Hospital operation analysis and target management system based on big data

Abstract

The invention discloses a hospital operation analysis and target management system based on big data, and relates to the technical field of hospital operation management, and the system employs an activity-based cost method ABC, introduces a non-linear sharing function and a multi-dimensional cost sharing model, enables the operation cost to be refined to a specific project of each department, and improves the operation efficiency. Equipment use, medicine consumption and working hours of medical staff are covered, cost allocation and updating are carried out through a dynamic adjustment mechanism and a system according to real-time resource use conditions, hospitals are helped to identify high-cost expenditure points, resource waste is reduced, transparency of operation cost and management efficiency are improved, an automatic feedback mechanism is introduced, and the system is convenient to use. According to the method, the predicted data and the actual data can be compared in real time, the operation abnormity can be identified, the adjustment suggestion is automatically triggered when the abnormity is found, the adjustment suggestion is pushed to the related management personnel for quick response, and compared with a manual examination method, the speed and the accuracy of management response are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hospital operation management, and in particular to a hospital operation analysis and target management system based on big data. Background Art

[0002] In modern hospital operation and management, operation cost analysis and PDCA closed-loop management are the core links to ensure rational allocation and efficient management of resources. However, although big data technology has gradually been applied to hospital operation and management, there are still many shortcomings in the combination of the two.

[0003] In terms of operating cost analysis, there are problems such as insufficient data refinement, inability to achieve accurate cost allocation, and lack of real-time performance. Existing cost analysis methods mainly rely on department-level budget reports, which are difficult to refine to each operational link. In operating rooms and inpatient departments, the costs of different items such as specific equipment use, drug consumption, and medical staff allocation are not accurately allocated, resulting in rough cost calculations and failure to discover resource waste in a timely manner. Hospital cost analysis mainly relies on quarterly or annual financial reports. When hospitals face a sudden increase in patient flow, equipment use, and a sharp increase in demand for drugs and consumables, it is difficult for management to discover which consumables have increased demand based on real-time data, and it is difficult to make cost control decisions, resulting in unreasonable resource allocation or high operating costs.

[0004] Secondly, in terms of PDCA closed-loop management, the lack of dynamic management and intelligent feedback mechanisms leads to delayed responses. The hospital's PDCA management cycle (plan, do, check, act) relies on the management's advance planning. Obviously, this type of plan cannot cope with situations such as emergency peaks and epidemic outbreaks. Lacking real-time data support, the PDCA management closed loop makes it difficult to adjust operational goals in real time, and the feedback cycle for operational problems is relatively long, resulting in a time lag between problem exposure and improvement.

[0005] It can be seen that in terms of cost control, there are problems with data refinement and lack of real-time performance. It is difficult for hospital management to quickly locate high-cost projects or unreasonable resource use in operations. Although annual or quarterly financial reports can reveal macro-level expenditure problems, the control response to specific projects is delayed, and timely intervention is not possible when problems arise. Due to the lag in cost data, hospitals often waste resources when they discover problems, and the optimization of resource allocation usually takes several months to adjust, which affects the overall operational efficiency of the hospital. There is a lack of refined analysis of detailed projects and overall department budget management, and it is impossible to effectively balance the budgets of various departments. It is possible that a department has a high demand for a certain drug, while other departments have redundancy.

[0006] However, PDCA management lacks an intelligent data feedback mechanism, making it difficult to achieve data-driven automated management. Even though some hospitals have introduced big data tools, the feedback is not automated and real-time enough, and management still relies on manual review and decision-making, resulting in a slow operational optimization process. Therefore, there is an urgent need for a hospital operation analysis and target management system based on big data to solve such problems. Summary of the invention

[0007] In view of the above existing problems, the present invention is proposed.

[0008] Therefore, the present invention provides a hospital operation analysis and target management system based on big data to solve the problems of lag and insufficient precision in operation cost analysis and PDCA closed-loop management in current hospital operation management.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0010] On the one hand, the present invention provides a hospital operation analysis and target management system based on big data, which includes:

[0011] Data collection module, used for operation data collection. The data sources are IoT devices, electronic medical record system EMR and operation system. Operation data includes financial data, patient flow, equipment utilization rate, drug consumption, bed utilization rate, equipment operation status (equipment utilization rate) and patient satisfaction.

[0012] A data dashboard that displays the hospital's key operating indicators, including bed occupancy rate, equipment operating status, and patient satisfaction;

[0013] An automated feedback system triggers adjustment suggestions when operational anomalies are identified;

[0014] Operation cost control module, used for operation cost control, allocating operation costs to departments, projects or equipment;

[0015] The operating cost control module adopts activity-based costing (ABC) to break down costs into individual items in a department, including equipment usage, drug consumption, and medical staff hours for specific surgeries. Based on real-time operational data, resources are reallocated to high-demand departments.

