Hospital logistics service resource integration and optimization method

Abstract

The invention discloses a hospital logistics service resource integration and optimization method, and particularly relates to the technical field of resource integration and optimization. The method comprises the following steps: acquiring service request data from a mobile terminal service platform, carrying out standardization processing on the service request data, establishing a multi-source service request feature analysis model, and carrying out differential analysis on data granularity, an emergency level marking mode and information completeness of service requests to construct a service request fusion model; the service request emergency level is re-evaluated in a unified manner through the fusion model, the service request and the service resource are dynamically matched in combination with the real-time load state of the service resource and the scheduling constraint condition, and the maintenance work order is automatically generated and distributed, so that the intelligent, real-time and refined optimization scheduling of the hospital logistics service resource is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of resource integration optimization, and more particularly to a hospital logistics service resource integration optimization method. BACKGROUND

[0002] Hospital logistics service management is an important part of the hospital operation system, and the operation efficiency of logistics service is directly related to the quality of medical service, the internal operation efficiency of the hospital and the satisfaction of patients.

[0003] In the traditional logistics service management mode commonly used in hospitals at present, there is a lack of effective data interconnection and intercommunication between different departments and service types, resulting in low response efficiency of logistics service, which cannot respond to the demands of medical staff and patients for logistics service in time, and restricts the fine operation of modern hospitals and the improvement of overall medical service efficiency. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a hospital logistics service resource integration optimization method to solve the problems in the above background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] A hospital logistics service resource integration optimization method, comprising the following steps:

[0007] S1: collecting service request data from a mobile terminal service platform, and performing standardized processing on the service request data to output a service request data set;

[0008] S2: analyzing the data granularity, emergency level marking method and information completeness difference of different service requests according to the service request data set, and outputting multi-source service request feature analysis data;

[0009] S3: based on the multi-source service request feature analysis data, establishing a service request fusion model, uniformly re-evaluating the emergency level of the service request, and outputting service request priority ranking data;

[0010] S4: based on the multi-source service request feature analysis data, using a multi-dimensional state feature correlation analysis method to calculate and verify the constraint conditions of the current load of the service resource, and outputting service resource real-time load state analysis data;

[0011] S5: according to the service request priority ranking data and the service resource real-time load state analysis data, using a multi-objective constraint optimization matching algorithm, establishing a dynamic correlation model between the service request features and the real-time state of the service resource, and outputting a service resource dynamic matching and scheduling scheme;

[0012] S6: Based on the service resource dynamic matching scheduling scheme, automatically generate and distribute maintenance work orders, and optimize the scheduling of logistics service resources in real time.

[0013] In a preferred embodiment, S1, specifically:

[0014] Collect service request data from the mobile terminal service platform; the service request data includes service request type, repair equipment name, repair equipment location, request submission time, and request emergency level;

[0015] Standardize the service request data and output a standardized service request data set.

[0016] In a preferred embodiment, the standardization process includes: uniformly encoding the repair equipment name according to the hospital's preset equipment classification standard; standardizing the repair equipment location based on the hospital's preset spatial location coding rules; converting the request submission time to a uniform time format; and re-labeling the request emergency level according to the hospital's unified emergency level evaluation standard.

[0017] In a preferred embodiment, S2, specifically:

[0018] Statistical data granularity of service request type in the standardized service request data set to obtain data granularity statistical results corresponding to each service request type;

[0019] Differential comparison of the emergency level marking method of the service request in the standardized service request data set to obtain the difference characteristics of the emergency level marking method of the service request;

[0020] Evaluate the information completeness of the service request submission time and the repair equipment location in the standardized service request data set to obtain the information completeness evaluation results of the service request submission time and the repair equipment location information;

[0021] Output multi-source service request feature analysis data including data granularity statistical results, emergency level marking method difference characteristics, and information completeness evaluation results.

[0022] In a preferred embodiment, S3, specifically:

[0023] Based on the multi-source service request feature analysis data, a service request fusion model is established, including:

[0024] Determine the granularity weight corresponding to each service request type according to the data granularity statistical results;

[0025] Determine the uniform marking rule of the service request emergency level according to the difference characteristics of the emergency level marking method;

[0026] setting an information completeness threshold of the service request according to the information completeness evaluation result;

[0027] According to the service request fusion model, the unified re-evaluation of the emergency level of each service request is carried out, including:

[0028] Adjusting the initial emergency level corresponding to the service request type by using the granularity weight;

[0029] According to the unified marking rule, the emergency levels of different sources are re-marked;

[0030] Lowering the emergency level of the service request that does not reach the information completeness threshold;

[0031] Outputting service request priority ranking data.

[0032] In a preferred embodiment, based on multi-source service request feature analysis data, a service request fusion model is established, specifically:

[0033] According to the data granularity statistical result, the granularity weight corresponding to each service request type is determined;

[0034] According to the difference characteristics of the emergency level marking mode, the unified marking rule of the emergency level of the service request is determined;

[0035] According to the information completeness evaluation result, the information completeness threshold of the service request is set.

[0036] In a preferred embodiment, according to the service request fusion model, the unified re-evaluation of the emergency level of each service request is carried out, specifically:

[0037] Adjusting the initial emergency level corresponding to the service request type by using the granularity weight;

[0038] According to the unified marking rule, the emergency levels of different sources are re-marked;

[0039] Lowering the emergency level of the service request that does not reach the information completeness threshold.

