus existing inventory. The larger of the two values ​​is used as the suggested replenishment quantity to balance ordering costs and inventory risks. The suggested replenishment quantity is rounded up to an integer multiple of the packaging specifications to ensure that the actual replenishment meets the constraints of logistics and packaging.

[0056] Step 14: Based on the runtime, number of operations, and alarm frequency in the equipment status data, determine the equipment that meets the corresponding thresholds and generate a candidate work order for preventive maintenance; by setting corresponding thresholds for runtime, number of operations, and alarm frequency, potentially malfunctioning equipment can be identified in advance, thereby achieving preventive maintenance.

[0057] Step 15: Merge replenishment candidate work orders and preventive maintenance candidate work orders into a unified candidate work order set. Merge replenishment candidate work orders with the same area identifier, the same time window, and the same material identifier, and use the sum of the suggested replenishment quantities before merging as the suggested replenishment quantities after merging, thereby improving work order processing efficiency and reducing the dispatch of duplicate tasks.

[0058] Through the above process, this embodiment can generate candidate work orders in both inventory and equipment operation and maintenance dimensions, ensuring the rationality and timeliness of replenishment triggering, while reducing equipment failure rate through preventive maintenance; work order merging improves the simplification of task scheduling and execution efficiency, thereby enhancing the intelligent level of enterprise logistics operation and maintenance and resource utilization.

[0059] In one embodiment of the present invention, the generation of a uniform urgency level includes:

[0060] Step 21: Standardize the candidate work orders and form data into a set of task records. Each task record includes a region identifier, work interest point, service level time limit, impact level, risk item, estimated working hours, and time window. Among them, the task record of replenishment tasks also includes existing inventory, safety stock, and reorder point fields to reflect the replenishment trigger conditions related to inventory status.

[0061] Step 22: For non-replenishment tasks, the ratio of the difference between the service level limit and the consumed time to the service level limit is used as the remaining service level ratio. This ratio reflects the relationship between the remaining available time of the task and the total time limit. The impact level is normalized, and the impact level minus the remaining service level ratio, the normalized impact level, and the risk item are weighted and summed using preset weight parameters to obtain the unified urgency of the non-replenishment task. 0 indicates sufficient inventory, 1 indicates insufficient inventory, and the median value linearly reflects inventory risk. The generation of the risk item includes: the ratio of the number of historical abnormal tasks to the total number of historical tasks as the historical abnormality rate; the ratio of the number of equipment alarms to the preset alarm threshold as the equipment risk; and the risk item is obtained by weighted summing and normalizing the results in conjunction with the risk level of the task's region.

[0062] Step 23: For replenishment tasks, the ratio of the difference between the existing inventory and the reorder point to the safety stock is determined as the inventory health. When the ratio is greater than one, it is taken as one; when the ratio is less than -1, it is taken as -1. The inventory health is subtracted from one and then divided by two to obtain the unified urgency of the replenishment task. This method can intuitively compare the inventory level with the reorder point and safety stock. When the inventory level is close to or lower than the reorder point, the unified urgency increases, thereby reflecting the urgency of the replenishment task.

[0063] Through the above process, this embodiment enables the urgency measurement of different types of tasks within a unified framework. For non-replenishment tasks, the unified urgency comprehensively considers time constraints, task impact level, and risk factors; for replenishment tasks, the unified urgency directly reflects the urgency of inventory status and replenishment needs. This method ensures the consistency and comparability of task urgency calculations, providing a reliable measurement basis for subsequent task sorting and scheduling.

[0064] In one embodiment of the present invention, merging tasks to generate a task list includes:

[0065] Step 31: Identify tasks in the task record set that have the same area identifier and the same estimated working hours within the same time window that are less than or equal to a preset working hour threshold, and the distance between the task interest points is less than or equal to a preset proximity threshold.