[0016] On the other hand, the present invention provides a hospital operation analysis and target management method based on big data, comprising:

[0017] Step S1, real-time data collection,

[0018] Based on the hospital's IoT devices, EMR system and operation system (finance, procurement, resource scheduling system), the operation data of the hospital's operation links are collected and integrated, and the operation data is updated in real time; the operation data includes financial data, patient flow, equipment utilization rate, drug consumption, bed utilization rate, equipment operation status (equipment utilization rate) and patient satisfaction;

[0019] Step S2: multi-dimensional analysis of operational data.

[0020] Based on historical financial data, patient flow, drug consumption and equipment utilization, multi-dimensional analysis is carried out, and dynamic visual data dashboards are used to display key hospital operating indicators in real time, including bed utilization, equipment operation status and patient satisfaction;

[0021] Step S3, PDCA closed-loop management,

[0022] In the PDCA closed-loop management, the focus is on optimizing the "check" and "action" links, and automatic feedback is provided based on the multi-dimensional analysis results of step S2. The automated feedback system is used to compare the results of the multi-dimensional analysis, i.e. the predicted data, with the actual operation data. When an operational anomaly is found, adjustment suggestions are automatically triggered;

[0023] Step S4, operating cost control,

[0024] Based on the operational abnormality data fed back in step S3, operational cost control is performed to allocate the operational costs to departments, projects or equipment.

[0025] Further, multidimensional analysis includes:

[0026] Resource demand forecasting, based on bed occupancy rate and seasonal patient peak data, predicts subsequent resource demand, including drug consumption and bed occupancy rate;

[0027] Financial analysis and forecasting: Based on historical financial data, forecast subsequent resource consumption trends, i.e. changes in financial data;

[0028] Operational efficiency analysis compares resource usage of different departments, including equipment operating status and drug consumption.

[0029] Furthermore, the resource demand prediction method in step S2 is:

[0030] Forecasting based on time series analysis combined with seasonal models,

[0031] Among them, R t+krepresents the resource demand forecast value at time t+k, that is, the future bed utilization rate or drug demand, α represents the weight coefficient of historical resource demand for future prediction, N represents the time window size used to calculate the historical average value, that is, the historical resource demand data from time t-1 to time tN, R t-i represents the resource demand at the ti-th moment, i.e., the i-th historical data point, β represents the influence coefficient of patient flow on future resource demand, M represents the time window size used to calculate the average patient flow, i.e., the historical patient flow data from t-1 to tM, P t-j represents the patient flow at time tj, γ represents the influence coefficient of seasonal fluctuation, S t represents the seasonal factors at the current time t, such as a sudden increase in patient traffic during the flu season or holidays, and ∈ represents noise;

[0032] Financial analysis and forecasting is based on historical financial data, combined with resource consumption and equipment usage, to predict future financial expenditure trends.

[0033] Among them, C t+k represents the predicted value of financial expenditure at the t+kth moment, δ represents the weight coefficient of historical financial expenditure on future prediction, Q represents the time window size used to calculate the average value of historical financial expenditure, and C t-i represents the financial expenditure data at the ti-th moment, η represents the influence coefficient of equipment utilization rate on financial expenditure, U t represents the equipment utilization rate at the current time t, that is, the equipment workload of each department of the hospital, μ represents the impact coefficient of drug consumption on financial expenditure, and D t represents the drug consumption at the current time t, ν represents the influence coefficient of human resource allocation on financial expenditure, E t represents the staffing situation at the current time t, that is, human resource input, and ξ represents noise.

[0034] Furthermore, the operational efficiency analysis in step S2 is used to identify inefficient resource usage, including low equipment utilization in a department and unreasonable consumption of a certain drug. Relative efficiency analysis is used to measure the relative operational efficiency of the department.

[0035] Among them, E i represents the operational efficiency index of the i-th department, reflecting the efficiency of the department’s resource utilization, R i represents the resource usage of the i-th department, which refers to the bed utilization rate or equipment utilization rate, P i represents the patient flow of the i-th department, U i represents the equipment utilization rate of the i-th department, reflecting the ratio of the actual working time of equipment use to the available working time, C irepresents the financial expenditure of the i-th department, D i represents the drug consumption of the i-th department, E i represents the staffing situation of the i-th department, w1, w2, w3, w4, w5, w6 represent the weight parameters of each operating factor;

[0036] Resource demand forecasting is based on historical demand, patient flow and seasonal fluctuations to predict future resource demand and ensure sufficient medical resources during peak hours. Financial analysis forecasting uses historical expenditure data and real-time equipment, drug and manpower consumption data to help hospitals predict future financial expenditures and optimize budget allocation. Operational efficiency analysis measures the relative efficiency of each department, locates operational bottlenecks, and provides data support for resource allocation.

[0037] Furthermore, operational anomalies include drug consumption, equipment utilization, and bed utilization exceeding expectations, and resource utilization of a department exceeding the budget;

[0038] The automated feedback system automatically generates adjustment suggestions, including redeploying medical staff, replenishing drug inventory, and opening new beds, and pushes them to relevant managers.

[0039] Furthermore, the PDCA closed-loop management method in step S3 is:

[0040] Calculate the difference between actual operating data and forecast data, and use weighted difference analysis to compare resource requirements in multiple dimensions, including beds, medicines, and equipment usage.