[0040] In a preferred embodiment, S4, specifically:

[0041] Collecting the task queue length, average task processing time and idle waiting time of each service resource in the current time window to form a service resource load original data matrix;

[0042] Performing normalization processing on the service resource load original data matrix to obtain a service resource load normalized index vector;

[0043] Multi-dimensional tensor splicing is performed on the service resource load normalized index vector and the multi-source service request feature analysis data to construct a multi-dimensional state feature tensor;

[0044] Based on the multi-dimensional state feature tensor, the load matching degree score between each service resource and the corresponding service request is calculated.

[0045] The load matching degree score is compared with the preset service resource capacity constraint, response time constraint and spatial accessibility constraint respectively, and service resource real-time load state analysis data is output.

[0046] In a preferred embodiment, S5, specifically:

[0047] Based on the service request priority ranking data and the service resource real-time load state analysis data, a mapping cost matrix of service requests and service resources is generated;

[0048] Based on the mapping cost matrix, a dynamic association model between service request features and service resource real-time state is established;

[0049] Based on the dynamic association model, a delay minimization objective function, a load balancing maximization objective function and a path distance minimization objective function are constructed to form a multi-objective optimization objective set;

[0050] The multi-objective optimization objective set is converted into a comprehensive evaluation function;

[0051] The comprehensive evaluation function is iteratively solved within the constraint set to obtain a service request and service resource matching decision variable matrix, and a service resource dynamic matching scheduling scheme is generated.

[0052] In a preferred embodiment, S6, specifically:

[0053] Based on the service resource dynamic matching scheduling scheme, a maintenance work order is automatically generated;

[0054] According to the service request type in the maintenance work order and the matched available service resource information, the maintenance work order is automatically assigned to the corresponding available service resource;

[0055] After the maintenance work order is assigned, according to the request submission time in the maintenance work order and the response time limit standard corresponding to the service request emergency level preset by the hospital, the maintenance work order is real-time scheduled to real-time optimize the scheduling of logistics service resources.

[0056] The technical effects and advantages of the hospital logistics service resource integration optimization method of the present application are:

[0057] By unified collection and standardized processing of mobile terminal service request, data format consistency and integrity are ensured; by differentiated analysis of granularity, emergency level and information completeness of standardized data, multi-dimensional feature support is provided for constructing fusion model based on multi-source features, unified re-evaluation of emergency level is provided, and objectification and dynamicization of service request priority are realized; combined with feature analysis results and historical capability indexes, service resource load and scheduling constraints are evaluated in real time, and resource availability monitoring precision is improved; based on priority sorting and real-time load state, request and available resources are dynamically matched, and resource allocation efficiency is optimized; by automatically generating and dispatching maintenance work orders, closed-loop scheduling process is improved, and request response timeliness and execution traceability are ensured. The response speed of logistics service, resource utilization and management transparency are improved, and the efficient and reliable logistics support demand of modern hospital is met. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 A hospital logistics service resource integration optimization method is given. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0060] EMBODIMENT

[0061] Figure 1 A hospital logistics service resource integration optimization method is given, which comprises the following steps:

[0062] S1: collecting service request data from a mobile terminal service platform, and performing standardized processing on the service request data to output a service request data set;

[0063] S2: according to the service request data set, analyzing the data granularity, emergency level marking method and information completeness difference of different service requests, and outputting multi-source service request feature analysis data;

[0064] S3: based on the multi-source service request feature analysis data, establishing a service request fusion model, uniformly re-evaluating the emergency level of the service request, and outputting service request priority sorting data;

[0065] S4: based on the multi-source service request feature analysis data, using a multi-dimensional state feature correlation analysis method to calculate and verify the constraint conditions of the current load of the service resource, and outputting service resource real-time load state analysis data;

[0066] S5: Based on the service request priority sorting data and the real-time load status analysis data of service resources, a multi-objective constraint optimization matching algorithm is used to establish a dynamic correlation model between service request characteristics and the real-time status of service resources, and output a dynamic matching and scheduling scheme for service resources.

[0067] S6: Based on the dynamic matching and scheduling scheme of service resources, it automatically generates and dispatches maintenance work orders and optimizes the scheduling of logistics service resources in real time.

[0068] S1: Collect service request data from the mobile service platform, standardize the service request data, and output a service request data set, including:

[0069] Collect service request data from mobile service platforms;

[0070] A mobile service platform refers to an application provided by the hospital for medical staff and those submitting service requests. Medical staff or those submitting service requests can use handheld mobile devices, such as smartphones or tablets, within the hospital's local area network or dedicated wireless network to submit service requests to the logistics service system. Service request data includes the service request type, the name of the device being repaired, the location of the device, the submission time, and the urgency level. For example, a doctor submits a service request to the logistics department through the hospital's mobile service platform: the service request type is "plumbing and electrical repair," the device name is "leaky water pipe," the location is "Room 301, East Side, Third Floor, Outpatient Building," the submission time is "10:45 AM, July 20, 2025," and the urgency level is "extremely urgent."

[0071] The service request data is normalized and output as a normalized service request data set.

[0072] Normalization is performed to standardize the format of service request data from different submitters. The service request data set is a unified formatted data set formed after normalization of request data submitted by all mobile service platforms. Normalization includes:

[0073] The names of the equipment reported for repair are uniformly coded according to the hospital's pre-defined equipment classification standards.

[0074] The equipment classification standard is a set of pre-established rules for classifying equipment within the hospital's logistics service system. This standard allows for the identification of various hospital devices using a unified coding system. For example, the hospital's equipment classification standard categorizes "central air conditioning indoor units" as a subcategory under the broader category of "HVAC equipment," coded as "NKT-001." Similarly, "medical ventilators" are categorized as a subcategory under the category of "medical equipment," coded as "YL-035." The coding method of the equipment classification standard is based on the hospital's pre-established coding system. Through unified coding, different expressions used by submitters in service request data can be standardized into unified equipment identifiers, ensuring a correspondence between the equipment names used by different submitters and the equipment identifiers used in system analysis.