[0066] Step 32: Merge the identified tasks to form a combined list, and set the unified urgency of the combined list to the maximum value of the unified urgency of the merged tasks to ensure that the combined list inherits the priority of the most urgent task, thereby avoiding the dilution of task urgency. The determination of the same time window is that the overlap between the scheduled execution time intervals of the tasks is not less than the preset time window threshold.

[0067] Step 33: Combine the tasks with the unified urgency calculation and the merged list to generate a task list with unified urgency. Bind a task identifier, unified urgency, region identifier, operation interest point, time window and task type to each record. When outputting replenishment tasks, bind the existing inventory, safety stock and reorder point for traceability.

[0068] Through the above process, this embodiment can effectively reduce task dispersion and redundancy in the same region and time period while ensuring accurate representation of task urgency, and achieve reasonable aggregation and list output of tasks, thereby improving the execution efficiency and resource utilization of logistics operation and maintenance scheduling.

[0069] In one embodiment of the present invention, based on a task list, personnel data, and personnel trajectory data, a comprehensive cost is calculated and a task sequence and work path for each employee are generated, including:

[0070] Step 41: The Euclidean distance between the current position and the task interest point in the personnel trajectory data is used as the spatial distance. The estimated working hours, spatial distance and remaining work capacity in the personnel data are normalized and weighted and summed to obtain the comprehensive cost, which reflects the execution cost of a single task when it is assigned to an employee.

[0071] Step 42: Calculate the ratio of the intersection of the skill set in the personnel data and the skill set required for the task to the number of skills required for the task to obtain the skill matching degree, and filter to obtain a set of candidate tasks; calculate the comprehensive cost of the candidate tasks, and sort them in ascending order of comprehensive cost to obtain the task sequence of employees. When there are the same comprehensive cost, the task with the earlier planned deadline is given priority.

[0072] Step 43: Based on the employee's task sequence and current position in the personnel trajectory data, a heuristic path search algorithm is used to generate a work path. A time window constraint is introduced during the path planning process. The time window constraint stipulates that the estimated completion time of the task must not be later than the planned deadline. When multiple feasible paths exist, the one with the shortest total travel distance is selected as the work path. The goal of path search is to generate a reasonable work path for the employee while satisfying the time window constraint.

[0073] Through the above process, this embodiment can comprehensively consider spatial distance, working hours, personnel workload and skill adaptability to achieve reasonable and accurate task allocation; at the same time, by combining time window constraints and path optimization, it ensures the timeliness of task execution and the efficiency of path planning, thereby improving the overall intelligence level and execution efficiency of enterprise logistics operation and maintenance task scheduling.

[0074] In one embodiment of the present invention, the system executes tasks on a mobile device based on the task sequence. Specifically, employees first complete a check-in operation on the mobile device interface. The check-in is used to confirm the employee's identity and task acceptance status, and to record the current location in the personnel trajectory data. Subsequently, employees execute the tasks sequentially according to the standard operating procedures corresponding to the tasks. The standard operating procedures refer to standardized execution processes pre-set for different task types to ensure consistency and compliance of operations. The task types include replenishment, quality inspection, or maintenance. During task execution, the system collects and updates existing inventory data in real time to reflect changes in inventory quantity caused by task completion, thereby ensuring the dynamic accuracy of inventory data.

[0075] During execution, the system collects and updates existing inventory data in real time. Specifically, the inventory before the update is superimposed with the inventory change caused by the task to obtain the updated inventory, thereby dynamically reflecting the impact of the task on the inventory status. When the inventory change exceeds a preset threshold, or when an operation fails during execution, or when there is missing data on site, the system automatically determines that the task has malfunctioned and immediately records the anomaly information. Based on this anomaly information, the system generates a corresponding anomaly work order. When generating the anomaly work order, the system inherits the task's area identifier, personnel trajectory data, and on-site collected data, ensuring that the background, location, and personnel information involved in the anomaly are completely preserved.