[0041] Where ΔX t Represents the overall operation deviation at time t, which is used to measure the deviation degree of multi-dimensional operation data. i,t represents the actual operation data of the i-th dimension, including the values ​​of bed utilization rate, drug consumption, and equipment utilization rate at time t, P i,t represents the forecast data of the i-th dimension, that is, the forecast result at the t-th time obtained in the multidimensional analysis, w i represents the weight of the i-th dimension data, that is, the contribution of this dimension to the overall deviation, the overall operation deviation ΔX t As a basis for determining whether an abnormality exists;

[0042] Using the standard deviation deviation model, set a predetermined threshold, when the overall operation deviation ΔX t When this threshold is exceeded, |ΔX t |>θ t , then anomaly detection is triggered, θ t represents the dynamically adjusted threshold, θ t Set θ based on the standard deviation of historical data t=σ·k, where σ is the standard deviation of historical operating data and k is the threshold coefficient;

[0043] When an operational anomaly is detected, adjustment suggestions are automatically triggered. Different adjustment measures are triggered for anomalies in different dimensions:

[0044] Resource allocation suggestions: If resource consumption (such as beds and medicines) exceeds expectations, resource allocation suggestions will be triggered.

[0045] Among them, R dp A represents the amount of resources that need to be deployed, i.e., the amount of additional beds or drug inventory. r,t represents the actual resource consumption value at time t, P r,t It represents the resource forecast value, i.e., the drug or bed demand forecast result in multidimensional analysis, and S represents the allocation coefficient;

[0046] Medical staff deployment suggestions: if the bed occupancy rate exceeds expectations, additional medical staff will be added.

[0047] Among them, N dp Indicates the number of medical staff that need to be deployed, A b,t represents the actual bed utilization rate at time t, P b,t represents the predicted bed occupancy rate, and H represents the medical staff deployment coefficient, that is, the proportion of medical staff deployed according to the change in bed occupancy rate;

[0048] Procurement recommendations: If drug consumption is significantly higher than expected, drug procurement recommendations will be triggered.

[0049] Among them, Q cg Indicates the number of additional drugs that need to be purchased, A d,t represents the actual drug consumption at time t, P d,t represents the predicted drug demand, and C represents the procurement coefficient.

[0050] Furthermore, the PDCA closed-loop management method in step S3 also includes:

[0051] Threshold θ t Dynamically adjust based on real-time data and historical standard deviation,

[0052] θ t+1 =λ·θ t +(1-λ)·σ t+1 , where θ t+1 represents the dynamic threshold at the next moment, λ represents the adjustment coefficient, σ t+1 Represents the forecast standard deviation for the next moment.

[0053] Furthermore, the operating cost control method is:

[0054] Using activity-based costing (ABC), costs are broken down into individual items in a department, including equipment usage for specific surgeries, drug consumption, and medical staff hours. Resources are reallocated to high-demand departments based on real-time operational data.

[0055] Furthermore, step S4 uses activity-based costing ABC to allocate costs:

[0056] Use nonlinear allocation function to refine cost allocation. Among them, C ijk represents the cost allocation of the jth project of the i-th department in the kth resource category, α represents the adjustment coefficient of activity cost allocation, which is used to adjust the cost allocation ratio between various projects, and V ijk represents the k-th resource usage of the j-th project in the i-th department, including drug consumption, equipment usage time, and staff hours. represents the total resource usage of the i-th department in all m projects, β represents the nonlinear weight coefficient of resource allocation, and adjusts the sensitivity of different resources to cost allocation. i It represents the total indirect cost of the i-th department, which is allocated to each project. The nonlinear weight β is used to refine the cost allocation for large-scale resource consumption, and projects with large resource usage will allocate more costs;

[0057] Resource consumption is broken down into multiple dimensions, including equipment, personnel, medicines, and consumables:

[0058] Among them, C ij represents the total shared cost of the jth project in the i-th department, λ k represents the weight of the k-th resource, V ijk represents the actual consumption of the kth resource in the jth project of the ith department, β k represents the nonlinear allocation coefficient of the k-th resource, C i represents the total indirect cost of the i-th department, It represents the total consumption of the kth resource of all projects in the ith department, which is used for standardized allocation;

[0059] Dynamically adjust cost allocation based on actual consumption data,

[0060] in, represents the apportioned cost at the next moment t+1, η represents the smoothing coefficient, determines the weight ratio of the current moment and the historical apportioned data, and controls the adjustment speed. Indicates the resource consumption at the next time t+1, combines the historical allocation data with the new resource consumption, and gradually adjusts the cost allocation results;

[0061] After the allocation is completed, the cost efficiency of each department is evaluated to measure the matching degree between resource consumption and output.

[0062] Among them, E ij represents the cost efficiency coefficient of the jth project of the i-th department, O ij represents the j-th project output of the i-th department, including surgical success rate and patient satisfaction, C ij represents the total shared cost of the jth project in the i-th department; the cost efficiency coefficient E ij Used to measure the efficiency of resource utilization.