[0075] The location of the reported equipment is described in a standardized manner based on the hospital's pre-defined spatial location coding rules;

[0076] Spatial location coding rules are pre-defined, unified identifiers for hospital spatial locations within the hospital's logistics service management system. These rules assign a unique code to each area and room within the hospital building, accurately pinpointing the location of equipment involved in a service request. For example, the location "Room 301, East Side, 3rd Floor, Outpatient Building" in a submitted service request would be uniformly coded as "MZL-03D-301" in the spatial location coding rules. Here, "MZL" represents "Outpatient Building," "03D" represents "East Side, 3rd Floor," and "301" represents the room number. Spatial location coding rules standardize and unify different descriptions of the location of equipment requested for repair by different submitters, enabling the hospital's logistics service department to quickly locate the equipment.

[0077] Use a unified time format to convert the service request submission time;

[0078] A unified time format refers to the standardized way of representing time as defined by the hospital's logistics service system. For example, a format using year, month, day, hour, and minute as basic elements. For instance, "July 20, 2025, 10:45 AM" is uniformly converted to "2025-07-20 10:45". Using a unified time format ensures that service request submission times are accurately recorded and uniformly analyzed, facilitating the management of service request data.

[0079] The urgency level of the service request is re-marked according to the hospital's unified urgency level assessment standards;

[0080] The hospital's standardized emergency level assessment criteria are pre-determined emergency classification rules by the hospital's logistics service department. These rules categorize service requests of varying urgency into multiple levels, such as "Very Urgent," "Urgent," "General," and "Low Urgent," each with a defined response timeframe. For example, a request initially marked as "Very Urgent" is reclassified as Level "I," requiring a response time of no more than 15 minutes; a request marked as "Urgent" is reclassified as Level "II," with a response time of no more than 30 minutes; a "General" request is marked as Level "III," with a response time of no more than 2 hours; and a request marked as "Low Urgent" is reclassified as Level "IV," with a response time of no more than 4 hours. This standardized reclassification of service request urgency levels unifies potential discrepancies between different submitters and ensures timely request processing.

[0081] S2: Based on the service request dataset, analyze the differences in data granularity, urgency level marking methods, and information completeness among different service requests, and output multi-source service request feature analysis data, including:

[0082] The data granularity of service request types in the standardized service request dataset is statistically analyzed to obtain the data granularity statistical results corresponding to each service request type;

[0083] The data granularity statistical results are derived by the hospital's logistics service department based on the quantity, specificity, and detail of each service request type in the standardized service request dataset, through statistical analysis and quantification. The data granularity statistical results reflect the distribution of service request types in terms of data scale, detail, and presentation. For example, from July 1st to July 31st, 2025, the hospital's logistics service department, through statistical analysis of the standardized service request dataset, obtained service request types including water and electricity repair, central air conditioning repair, elevator repair, and others. The statistical results show that the data granularity statistical results for the "water and electricity repair" request type are 1483, for the "central air conditioning repair" request type are 905, and for the "elevator repair and others" request type are 705. The statistical process records data according to the service request type tagged in each request in the standardized service request dataset, and the method for statistical data granularity follows the principles of service request type classification and quantity recording.

[0084] The emergency level marking methods of service requests in the standardized service request dataset are compared to identify differences, and the differences in the emergency level marking methods of service requests are obtained.

[0085] The differences in emergency level labeling methods are the results of a comparison of the different emergency level labeling methods used by the hospital's logistics service system to classify service requests in a standardized service request dataset. These differences highlight inconsistencies in how different service request submitters label emergency levels. By clarifying these differences, the system can reveal the variations in the definition of urgency levels before standardization, thus emphasizing the necessity of a unified labeling rule after standardization. For example, when a doctor submits a plumbing / electrical repair request, the initial emergency level is labeled "extremely urgent," while a nurse submitting the same type of plumbing / electrical repair request initially labels it only "moderate." Through comparison, the hospital's logistics service department found significant differences in the initial emergency level labeling based on subjective judgments by the requesters. By comparing the emergency level labeling methods of all requests in the standardized service request dataset from July 2025, the hospital's logistics service department identified a significant characteristic of these differences: for requests involving similar faults, equipment, and locations, the initial emergency level labeling by submitters differed by more than 30%. The differences highlight the necessity for the hospital's logistics service management system to re-standardize the marking of emergency levels.

[0086] The completeness of information regarding the service request submission time and the location of the reported equipment in the standardized service request dataset is evaluated to obtain the information completeness evaluation results for the service request submission time and the location of the reported equipment.

[0087] The information completeness assessment result is generated by the hospital's logistics service system based on each service request in the standardized service request dataset. This involves checking the completeness of the service request submission time and the location information of the equipment being repaired, thereby determining the degree of information completeness. The information completeness assessment of the service request submission time checks the accuracy of the time data in each request, based on the time format requirements preset by the hospital's logistics service system. The information completeness assessment of the location information of the equipment being repaired checks the accuracy of the location code and location description, as well as the completeness of the correspondence between the location data and the actual hospital building location. For example, if the hospital's logistics service department assesses a service request in the standardized service request dataset with a submission time of "2025-07-20 10:45" and a location code for the equipment being repaired of "MZL-03D-301", and the check confirms that both the submission time and location code information meet the hospital's unified standards, the information completeness assessment result is complete and valid. Using the above evaluation methods, the hospital's logistics service department conducted an information completeness assessment on all service request data during July 2025, obtaining an assessment result of 98.5% completeness of service request submission time information and 92.7% completeness of equipment location information.