[0076] In one embodiment of the present invention, the system verifies the authenticity of tasks and conducts quality inspection and reconciliation based on task execution records to ensure the effectiveness of task execution and the consistency of inventory data. Specifically, the system first compares the executor, execution time, and execution location in the task execution record with the corresponding information in the task plan table to determine whether the task was completed by the personnel specified in the plan at the predetermined time and location, thereby verifying the authenticity of the task. Furthermore, the system also checks whether the task execution record completely includes the mandatory data specified in the standard operating procedures. These standard operating procedures are operational requirements set to ensure the standardization and traceability of task execution, and the mandatory data includes confirmation of key steps and the collection results of necessary parameters. If the task execution record lacks the aforementioned mandatory data, the task is determined to be non-compliant with the authenticity verification requirements.

[0077] After verifying authenticity, the system proceeds to the quality inspection and reconciliation stage. This stage compares the actual inventory data with the existing inventory data, identifies discrepancies, and automatically generates an inventory discrepancy analysis work order. The actual inventory data refers to the inventory data obtained through physical inventory checks, accurately reflecting the quantity of materials on-site. When generating the inventory discrepancy analysis work order, it inherits the task's area identifier, material identifier, and personnel trajectory data to fully reflect the background environment of the discrepancy, the materials involved, and relevant responsible personnel information. Through this work order, managers can quickly locate the source of the problem and conduct targeted verification and correction.

[0078] Through the above process, this embodiment can effectively ensure the authenticity and compliance of task execution, and avoid system deviations caused by false execution or missing data; at the same time, by comparing the actual inventory with the existing inventory, the consistency of inventory data is verified, and a clear basis is provided for subsequent problem handling in the form of discrepancy work orders, thereby improving the reliability and management accuracy of enterprise logistics operation and maintenance.

[0079] In one embodiment of the present invention, the inventory forecasting parameters, urgency weight, and overall cost weight are updated based on completed task data, historical execution data, and inventory forecasting data, including:

[0080] Step 51: Extract the latest demand and execution cycle based on the completed task data and historical execution data, and correct the inventory forecast parameters in combination with inventory forecast data. The inventory forecast parameters include future demand, safety stock and reorder point; wherein, future demand and safety stock are updated using an exponential smoothing method, and the reorder point is calculated.

[0081] Step 52: Calculate the ratio of actual task time to planned time to obtain the time consumption deviation; calculate the proportion of stockouts to the total number of tasks to obtain the stockout deviation; construct a deviation function based on the weighted average of the time consumption deviation and the stockout deviation; normalize the deviation function by adjusting the adjustment coefficient; and correct the urgency weights using the normalized deviation function so that the urgency weights can be dynamically adjusted according to changes in task execution deviation, ensuring that the unified urgency calculation is more consistent with the actual situation of task execution; it should be noted that the urgency weights are weighting parameters used for the unified urgency calculation of non-replenishment tasks, corresponding to the three dimensions of remaining service level ratio, impact level, and risk item, respectively, to balance the relative importance of each factor in the urgency calculation; since the unified urgency of replenishment tasks is directly obtained through a linear mapping of inventory health, it does not involve urgency weights and does not need to be dynamically updated;

[0082] Step 53: When the difference between the comprehensive cost and the preset target cost exceeds the preset difference, the comprehensive cost weight is corrected using a gradient descent method, so that the comprehensive cost gradually approaches the target cost, thereby improving the rationality of resource allocation and the economy of overall operation and maintenance. The comprehensive cost weight is a weighted parameter used in the comprehensive cost calculation, considering estimated working hours, spatial distance, and remaining personnel working capacity.

[0083] Through the above process, this embodiment realizes the linkage update of inventory forecasting parameters, urgency weights, and comprehensive cost weights. Driven by task execution feedback, it can continuously optimize the forecasting and scheduling model, ensure the accuracy of inventory management and the rationality of task allocation, and thus improve the adaptability and optimization level of the enterprise's intelligent logistics operation and maintenance platform.