[0063] The beneficial effects of the present invention are:

[0064] The present invention, through the activity-based costing method ABC, introduces nonlinear allocation functions and multi-dimensional cost allocation models, refines the operating costs to specific items of each department, covers equipment use, drug consumption, and medical staff hours. Through the dynamic adjustment mechanism, the system updates the cost allocation according to the real-time resource usage, helps hospitals identify high-cost expenditure points, which not only reduces resource waste, but also improves the transparency of operating costs and management efficiency.

[0065] The present invention collects key operational data such as patient flow, equipment utilization, and drug consumption in real time through IoT devices, electronic medical record systems EMR, and other operational systems, and displays them to management in real time through a multi-dimensional analysis dashboard. Through intelligent PDCA closed-loop management, adjustment suggestions are triggered in a timely manner based on the deviation between real-time data and predicted data, helping management to quickly make response decisions when faced with increased patient flow, frequent equipment use, or a sharp increase in drug consumption. Compared with previously lagging financial reports, real-time feedback significantly shortens the lag between problem exposure and decision-making.

[0066] The present invention combines historical data with seasonal fluctuations to accurately predict future resource needs. When encountering emergencies, such as emergency peaks or epidemic outbreaks, it can monitor operational conditions in real time and automatically trigger resource allocation suggestions. It can also allocate medical resources in a timely manner based on the difference between predicted and actual data, and can respond quickly when faced with emergencies.

[0067] The present invention introduces an automated feedback mechanism that can compare predicted data with actual data in real time, identify operational anomalies, automatically trigger adjustment suggestions when anomalies are found, and push them to relevant managers for rapid response. Compared with manual review methods, the speed and accuracy of management response are improved.

[0068] The present invention, based on the activity-based costing method ABC, allocates costs to each specific project, and combines it with operational anomaly detection to accurately locate high-cost projects and unreasonable resource usage points in operations, and makes dynamic adjustments based on actual operational data to reallocate resources to departments with high demand to avoid idle resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0070] Figure 1 It is a schematic diagram of the hospital operation analysis and target management system structure of the present invention;

[0071] Figure 2 It is a flowchart of the hospital operation analysis and target management method of the present invention. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0073] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0074] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0075] Example 1, reference Figure 1 , this embodiment provides a hospital operation analysis and target management system based on big data, including:

[0076] Data collection module, used for operation data collection. The data sources are IoT devices, electronic medical record system EMR and operation system. Operation data includes financial data, patient flow, equipment utilization rate, drug consumption, bed utilization rate, equipment operation status (equipment utilization rate) and patient satisfaction.

[0077] A data dashboard that displays the hospital's key operating indicators, including bed occupancy rate, equipment operating status, and patient satisfaction;

[0078] An automated feedback system triggers adjustment suggestions when operational anomalies are identified;

[0079] Operation cost control module, used for operation cost control, allocating operation costs to departments, projects or equipment;

[0080] The operating cost control module adopts activity-based costing (ABC) to break down costs into individual items in a department, including equipment usage, drug consumption, and medical staff hours for specific surgeries. Based on real-time operational data, resources are reallocated to high-demand departments.

[0081] Example 2, reference Figure 2 This embodiment provides a hospital operation analysis and target management method based on big data, comprising the following steps:

[0082] Step S1, real-time data collection,

[0083] Based on the hospital's IoT devices, EMR system and operation system (finance, procurement, resource scheduling system), the operation data of the hospital's operation links are collected and integrated, and the operation data is updated in real time; the operation data includes financial data, patient flow, equipment utilization rate, drug consumption, bed utilization rate, equipment operation status (equipment utilization rate) and patient satisfaction;

[0084] Specifically, based on real-time data collection from IoT devices, electronic medical record systems (EMRs), and various hospital operation systems (finance, procurement, and resource scheduling systems), comprehensive and real-time updated operation data is obtained, including patient flow, bed occupancy rate, drug consumption, equipment operating status, and patient satisfaction indicators, to gain a comprehensive grasp of the operation status. Compared with previous decision-making based on quarterly or annual report data, this method significantly improves the hospital's response speed, allowing management to grasp the actual consumption of resources in a short period of time and make more timely adjustments.

[0085] Step S2: multi-dimensional analysis of operational data.

[0086] Based on historical financial data, patient flow, drug consumption and equipment utilization, multi-dimensional analysis is carried out, and dynamic visual data dashboards are used to display key hospital operating indicators in real time, including bed utilization, equipment operation status and patient satisfaction;

[0087] Multidimensional analysis includes:

[0088] Resource demand forecasting, based on bed occupancy rate and seasonal patient peak data, predicts subsequent resource demand, including drug consumption and bed occupancy rate;

[0089] Financial analysis and forecasting: Based on historical financial data, forecast subsequent resource consumption trends, i.e. changes in financial data;

[0090] Operational efficiency analysis, comparing resource usage of different departments, including equipment operating status and drug consumption;

[0091] The resource demand prediction method in step S2 is:

[0092] Forecasting based on time series analysis combined with seasonal models,

[0093] Among them, R t+k represents the resource demand forecast value at time t+k, that is, the future bed utilization rate or drug demand, α represents the weight coefficient of historical resource demand for future prediction, N represents the time window size used to calculate the historical average value, that is, the historical resource demand data from time t-1 to time tN, R t-i represents the resource demand at the ti-th moment, i.e., the i-th historical data point, β represents the influence coefficient of patient flow on future resource demand, M represents the time window size used to calculate the average patient flow, i.e., the historical patient flow data from t-1 to tM, P t-j represents the patient flow at time tj, γ represents the influence coefficient of seasonal fluctuation, S t represents the seasonal factors at the current time t, such as a sudden increase in patient traffic during the flu season or holidays, and ∈ represents noise;

[0094] Financial analysis and forecasting is based on historical financial data, combined with resource consumption and equipment usage, to predict future financial expenditure trends.

[0095] Among them, C t+k represents the predicted value of financial expenditure at the t+kth moment, δ represents the weight coefficient of historical financial expenditure on future prediction, Q represents the time window size used to calculate the average value of historical financial expenditure, and C t-i represents the financial expenditure data at the ti-th moment, η represents the influence coefficient of equipment utilization rate on financial expenditure, U t represents the equipment utilization rate at the current time t, that is, the equipment workload of each department of the hospital, μ represents the impact coefficient of drug consumption on financial expenditure, and D t represents the drug consumption at the current time t, ν represents the influence coefficient of human resource allocation on financial expenditure, E t represents the staffing situation at the current time t, i.e., human resource input, and ξ represents noise;

[0096] The operational efficiency analysis in step S2 is used to identify inefficient resource usage, including low equipment utilization in a department and unreasonable consumption of certain drugs. Relative efficiency analysis is used to measure the relative operational efficiency of the department.

[0097] Among them, E i represents the operational efficiency index of the i-th department, reflecting the efficiency of the department’s resource utilization, R i represents the resource usage of the i-th department, which refers to the bed utilization rate or equipment utilization rate, P i represents the patient flow of the i-th department, U i represents the equipment utilization rate of the i-th department, reflecting the ratio of the actual working time of equipment use to the available working time, C i represents the financial expenditure of the i-th department, D i represents the drug consumption of the i-th department, E i represents the staffing situation of the i-th department, w1, w2, w3, w4, w5, w6 represent the weight parameters of each operating factor;

[0098] Resource demand forecasting is based on historical demand, patient flow and seasonal fluctuations to predict future resource demand and ensure sufficient medical resources during peak hours. Financial analysis forecasting uses historical expenditure data and real-time equipment, drug and manpower consumption data to help hospitals predict future financial expenditures and optimize budget allocation. Operational efficiency analysis measures the relative efficiency of each department, locates operational bottlenecks and provides data support for resource allocation.

[0099] Specifically, in terms of multi-dimensional analysis, time series analysis is combined with seasonal models to predict future resource demands, financial expenditure trends, and operational efficiency. By comprehensively considering historical data, real-time data, and seasonal fluctuations, multi-dimensional information such as bed usage, drug demand, equipment utilization, and financial expenditure can be accurately predicted. Through these forecast results, hospitals can make preparations in advance when patient flow increases dramatically or resource consumption exceeds expectations.

[0100] Step S3, PDCA closed-loop management,

[0101] In the PDCA closed-loop management, the focus is on optimizing the "check" and "action" links, and automatic feedback is provided based on the multi-dimensional analysis results of step S2. The automated feedback system is used to compare the results of the multi-dimensional analysis, i.e. the predicted data, with the actual operation data. When an operational anomaly is found, adjustment suggestions are automatically triggered;

[0102] Operational anomalies include drug consumption, equipment utilization, and bed utilization exceeding expectations, and resource utilization of a department exceeding the budget;

[0103] The automated feedback system automatically generates adjustment suggestions, including redeploying medical staff, replenishing drug stocks, and opening new beds, and pushes them to relevant managers;

[0104] The PDCA closed-loop management method in step S3 is:

[0105] Calculate the difference between actual operating data and forecast data, and use weighted difference analysis to compare resource requirements in multiple dimensions, including beds, medicines, and equipment usage.

[0106] Where ΔX t Represents the overall operation deviation at time t, which is used to measure the deviation degree of multi-dimensional operation data. i,t represents the actual operation data of the i-th dimension, including the values ​​of bed utilization rate, drug consumption, and equipment utilization rate at time t, P i,t represents the forecast data of the i-th dimension, that is, the forecast result at the t-th time obtained in the multidimensional analysis, w i represents the weight of the i-th dimension data, that is, the contribution of this dimension to the overall deviation, the overall operation deviation ΔX t As a basis for determining whether an abnormality exists;

[0107] Using the standard deviation deviation model, set a predetermined threshold, when the overall operation deviation ΔX t When this threshold is exceeded, |ΔX t |>θ t , then anomaly detection is triggered, θ t represents the dynamically adjusted threshold, θ t Set θ based on the standard deviation of historical data t =σ·k, where σ is the standard deviation of historical operating data and k is the threshold coefficient;

[0108] When an operational anomaly is detected, adjustment suggestions are automatically triggered. Different adjustment measures are triggered for anomalies in different dimensions:

[0109] Resource allocation suggestions: If resource consumption (such as beds and medicines) exceeds expectations, resource allocation suggestions will be triggered.