[0088] The output includes multi-source service request feature analysis data, including data granularity statistical results, differences in emergency level labeling methods, and information completeness assessment results;

[0089] The multi-source service request feature analysis data is a data analysis result formed by the hospital logistics service system based on the statistical results of data granularity, the differences in emergency level marking methods, and the information completeness assessment results.

[0090] S3: Based on multi-source service request feature analysis data, establish a service request fusion model, uniformly re-evaluate the urgency level of service requests, and output service request priority ranking data, including:

[0091] Based on multi-source service request feature analysis data, a service request fusion model is established, including:

[0092] Determine the granularity weight for each service request type based on the data granularity statistics results;

[0093] Granularity weight is a weight value determined by the hospital's logistics service management system based on the data quantity, specificity, and frequency of each service request type in the granularity statistics. Granularity weight reflects the importance and processing priority of service request types; for example, service request types with a large number of data items typically correspond to higher granularity weights, while those with fewer data items correspond to lower granularity weights. For instance, statistics from July 2025 show that the granularity statistics for the "Water and Electricity Repair" service request type were 1483, "Central Air Conditioning Repair" had 905, and "Elevator and Others" had 705. The hospital's logistics service management system calculated the granularity weight for the "Water and Electricity Repair" service request type to be 0.48, for "Central Air Conditioning Repair" to be 0.29, and for "Central Air Conditioning Repair" to be 0.23. The granularity weight is determined by calculating the proportion of data quantity, i.e., the granularity weight of a service request type is the ratio obtained by dividing the number of service request type data by the total number of service request data items.

[0094] A unified marking rule for the emergency level of a service request is determined based on the differences in emergency level marking methods;

[0095] The unified labeling rules are established by the hospital's logistics service management system based on the analysis of differences in emergency level labeling methods. These rules eliminate discrepancies in emergency levels arising from subjective judgment by service request submitters, standardizing emergency levels across different initial labels. For example, statistics from July 2025 revealed that the initial emergency level labeling of service requests varied by as much as 30%, with similar requests labeled as "extremely urgent" by doctors and "general" by nurses. Based on this analysis, the hospital's logistics service department developed unified labeling rules. The unified labeling rules are defined as follows: When the proportion of a certain service request type initially labeled "extremely urgent" in historical data exceeds 20%, all requests of this type are uniformly labeled as Level "I," with a response time limit of 15 minutes or less; those exceeding 10% but less than 20% are uniformly labeled as Level "II," with a response time limit of 30 minutes or less; those exceeding 5% but less than 10% are uniformly labeled as Level "III," with a response time limit of 2 hours or less; and those below 5% are uniformly labeled as Level "IV," with a response time limit of 4 hours or less. Unified labeling rules enable consistent adjustments to initial labeling differences, ensuring a uniform and objective expression of service request urgency levels.

[0096] Set the information integrity threshold for service requests based on the information completeness assessment results;

[0097] The information integrity threshold is a standard set by the hospital's logistics service management system to judge the completeness of information based on the evaluation results of the submission time information and the location information of the equipment being repaired. The information integrity threshold is used to determine whether the submission time information and equipment location code information in a service request are sufficiently complete to support the timely response of the hospital's logistics service department. The information integrity threshold is defined as follows: a request with both complete and valid submission time information and equipment location code information is considered to have reached the threshold; a request with either incomplete information is considered to have failed to reach the threshold. For example, in all requests in July 2025, the information completeness evaluation showed that 98.5% of requests had complete submission times and 92.7% had complete equipment location information. The hospital's logistics service management system stipulates that both the equipment location information and the submission time information in a service request must be complete and consistent with the hospital's standards to reach the information integrity threshold; otherwise, it is considered to have failed to reach the threshold.

[0098] Based on the service request fusion model, a unified reassessment of the urgency level of each service request is performed, including:

[0099] Adjust the initial urgency level corresponding to the service request type using granular weights;

[0100] Adjusting the initial urgency level of service requests using granular weights means that the hospital's logistics service management system quantifies and adjusts the initial urgency level of each service request type according to granular weights, ensuring that service request types with higher data granularity correspond to higher priority levels. For example, a water and electricity repair service request with a granular weight of 0.5 and an initial level of "II" can be upgraded one level to "I" according to the weight adjustment rules, thus increasing its processing priority.

[0101] Re-label the emergency levels from different sources according to the unified labeling rules;

[0102] The hospital's logistics service management system re-marks service requests according to a unified marking rule to determine their urgency level. For example, a water and electricity repair request that was initially marked as "general" may be re-marked as "very urgent" if the proportion of historical data marked as "very urgent" exceeds 20% according to the rule.

[0103] Downgrade the urgency level of service requests that do not meet the information integrity threshold;

[0104] If the information in the request is incomplete, such as the device location only being marked as "outpatient building" and not meeting the information completeness threshold set by the hospital, the logistics service department will downgrade the request originally marked as "very urgent" to "urgent", thus reducing the priority of the service request response.

[0105] Output service request priority sorting data;

[0106] The service request priority ranking data is a final priority list of service requests formed after unified re-evaluation. The hospital's logistics service department arranges the service request response order according to the ranking data. The ranking data uses the urgency level, which has been uniformly adjusted by the fusion model, as the ranking standard and includes the ranking order of all requests.

[0107] S4: Based on multi-source service request feature analysis data, a multi-dimensional state feature correlation analysis method is used to calculate and verify the current load status of service resources, and output real-time load status analysis data of service resources, including:

[0108] Collect the task queue length, average task processing time and idle waiting time of each service resource within the current time window to form the original data matrix of service resource load.