[0084] It should be noted that the interval and threshold sizes are set for ease of comparison. The size of the threshold depends on the amount of sample data and the base number set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless calculations, and the formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0085] The embodiments of the present invention have been described above, but the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of the present embodiments, all of which are within the protection scope of the present embodiments.

Claims

1. An intelligent enterprise logistics operation and maintenance platform based on the Internet of Things and mobile Internet, characterized in that, include: The basic data modeling module is used to acquire real-time inventory data, equipment status data, personnel data, and task execution records, establish a basic data system for enterprise logistics scenarios, and generate form data and personnel trajectory data based on real-time inventory data and personnel data. The demand forecasting generation module is used to calculate future demand, safety stock, and reorder point based on historical demand data, real-time inventory data, and equipment status data, and to generate candidate work orders. The task urgency assessment module is used to calculate a uniform urgency level based on candidate work orders and form data, and to merge tasks to generate a task list, including: Step 21: Standardize the candidate work orders and form data into a set of task records. Each task record includes a region identifier, work interest point, service level time limit, impact level, risk item, estimated working hours, and time window. Among them, the task record for replenishment tasks also includes existing inventory, safety stock, and reorder point fields. Step 22: For non-replenishment tasks, the ratio of the difference between the service level limit and the consumed time to the service level limit is used as the remaining service level ratio; the impact level is normalized, and the impact level minus the remaining service level ratio, the normalized impact level, and the risk item are weighted and summed using preset weight parameters to obtain the unified urgency of the non-replenishment task; the formula for calculating the unified urgency is: U represents the uniform urgency level. , and These represent the first, second, and third weight parameters in the preset weight parameters, respectively. Indicates the percentage of remaining service level. Indicates the impact level of non-replenishment tasks. Indicates risk items not related to replenishment tasks; Specifically, the historical anomaly rate is obtained by taking the ratio of the number of historical abnormal tasks to the total number of historical tasks; the equipment risk is taken by taking the ratio of the number of equipment alarms to the preset alarm threshold; and the risk item is obtained by taking a weighted sum and normalizing it in combination with the risk level of the area where the task is located. Step 23: For replenishment tasks, the ratio of the difference between the existing inventory and the reorder point to the safety stock is determined as the inventory health; the sum of one minus the inventory health and then divided by two is used as the unified urgency of the replenishment task. The personnel task matching module is used to calculate the overall cost and generate the task sequence and work path for each employee based on the task list, personnel data, and personnel trajectory data, including: Step 41: Use the Euclidean distance between the current position and the position of the task interest point in the personnel trajectory data as the spatial distance, normalize the estimated working hours, spatial distance and remaining working capacity in the personnel data, and perform a weighted summation to obtain the comprehensive cost. Step 42: Calculate the ratio of the intersection of the skill set in the personnel data and the skill set required for the task to the number of skills required for the task to obtain the skill matching degree, and filter to obtain a set of candidate tasks; calculate the comprehensive cost of the candidate tasks, and sort them in ascending order of comprehensive cost to obtain the task sequence of the employees; Step 43: Based on the current position in the employee's task sequence and personnel trajectory data, a heuristic path search algorithm is used to generate the work path. When there are multiple feasible paths, the one with the shortest total travel distance is selected as the work path. The mobile execution control module is used to execute tasks on mobile devices according to task sequences, complete check-in, execute according to standard operating procedures and update existing inventory in real time, record anomalies and generate anomaly work orders, and inherit area identifiers, personnel trajectory data and on-site data; The verification and quality inspection reconciliation module is used to verify the authenticity of tasks based on task execution records and perform quality inspection reconciliation. It compares the actual inventory with the existing inventory and generates inventory difference analysis work orders, and associates regional identifiers, material identifiers and personnel trajectory data. The parameter optimization and update module is used to update inventory forecasting parameters, urgency weights, and overall cost weights based on completed task data, historical execution data, and inventory forecasting data. This includes: Step 51: Extract the latest demand and execution cycle based on the completed task data and historical execution data, and adjust the inventory forecast parameters in combination with inventory forecast data. The inventory forecast parameters include future demand, safety stock and reorder point; wherein, future demand and safety stock are updated using an exponential smoothing method. Step 52: Calculate the ratio of actual task time to planned task time to obtain the time consumption deviation; calculate the proportion of stockouts to the total number of tasks to obtain the stockout deviation; construct a deviation function based on the weighted average of the time consumption deviation and the stockout deviation; normalize the deviation function by adjusting the adjustment coefficient and correct the urgency weight. Step 53: When the difference between the comprehensive cost and the preset target cost exceeds the preset difference, the comprehensive cost weight is corrected using the gradient descent method.