[0110] Among them, R dp A represents the amount of resources that need to be deployed, i.e., the amount of additional beds or drug inventory. r,t represents the actual resource consumption value at time t, P r,t It represents the resource forecast value, i.e., the drug or bed demand forecast result in multidimensional analysis, and S represents the allocation coefficient;

[0111] Medical staff deployment suggestions: if the bed occupancy rate exceeds expectations, additional medical staff will be added.

[0112] Among them, N dp Indicates the number of medical staff that need to be deployed, A b,t represents the actual bed utilization rate at time t, P b,t represents the predicted bed occupancy rate, and H represents the medical staff deployment coefficient, that is, the proportion of medical staff deployed according to the change in bed occupancy rate;

[0113] Procurement recommendations: If drug consumption is significantly higher than expected, drug procurement recommendations will be triggered.

[0114] Among them, Q cg Indicates the number of additional drugs that need to be purchased, A d,t represents the actual drug consumption at time t, P d,t represents the predicted demand for drugs, and C represents the purchase coefficient;

[0115] The PDCA closed-loop management method in step S3 also includes:

[0116] Threshold θ t Dynamically adjust based on real-time data and historical standard deviation,

[0117] θ t+1 =λ·θ t +(1-λ)·σ t+1 , where θ t+1 represents the dynamic threshold at the next moment, λ represents the adjustment coefficient, σ t+1 Indicates the forecast standard deviation for the next moment;

[0118] Specifically, the "check" and "action" links in the PDCA closed-loop management are combined with the automated feedback mechanism to automatically compare multi-dimensional analysis prediction data with actual operation data. When operational abnormalities, such as drug consumption exceeding expectations, equipment utilization rate being too low, and a surge in bed occupancy rate, are detected, the system will automatically generate corresponding adjustment suggestions and push them to relevant managers, greatly shortening the time it takes to expose and resolve problems, allowing hospitals to quickly adjust their configurations when resources are tight and avoid further expansion of operational risks.

[0119] Step S4, operating cost control,

[0120] Based on the abnormal operation data fed back in step S3, the operation cost is controlled and the operation cost is allocated to departments, projects or equipment;

[0121] The operating cost control methods are:

[0122] Using activity-based costing (ABC), we break down costs into individual items in a department, including equipment usage, drug consumption, and medical staff hours for specific surgeries. Based on real-time operational data, we reallocate resources to departments with high demand.

[0123] Step S4 uses activity-based costing ABC to allocate costs:

[0124] Use nonlinear allocation function to refine cost allocation. Among them, C ijk represents the cost allocation of the jth project of the i-th department in the kth resource category, α represents the adjustment coefficient of activity cost allocation, which is used to adjust the cost allocation ratio between various projects, and V ijk represents the k-th resource usage of the j-th project in the i-th department, including drug consumption, equipment usage time, and staff hours. represents the total resource usage of the i-th department in all m projects, β represents the nonlinear weight coefficient of resource allocation, and adjusts the sensitivity of different resources to cost allocation. i It represents the total indirect cost of the i-th department, which is allocated to each project. The nonlinear weight β is used to refine the cost allocation for large-scale resource consumption, and projects with large resource usage will allocate more costs;

[0125] Resource consumption is broken down into multiple dimensions, including equipment, personnel, medicines, and consumables:

[0126] Among them, C ij represents the total shared cost of the jth project in the i-th department, λ k represents the weight of the k-th resource, V ijk represents the actual consumption of the kth resource in the jth project of the ith department, β k represents the nonlinear allocation coefficient of the k-th resource, C i represents the total indirect cost of the i-th department, It represents the total consumption of the kth resource of all projects in the ith department, which is used for standardized allocation;

[0127] Dynamically adjust cost allocation based on actual consumption data,

[0128] in, represents the apportioned cost at the next moment t+1, η represents the smoothing coefficient, determines the weight ratio of the current moment and the historical apportioned data, and controls the adjustment speed. Indicates the resource consumption at the next time t+1, combines the historical allocation data with the new resource consumption, and gradually adjusts the cost allocation results;

[0129] After the allocation is completed, the cost efficiency of each department is evaluated to measure the matching degree between resource consumption and output.

[0130] Among them, E ij represents the cost efficiency coefficient of the jth project of the i-th department, O ij represents the j-th project output of the i-th department, including surgical success rate and patient satisfaction, C ij represents the total shared cost of the jth project in the i-th department; the cost efficiency coefficient E ij Used to measure the efficiency of resource utilization;

[0131] Specifically, by adopting the activity-based costing method ABC, operating costs are broken down into individual items in each department, such as the use of equipment for specific surgeries, drug consumption, and medical staff hours. A more refined cost allocation method can more accurately reflect the actual cost of various resources, and can also help hospitals identify high-cost links, thereby more effectively controlling costs.