[0109] The collection of task queue length, average task processing time, and idle waiting time for each service resource within the current time window refers to the information recorded by the hospital logistics service resource management system on the real-time operational status of each service resource currently used by the hospital's logistics department to meet service requests. Service resources refer to various logistics personnel, equipment, or other logistical support tools capable of responding to and processing service requests such as water and electricity maintenance, central air conditioning maintenance, elevator maintenance, and others within the hospital. The current time window is a pre-defined data statistics period defined by the hospital logistics service management system. Task queue length refers to the number of tasks waiting to be processed or being processed for a particular service resource within the current statistical period. Average task processing time is the average processing time consumed for each service task completed by each service resource within the current time window, as statistically analyzed by the hospital logistics service resource management system. Idle waiting time refers to the total idle waiting time accumulated for each service resource in an inactive state within the current statistical period. The task queue length, average task processing time, and idle waiting time are arranged in an orderly manner in the form of a matrix to form the original data matrix of service resource load. The rows of the matrix represent each type of service resource, and the columns represent the above three status indicators for each type of service resource.

[0110] The original data matrix of service resource load is normalized to obtain the normalized index vector of service resource load.

[0111] Normalization of the raw data matrix of service resource load refers to transforming all indicators in the raw data matrix of service resource load into a range of 0 to 1, so as to eliminate the difference in dimensionality between different service resource indicators and ensure that the indicators are comparable.

[0112] The service resource load normalization index vector is concatenated with the multi-source service request feature analysis data to form a multi-dimensional tensor, thus constructing a multi-dimensional state feature tensor.

[0113] The service resource load normalization index vector and multi-source service request feature analysis data—namely, data granularity statistical results, differences in emergency level labeling methods, and information completeness assessment results—are integrated in the form of a tensor to form a multidimensional state feature tensor. The method for constructing the multidimensional state feature tensor is to use the service resource normalization index vector as one dimension, and the feature analysis data such as data granularity, differences in emergency level labeling, and information completeness assessment as the other dimensions.

[0114] Based on the multidimensional state feature tensor, the load matching score between each service resource and the corresponding service request is calculated.

[0115] The multidimensional state feature tensor load matching score is calculated by matching each service resource in the tensor with each service request to quantify the degree to which the service resource meets the service request requirements. For example, a weighted Euclidean distance method can be used, where the service resource load state value of each element is correlated with the service request feature value to calculate a distance. The smaller the distance value, the better the match between the service resource load and the service request features, and the higher the matching score.

[0116] The load matching score is compared with the preset service resource capacity constraints, response time constraints, and spatial reachability constraints, and the real-time load status analysis data of the service resources is output.

[0117] The comparison of load matching score with capacity constraints, response time constraints, and spatial reachability constraints refers to the hospital's logistics service resource management system using pre-set constraint standards to determine whether the matching score meets the actual capacity, time, and space requirements, thereby generating real-time load status analysis data.

[0118] S5: Based on service request priority ranking data and real-time service resource load status analysis data, a multi-objective constraint optimization matching algorithm is used to establish a dynamic correlation model between service request characteristics and real-time service resource status, and output a dynamic matching and scheduling scheme for service resources, including:

[0119] Based on service request priority ranking data and real-time service resource load status analysis data, a mapping cost matrix between service requests and service resources is generated.

[0120] The mapping cost matrix is ​​a two-dimensional matrix formed by the hospital's logistics service resource management system using service request priority ranking data and real-time service resource load status analysis data to calculate the matching degree between service requests and service resources. The system arranges each service request in the row direction of the matrix according to its priority, and arranges all service resources in the column direction, so that each matrix element corresponds to one service request and one service resource. The mapping cost is a value calculated using multi-dimensional data such as service request priority and real-time service resource load status, used to describe the overall cost incurred in allocating a service request to a specific service resource. By calculating the cost for each service request and each type of service resource, a complete mapping cost matrix is ​​formed. Each element of the matrix represents the cost value under the corresponding service request and service resource combination. The mapping cost matrix reflects the mapping relationship between service requests and service resources.

[0121] Based on the mapping cost matrix, a dynamic correlation model between service request characteristics and real-time status of service resources is established.

[0122] The hospital's logistics service resource management system utilizes a mapping cost matrix to construct a data calculation model—a dynamic correlation model—that can continuously update the relationship between service request characteristics and the real-time load status of service resources. Service request characteristics refer to multi-dimensional feature data such as the type of service request, urgency level, information completeness, request location, and request submission time. Real-time service resource status refers to real-time load data such as the current task queue length, average processing time, and idle waiting time. The dynamic correlation model establishes a real-time correspondence between changes in service request characteristics and changes in the real-time status of service resources by analyzing each element within the mapping cost matrix. For example: A service request for repair of the indoor unit of the central air conditioning in Clinic 301, East Side, 3rd Floor, Outpatient Building, Emergency Level I, submitted at 10:45 AM on July 20, 2025. The service request features include a spatial location code of "MZL-03D-301", a granularity weight of 0.5, and an information completeness of 1.0; it is currently unprocessed. The service resource, service technician Wang, has a real-time status data queue length of 3, an average processing time of 22.5 minutes, and an idle time of 5 minutes. Based on the cost value of 0.1 in the mapping cost matrix, requests with similar locations, higher weights, and complete information are strongly associated with resources with shorter queue lengths and longer idle times, forming a dynamic association rule between service requests and resources. Summarizing all dynamic association rules constitutes the dynamic association model. The model's dynamism is reflected in the fact that whenever the service request feature data or service resource status data changes, the model immediately recalculates the mapping cost matrix and updates the association rules based on the latest data, thereby achieving real-time adjustment of the matching relationship between service requests and service resources.