2. The enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet according to claim 1, characterized in that, Based on historical demand data, real-time inventory data, and equipment status data, calculate future demand, safety stock, and reorder point to generate candidate work orders, including: Step 11: Calculate future demand using exponential smoothing based on historical demand data, and update the daily average demand and demand standard deviation by setting a rolling window, and truncate outliers in the window. Step 12: Obtain the safety stock by multiplying the service level quantile by the square root of the demand standard deviation and the replenishment lead time; and determine the reorder point by adding the safety stock to the product of the average daily demand and the replenishment lead time. Step 13: When the sum of existing inventory and in-transit inventory minus allocated inventory is less than or equal to the reorder point, a replenishment candidate work order is generated. The economic order quantity is compared with the result of the reorder point plus safety stock minus existing inventory. The larger of the two values ​​is taken as the suggested replenishment quantity, which is then rounded up to an integer multiple of the packaging size. The formula for calculating the suggested replenishment quantity is as follows: Q represents the suggested replenishment quantity, max represents the maximum value function, E represents the economic order quantity, R represents the reorder point, and S represents the safety stock. Indicates current inventory; Step 14: Based on the running time, number of operations and alarm frequency in the equipment status data, determine the equipment that meets the corresponding threshold and generate a candidate work order for preventive maintenance; Step 15: Merge replenishment candidate work orders and preventive maintenance candidate work orders into a unified candidate work order set. Merge replenishment candidate work orders with the same area identifier, the same time window, and the same material identifier, and use the sum of the suggested replenishment quantities before merging as the suggested replenishment quantity after merging.

3. The enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet according to claim 1, characterized in that, The tasks are merged to generate a task list, including: Step 31: Identify tasks in the task record set that have the same area identifier and the same estimated working hours within the same time window that are less than or equal to a preset working hour threshold, and the distance between the task interest points is less than or equal to a preset proximity threshold. Step 32: Merge the identified tasks to form a combined list, and set the uniform urgency of the combined list to the maximum uniform urgency of the merged tasks. Step 33: Combine the tasks with the unified urgency calculation and the merged list to generate a task list with unified urgency, and bind a task identifier, unified urgency, region identifier, task interest point, time window and task type to each record.

4. The enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet according to claim 1, characterized in that, The updated inventory is obtained by adding the inventory before the update and the inventory change caused by the task. When the inventory change exceeds the preset change threshold, the operation steps fail, or the data is missing, the system records the anomaly and automatically generates an exception work order.

5. The enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet according to claim 1, characterized in that, The authenticity of the task execution is determined by comparing the executor, execution time, and execution location in the task execution record with the corresponding information in the task plan table; and the task execution record is checked to see if it contains the required data in the standard operating procedure.

6. The enterprise logistics intelligent operation and maintenance platform based on the Internet of Things and mobile Internet according to claim 1, characterized in that, Extract the actual inventory data and the existing inventory from the inventory data, and calculate the difference between the two. When the absolute value of the difference exceeds the preset difference threshold, generate an inventory difference analysis work order, and record the task identifier, region identifier, material identifier, difference amount and timestamp in the inventory difference analysis work order.

Images