[0132] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A hospital operation analysis and target management system based on big data, characterized by: include, Data collection module, used for operation data collection. The data sources are IoT devices, EMR and operation systems. Operation data include financial data, patient flow, equipment utilization, drug consumption, bed utilization, equipment operation status and patient satisfaction. A data dashboard that displays the hospital's key operating indicators, including bed occupancy rate, equipment operating status, and patient satisfaction; An automated feedback system triggers adjustment suggestions when operational anomalies are identified; Operation cost control module, used for operation cost control, allocating operation costs to departments, projects or equipment; The operating cost control module adopts activity-based costing (ABC) to break down costs into individual items in a department, including equipment usage, drug consumption, and medical staff hours for specific surgeries. Based on real-time operational data, resources are reallocated to high-demand departments.

2. A hospital operation analysis and target management method based on big data, characterized in that: Based on the big data-based hospital operation analysis and target management system described in claim 1, The following steps are included: Step S1, real-time data collection, collecting operational data of the hospital's operational links based on the hospital's IoT devices, electronic medical record system EMR and operation system, and integrating the operational data, and updating the operational data in real time; Step S2, multi-dimensional analysis of operational data, based on historical financial data, patient flow, drug consumption and equipment utilization rate, multi-dimensional analysis is performed, and a dynamic visual data dashboard is used to display key hospital operational indicators in real time, including bed utilization rate, equipment operation status and patient satisfaction; Step S3, PDCA closed-loop management, based on the multi-dimensional analysis results of step S2, automatic feedback is provided. The results of the multi-dimensional analysis, i.e., the predicted data, are compared with the actual operation data using an automated feedback system. When an operation abnormality is found, adjustment suggestions are automatically triggered. Step S4, operating cost control, performs operating cost control based on the abnormal operating data fed back in step S3, and allocates the operating cost to departments, projects or equipment.

3. The hospital operation analysis and target management method based on big data according to claim 2 is characterized in that: Multidimensional analysis includes: Resource demand forecasting, based on bed occupancy rate and seasonal patient peak data, predicts subsequent resource demand, including drug consumption and bed occupancy rate; Financial analysis and forecasting: Based on historical financial data, forecast subsequent resource consumption trends, i.e. changes in financial data; Operational efficiency analysis compares resource usage of different departments, including equipment operating status and drug consumption.

4. The hospital operation analysis and target management method based on big data according to claim 3 is characterized in that: The resource demand prediction method in step S2 is: Forecasting based on time series analysis combined with seasonal models, Among them, R t+k represents the resource demand forecast value at time t+k, that is, the future bed utilization rate or drug demand, α represents the weight coefficient of historical resource demand for future prediction, N represents the time window size used to calculate the historical average value, that is, the historical resource demand data from time t-1 to time tN, R t-i represents the resource demand at the ti-th moment, i.e., the i-th historical data point, β represents the influence coefficient of patient flow on future resource demand, M represents the time window size used to calculate the average patient flow, i.e., the historical patient flow data from t-1 to tM, P t-j represents the patient flow at time tj, γ represents the influence coefficient of seasonal fluctuation, S t represents the seasonal factor at the current time t, ∈ represents noise; Financial analysis and forecasting is based on historical financial data, combined with resource consumption and equipment usage, to predict future financial expenditure trends. Among them, C t+k represents the predicted value of financial expenditure at the t+kth moment, δ represents the weight coefficient of historical financial expenditure on future prediction, Q represents the time window size used to calculate the average value of historical financial expenditure, and C t-i represents the financial expenditure data at the ti-th moment, η represents the influence coefficient of equipment utilization rate on financial expenditure, U t represents the equipment utilization rate at the current time t, that is, the equipment workload of each department of the hospital, μ represents the impact coefficient of drug consumption on financial expenditure, and D t represents the drug consumption at the current time t, ν represents the influence coefficient of human resource allocation on financial expenditure, E t represents the staffing situation at the current time t, that is, human resource input, and ξ represents noise.

5. The hospital operation analysis and target management method based on big data according to claim 4 is characterized in that: The operational efficiency analysis in step S2 is used to identify inefficient resource usage, including low equipment utilization in a department and unreasonable consumption of certain drugs. Relative efficiency analysis is used to measure the relative operational efficiency of the department. Among them, E i represents the operational efficiency index of the i-th department, reflecting the efficiency of the department’s resource utilization, R i represents the resource usage of the i-th department, which refers to the bed utilization rate or equipment utilization rate, P i represents the patient flow of the i-th department, U i represents the equipment utilization rate of the i-th department, reflecting the ratio of the actual working time of equipment use to the available working time, C i represents the financial expenditure of the i-th department, D i represents the drug consumption of the i-th department, E i represents the staffing situation of the i-th department, and w1, w2, w3, w4, w5, and w6 represent the weight parameters of each operating factor.