[0123] Based on the dynamic correlation model, we construct three objective functions: latency minimization, load balancing maximization, and path distance minimization, forming a multi-objective optimization objective set.

[0124] The latency minimization objective function is a mathematical function constructed by the hospital's logistics service resource management system to minimize the total latency required for all service requests from submission to completion. Latency time is a quantification of the service request response speed. For example, the ideal latency for an emergency level I central air conditioning main duct repair request is within 15 minutes. The hospital wants the actual processing time of the request to be as short as possible, so the function is established with the minimum actual response latency as the optimization objective. The load balancing maximization objective function is constructed with the goal of distributing the current load levels of all service resources as evenly as possible. For example, if the hospital's logistics department has 10 maintenance technicians, the objective function sets the load of each technician to be as close as possible, i.e., a load value between 0.4 and 0.6 is considered optimal. The path distance minimization objective function is constructed with the goal of minimizing the total distance the service resource needs to travel to perform a task. For example, the distance between maintenance technician Wang and clinic 301 is 200 meters, and the distance to Li is 800 meters. The function objective sets the minimum sum of the actual physical distances between resources and service requests. These objective functions together constitute a multi-objective optimization objective set to comprehensively evaluate the overall rationality of service request matching.

[0125] The delay minimization objective function is a mathematical expression established by the hospital's logistics service resource management system to optimize the sum of all service request delays from the time a request is submitted to the time the corresponding service resource actually completes its task. The mathematical expression is:

[0126] The objective function for minimizing latency is equal to the sum of the differences between the actual completion time of all service requests and the submission time of the corresponding service request.

[0127] Assume the hospital's logistics service management system currently has a total of service requests, represented by the uppercase letter N. Each service request has a unique identifier, denoted sequentially as service request number 1 to service request number N. For any service request number i, where the integer i satisfies the value range from 1 to the uppercase letter N, the submission time of service request number i is recorded as the submission time of service request number i. The actual completion time of the service resource corresponding to service request number i is a timestamp in a uniform format recorded when the service resource corresponding to service request number i actually finishes processing the service request.

[0128] For example, if the submission time of service request sequence number 1 is 10:45 on 2025-07-20 and the actual completion time is 11:00 on 2025-07-20, then the delay time of service request sequence number 1 is 15 minutes. And so on, the delay minimization objective function expression is to sum the delay times corresponding to all service request sequences from 1 to N, thereby determining the total delay time of the delay minimization objective function.

[0129] The load balancing maximization objective function is a mathematical expression constructed by the hospital's logistics service resource management system to achieve a balanced distribution of load across all service resources. The mathematical expression is as follows:

[0130] The objective function for maximizing load balancing is to maximize the load balancing index of all service resources, which is equal to one minus the variance of the load of all service resources.

[0131] Assume the total number of service resources used by the hospital's logistics department to fulfill service requests is represented by the uppercase letter M. Each service resource also has a unique identifier, denoted sequentially as service resource number 1 to service resource number M. For any service resource number j, where the integer j satisfies the value range from 1 to the uppercase letter M, the real-time load status value of service resource number j is the normalized index of the service resource in the real-time load status analysis data, denoted as the real-time load status value of service resource number j. The average load of all service resources is the sum of the real-time load status values ​​of all service resources divided by the total number of service resources M. Therefore, the variance of the service resource load is the sum of the squares of the differences between the real-time load status values ​​corresponding to each service resource number from 1 to M and the average load of all service resources, divided by the total number of service resources M.

[0132] For example, if a hospital currently has ten maintenance technicians, and each technician's real-time load values ​​are 0.4, 0.5, 0.6, 0.5, 0.55, 0.45, 0.5, 0.5, 0.45, and 0.55 respectively, then the average load of all service resources is the sum of these values ​​divided by 10, which equals 0.5. The load variance is calculated by summing the squares of the differences between each technician's load value and 0.5, then dividing by 10. The load balancing objective function is the load balancing index obtained by subtracting the variance value.

[0133] The path distance minimization objective function is a mathematical expression established by the hospital logistics service resource management system to minimize the total distance of the travel paths required for all service resources to satisfy service requests. The mathematical expression is as follows:

[0134] The objective function for minimizing path distance is: to minimize the sum of the actual physical path distances between all service requests and the corresponding service resources allocated.

[0135] For any service request sequence number i and its corresponding allocated service resource sequence number j, the position of service request sequence number i is the spatial location code after the aforementioned normalization process, and the current real-time location of service resource sequence number j adopts the same coding standard. The actual path distance between service request sequence number i and corresponding service resource sequence number j refers to the actual physical distance calculated using the distance data between hospital spatial location codes pre-stored in the hospital logistics service resource management system. The objective function for minimizing the path distance is the sum of the path distances required for all allocated service resources corresponding to service request sequences i to N to actually move to the service request location, with the optimization objective being to minimize the total distance.

[0136] For example, maintenance technician Wang is located at MZL-03D-302 and needs to go to clinic 301 (MZL-03D-301). The system presets the distance between locations to be 50 meters. Another maintenance technician, Li, is currently located at B101 on the first floor of the ward building. To go to clinic 301, he needs to move 800 meters. The objective function of minimizing the path distance calculates the distances of each item by summing them up based on the correspondence between the service request and the service resource decision variable matrix. The goal is to minimize the total distance, thereby improving service efficiency.

[0137] Transform the multi-objective optimization objective set into a comprehensive evaluation function;

[0138] The hospital's logistics service resource management system utilizes linear weighting to combine the objective functions of latency minimization, load balancing maximization, and path distance minimization into a single comprehensive evaluation function by assigning weight coefficients. For example, if the latency minimization weight is set to 0.5, the load balancing weight to 0.3, and the path distance weight to 0.2, then the three objective functions are multiplied by their respective weight coefficients and the results are summed to uniformly measure the overall quality of service request and service resource matching.

[0139] The comprehensive evaluation function is iteratively solved within the constraint set to obtain the service request and service resource matching decision variable matrix, and a dynamic matching and scheduling scheme for service resources is generated.

[0140] The constraint set comprises pre-defined constraints from the hospital, including service resource capacity constraints (maximum number of tasks each resource can handle simultaneously), response time constraints (response times corresponding to different urgency levels), and spatial reachability constraints (service resources can reach the requested location within a predetermined response time). Integer programming is used to iteratively solve the comprehensive evaluation function within the constraint set to determine the optimal service request and service resource matching decision variable matrix. The elements of the decision variable matrix are either 0 or 1; for example, a request for repairing the central air conditioning indoor unit in Room 301 of the outpatient building is assigned as 1 to Wang and 0 to Li. Once the matrix is ​​determined, a dynamic matching and scheduling scheme for the hospital's logistical service resources is formed.

[0141] S6: Based on a dynamic matching and scheduling scheme for service resources, automatically generate and dispatch maintenance work orders, and optimize the scheduling of logistical service resources in real time, including:

[0142] Based on a dynamic matching and scheduling scheme for service resources, maintenance work orders are automatically generated.

[0143] The dynamic matching and scheduling scheme for service resources is a scheme formed by the hospital's logistics service management system based on service request priority ranking data and real-time load status analysis data of service resources. It records the matching relationship between each service request and its corresponding service resource. Maintenance work orders are task documents automatically generated electronically by the hospital's logistics service management system. They record the content of each service request and information on the matching service resources, guiding the actual maintenance work. Maintenance work orders include the service request type, the name of the equipment to be repaired, the location of the equipment, the request submission time, the urgency level of the request, the department information of the person submitting the service request, and the service resource information matching each service request. Service request types include, for example, water and electricity repair, central air conditioning repair, or elevator repair, etc. Equipment names include, for example, central air conditioning indoor units, medical ventilators, etc. The location of the equipment to be repaired uses the hospital's preset spatial location coding rules, for example, "MZL-03D-301" represents room 301 on the east side of the third floor of the outpatient building. The request submission time is a standardized format, for example, "2025-07-20 10:45". The urgency level of the request is marked as "I", "II", "III", or "IV". The department information of the person submitting the service request, such as the name of internal medicine, surgery, or outpatient department, is required. The service resource information matching the service request includes the name, skills, qualifications, current location, and contact information of the service personnel. The maintenance work order generation process is automated, requiring no manual intervention, and is completed automatically based on the matching relationship of the dynamic matching and scheduling scheme of service resources.

[0144] Based on the service request type in the maintenance work order and the matching available service resource information, the maintenance work order is automatically dispatched to the corresponding available service resource;

[0145] The automated work order dispatch process involves the hospital's logistics service management system automatically pushing generated work orders to the mobile terminal devices (such as mobile phones or smart terminals) of the matching service personnel via its internal communication network. The automated dispatch process includes the system confirming the service request type in the work order; for example, "plumbing and electrical repairs" are dispatched to maintenance personnel, "central air conditioning repairs" to sanitation maintenance personnel, and "elevator and other" to consumables delivery personnel. The dispatch recipient is automatically identified based on the name and contact information of the matching service personnel recorded in the work order, eliminating the need for manual selection. Each engineer has a corresponding weight value, calculated based on historical service performance data. For any given repair request, the total weight value for each service personnel is calculated by comprehensively considering their current workload, distance traveled, historical response speed, and service evaluation. The specific calculation formula is as follows:

[0146] Total task weight value = α × (1 - task load rate) + β × (1 - task distance / maximum response distance) + γ × historical average response speed index + δ × historical service evaluation index;

[0147] In this system, α, β, γ, and δ are weighting coefficients, set according to the hospital's preset optimization strategy. The task load rate is the ratio of the number of currently assigned but incomplete tasks to the set load threshold; the task distance is the actual distance between the current location of the service resource personnel and the location of the service request; the maximum response distance is the response distance constraint standard stipulated by the hospital; the historical average response speed index is the ratio of the average response speed of this service resource personnel over the past three months to the average response speed of all service resource personnel; and the historical service evaluation index is the ratio of the average user feedback evaluation score after service completion over the past three months to the average evaluation score of all service resource personnel in the hospital. Based on the calculated total task weight value, the service resource personnel with the highest weight value are prioritized for automatic dispatch of maintenance work orders. Upon receiving the notification, the dispatched maintenance personnel confirm receipt of the task via their mobile terminal, record the work order dispatch status as "dispatched," and record the dispatch time and receipt status.

[0148] After a maintenance work order is dispatched, it is scheduled in real time according to the request submission time in the maintenance work order and the response time limit standard corresponding to the service request urgency level preset by the hospital, so as to optimize the scheduling of logistics service resources in real time.

[0149] Real-time dispatch allows the hospital's logistics service management system to monitor the processing progress of maintenance work orders in real time, ensuring that service resources adhere to the response time limits corresponding to the urgency level marked on the work order. Response time limits are determined based on the hospital's preset urgency levels for service requests: Level "I" requires arrival within 15 minutes, Level "II" within 30 minutes, Level "III" within 2 hours, and Level "IV" within 4 hours. Real-time dispatch automatically tracks the real-time distance between the current location of maintenance personnel and the location of the equipment requiring repair. If maintenance personnel fail to arrive within the specified time limit, an overdue alarm notification is automatically sent to the logistics service management personnel, who are also notified to urgently dispatch the maintenance task to a nearby backup maintenance personnel for emergency response, ensuring that service requests are effectively processed within the preset response time limits.