6. The hospital operation analysis and target management method based on big data according to claim 5 is characterized in that: Operational anomalies include drug consumption, equipment utilization, and bed utilization exceeding expectations, and resource utilization of a department exceeding the budget; The automated feedback system automatically generates adjustment suggestions, including redeploying medical staff, replenishing drug inventory, and opening new beds, and pushes them to relevant managers.

7. The hospital operation analysis and target management method based on big data according to claim 6 is characterized in that: The PDCA closed-loop management method in step S3 is: Calculate the difference between actual operating data and forecast data, and use weighted difference analysis to compare resource requirements in multiple dimensions, including beds, medicines, and equipment usage. Where ΔX t Represents the overall operation deviation at time t, which is used to measure the deviation degree of multi-dimensional operation data. i,t represents the actual operation data of the i-th dimension, including the values ​​of bed utilization rate, drug consumption, and equipment utilization rate at time t, P i,t represents the forecast data of the i-th dimension, that is, the forecast result at the t-th time obtained in the multidimensional analysis, w i Represents the weight of the i-th dimension data, that is, the contribution of this dimension to the overall deviation; Using the standard deviation deviation model, set a predetermined threshold, when the overall operation deviation ΔX t When this threshold is exceeded, |ΔX t |>θ t , then anomaly detection is triggered, θ t represents the dynamically adjusted threshold, θ t Set θ based on the standard deviation of historical data t =σ·k, where σ is the standard deviation of historical operating data and k is the threshold coefficient; When an operational anomaly is detected, adjustment suggestions are automatically triggered. Different adjustment measures are triggered for anomalies in different dimensions: Resource allocation suggestions: If resource consumption exceeds expectations, resource allocation suggestions will be triggered. Among them, R dp A represents the amount of resources that need to be deployed, i.e., the amount of additional beds or drug inventory. r,t represents the actual resource consumption value at time t, P r,t It represents the resource forecast value, i.e., the drug or bed demand forecast result in multidimensional analysis, and S represents the allocation coefficient; Medical staff deployment suggestions: if the bed occupancy rate exceeds expectations, additional medical staff will be added. Among them, N dp Indicates the number of medical staff that need to be deployed, A b,t represents the actual bed utilization rate at time t, P b,t represents the predicted bed occupancy rate, and H represents the medical staff deployment coefficient, that is, the proportion of medical staff deployed according to the change in bed occupancy rate; Procurement recommendations: If drug consumption is significantly higher than expected, drug procurement recommendations will be triggered. Among them, Q cg Indicates the number of additional drugs that need to be purchased, A d,t represents the actual drug consumption at time t, P d,t represents the predicted drug demand, and C represents the procurement coefficient.

8. The hospital operation analysis and target management method based on big data according to claim 7 is characterized in that: The PDCA closed-loop management method in step S3 also includes: Threshold θ t Dynamically adjust based on real-time data and historical standard deviation, θ t+1 =λ·θ t +(1-λ)·σ t+1 , where θ t+1 represents the dynamic threshold at the next moment, λ represents the adjustment coefficient, σ t+1 Represents the forecast standard deviation for the next moment.

9. The hospital operation analysis and target management method based on big data according to claim 8, characterized in that: The operating cost control methods are: Using activity-based costing (ABC), costs are broken down into individual items in a department, including equipment usage for specific surgeries, drug consumption, and medical staff hours. Resources are reallocated to high-demand departments based on real-time operational data.

10. The hospital operation analysis and target management method based on big data according to claim 9, characterized in that: Step S4 uses activity-based costing ABC to allocate costs: Use nonlinear allocation function to refine cost allocation. Among them, C ijk represents the cost allocation of the jth project of the i-th department in the kth resource category, α represents the adjustment coefficient of activity cost allocation, which is used to adjust the cost allocation ratio between various projects, and V ijk represents the k-th resource usage of the j-th project in the i-th department, including drug consumption, equipment usage time, and staff hours. represents the total resource usage of the i-th department in all m projects, β represents the nonlinear weight coefficient of resource allocation, and adjusts the sensitivity of different resources to cost allocation. i represents the total indirect cost of the i-th department, allocated to each project; Resource consumption is broken down into multiple dimensions, including equipment, personnel, medicines, and consumables: Among them, C ij represents the total shared cost of the jth project in the i-th department, λ k represents the weight of the k-th resource, V ijk represents the actual consumption of the kth resource in the jth project of the ith department, β k represents the nonlinear allocation coefficient of the k-th resource, C i represents the total indirect cost of the i-th department, It represents the total consumption of the kth resource of all projects in the ith department, which is used for standardized allocation; Dynamically adjust cost allocation based on actual consumption data, in, represents the apportioned cost at the next moment t+1, η represents the smoothing coefficient, determines the weight ratio of the current moment and the historical apportioned data, and controls the adjustment speed. Indicates the resource consumption at the next time t+1; After the allocation is completed, the cost efficiency of each department is evaluated to measure the matching degree between resource consumption and output. Among them, E ij represents the cost efficiency coefficient of the jth project of the i-th department, O ij represents the j-th project output of the i-th department, including surgical success rate and patient satisfaction, C ij Represents the total shared cost of the jth project in the ith department.

Images