[0150] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0151] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0152] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0153] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0154] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0155] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0156] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0157] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0159] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for integrating and optimizing hospital logistics service resources, characterized in that, Includes the following steps: S1: Collect service request data from the mobile service platform, standardize the service request data, and output a set of service request data. S2: Based on the service request data set, analyze the differences in data granularity, urgency level marking method, and information completeness of different service requests, and output multi-source service request feature analysis data; S3: Based on the multi-source service request feature analysis data, establish a service request fusion model, uniformly re-evaluate the urgency level of service requests, and output service request priority ranking data; S4: Based on the multi-source service request feature analysis data, a multi-dimensional state feature correlation analysis method is used to calculate and verify the current load status of service resources, and output real-time load status analysis data of service resources. S5: Based on the service request priority sorting data and the real-time load status analysis data of service resources, a multi-objective constraint optimization matching algorithm is used to establish a dynamic correlation model between service request characteristics and the real-time status of service resources, and output a dynamic matching and scheduling scheme for service resources. S6: Based on the dynamic matching and scheduling scheme of service resources, it automatically generates and dispatches maintenance work orders and optimizes the scheduling of logistics service resources in real time.

2. The method for integrating and optimizing hospital logistics service resources according to claim 1, characterized in that, S1, specifically: Collect service request data from the mobile service platform; the service request data includes the service request type, the name of the device being repaired, the location of the device being repaired, the time the request was submitted, and the urgency level of the request; The service request data is normalized and output as a normalized service request data set.

3. The method for integrating and optimizing hospital logistics service resources according to claim 2, characterized in that, The standardized processing includes: uniformly coding the names of the equipment reported for repair according to the hospital's preset equipment classification standards; standardizing the description of the location of the equipment reported for repair based on the hospital's preset spatial location coding rules; converting the request submission time using a unified time format; and re-marking the urgency level of the request according to the hospital's unified urgency level assessment standards.

4. The method for integrating and optimizing hospital logistics service resources according to claim 3, characterized in that, S2, specifically: The data granularity of service request types in the standardized service request dataset is statistically analyzed to obtain the data granularity statistical results corresponding to each service request type; The emergency level marking methods of service requests in the standardized service request dataset are compared to identify differences, and the differences in the emergency level marking methods of service requests are obtained. The completeness of information regarding the service request submission time and the location of the reported equipment in the standardized service request dataset is evaluated to obtain the information completeness evaluation results for the service request submission time and the location of the reported equipment. The output includes multi-source service request feature analysis data, including data granularity statistical results, differences in emergency level labeling methods, and information completeness assessment results.

5. The method for integrating and optimizing hospital logistics service resources according to claim 4, characterized in that, S3, specifically: A service request fusion model is established based on multi-source service request feature analysis data; Based on the service request fusion model, each service request is re-evaluated for its urgency level, and the service request priority ranking data is output.

6. The method for integrating and optimizing hospital logistics service resources according to claim 5, characterized in that, Based on multi-source service request feature analysis data, a service request fusion model is established, specifically as follows: Determine the granularity weight for each service request type based on the data granularity statistics results; A unified marking rule for the emergency level of a service request is determined based on the differences in emergency level marking methods; Set the information integrity threshold for service requests based on the information completeness assessment results.

7. The method for integrating and optimizing hospital logistics service resources according to claim 6, characterized in that, Based on the service request fusion model, each service request undergoes a unified reassessment of its urgency level, specifically as follows: Adjust the initial urgency level corresponding to the service request type using granular weights; Re-label the emergency levels from different sources according to the unified labeling rules; The urgency level of service requests that do not meet the information integrity threshold will be reduced.

8. The method for integrating and optimizing hospital logistics service resources according to claim 7, characterized in that, S4, specifically: Collect the task queue length, average task processing time and idle waiting time of each service resource within the current time window to form the original data matrix of service resource load. The original data matrix of service resource load is normalized to obtain the normalized index vector of service resource load. The service resource load normalization index vector is concatenated with the multi-source service request feature analysis data to form a multi-dimensional tensor, thus constructing a multi-dimensional state feature tensor. Based on the multidimensional state feature tensor, the load matching score between each service resource and the corresponding service request is calculated. The load matching score is compared with the preset service resource capacity constraints, response time constraints, and spatial reachability constraints, and the real-time load status analysis data of the service resources is output.

9. The method for integrating and optimizing hospital logistics service resources according to claim 8, characterized in that, S5, specifically: Based on service request priority ranking data and real-time service resource load status analysis data, a mapping cost matrix between service requests and service resources is generated. Based on the mapping cost matrix, a dynamic correlation model between service request characteristics and real-time status of service resources is established. Based on the dynamic correlation model, we construct three objective functions: latency minimization, load balancing maximization, and path distance minimization, forming a multi-objective optimization objective set. Transform the multi-objective optimization objective set into a comprehensive evaluation function; The comprehensive evaluation function is iteratively solved within the constraint set to obtain the decision variable matrix for matching service requests and service resources, and a dynamic matching and scheduling scheme for service resources is generated.

10. The method for integrating and optimizing hospital logistics service resources according to claim 9, characterized in that, S6, specifically: Based on a dynamic matching and scheduling scheme for service resources, maintenance work orders are automatically generated. Based on the service request type in the maintenance work order and the matching available service resource information, the maintenance work order is automatically dispatched to the corresponding available service resource; After a maintenance work order is dispatched, it is scheduled in real time according to the request submission time in the work order and the response time limit standard corresponding to the hospital's preset service request urgency level, so as to optimize the scheduling of logistics service resources in real time.

Images