CN112382375A - Hospital doctor seeing process optimization method based on difference detection - Google Patents

Hospital doctor seeing process optimization method based on difference detection Download PDF

Info

Publication number
CN112382375A
CN112382375A CN202011172348.0A CN202011172348A CN112382375A CN 112382375 A CN112382375 A CN 112382375A CN 202011172348 A CN202011172348 A CN 202011172348A CN 112382375 A CN112382375 A CN 112382375A
Authority
CN
China
Prior art keywords
optimization
candidate
flow
execution time
candidate optimization
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011172348.0A
Other languages
Chinese (zh)
Other versions
CN112382375B (en
Inventor
王佳星
周武源
杨旭升
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN202011172348.0A priority Critical patent/CN112382375B/en
Publication of CN112382375A publication Critical patent/CN112382375A/en
Application granted granted Critical
Publication of CN112382375B publication Critical patent/CN112382375B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Development Economics (AREA)
  • Epidemiology (AREA)
  • Game Theory and Decision Science (AREA)
  • Biomedical Technology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Primary Health Care (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A hospital visit procedure optimization method based on difference detection comprises the following steps: carrying out difference detection on two aspects of structure and behavior on two different versions of the hospital visit flow; connecting the same parts of the hospital visit processes of two different versions and the different parts at different positions to establish all possible candidate optimization processes; calculating the optimization quality, the longest execution time and the time stability rate of each candidate optimization flow, filtering flows of which the optimization quality and the time stability rate are smaller than a given value and the longest execution time is larger than the given value, and adding the rest optimization flows into a candidate set; selecting a process with the minimum cost in the candidate set; and if the cost of the selected candidate flow with the lowest cost is less than the cost of two flows, the flow with the lowest cost is the optimal optimized flow, otherwise, the optimized flow is failed to create. The invention automatically optimizes the process by utilizing the existing process resources, thereby saving the time and the cost for optimizing the process while ensuring the treatment time of the patient.

Description

Hospital doctor seeing process optimization method based on difference detection
Technical Field
The invention belongs to the field of business process management, relates to a difference detection and optimization method among process models, and particularly relates to a hospital visit process optimization method based on difference detection.
Background
With the development of Business Process Management (BPM) technology, more and more enterprises and organizations adopt Business Process Improvement (BPI) technology to identify repeated work in a Process, and unreasonable places for structure and resource allocation, so as to improve the Process, so that the Process is executed more efficiently, the cost is saved, and the customer requirements can be met. The traditional BPI method optimizes the process in a manual or semi-automatic mode, namely, a process optimization expert finds out parts needing optimization in the process by means of own experience and professional knowledge and optimizes the parts manually. However, as the market demand changes, the process has become more and more complex, i.e. the variety of nodes is increasing, and the control flow structure becomes more and more complex. The manual or semi-automatic method is adopted to improve the process, so that the time and the labor are consumed, the cost is high, and errors are easy to occur. Thus, there is a strong need for an automated technique to optimize the process.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a hospital clinic procedure optimization method based on difference detection.
In order to automatically, efficiently, accurately and quickly optimize the hospital clinic visit process, the invention provides a hospital clinic visit process optimization method based on difference detection. The method comprises the steps of carrying out difference detection on flows of two different versions of a hospital visiting flow, constructing all possible candidate optimization flows based on difference part combinations of different difference positions, designing optimization indexes to filter the candidate optimization flows which do not meet requirements, finding out the flow with the minimum execution cost and the cost less than that of the flow of the existing version from the rest candidate optimization flows as the optimal optimization flow, breaking through the limitations of manual or semi-automatic flow optimization, long time consumption, high cost and easy error of the existing method, carrying out automatic flow optimization by using existing flow resources, greatly saving the time and cost of flow optimization while guaranteeing the visiting time of patients, and improving the actual availability.
A hospital clinic visit process optimization method based on difference detection comprises the following steps:
(1) two different versions of the procedure for hospital visits are entered: a first treatment flow and a second treatment flow;
(2) the difference detection is carried out on the first treatment process and the second treatment process, namely different parts in the first treatment process and the second treatment process are found out and are described by using a difference mode, and the steps specifically comprise the following steps:
(2.1) structural difference detection, namely finding out nodes which only appear in one process but not in the other process;
and (2.2) detecting the behavior difference, namely finding out the change of the execution relationship between the two nodes mapped in the first visit process and the second visit process, wherein the execution relationship comprises sequence, selection, parallel and circulation.
(3) If there are n differences between the first and second treatment procedures, one difference will correspond to 2 different parts in the two procedures, and 2 can be established by connecting the same parts and different parts at all positions of the first and second treatment proceduresn-2 candidate optimization procedures, wherein the 2 procedures subtracted are visit procedure one and visit procedure two;
(4) judging whether unprocessed candidate optimization flows exist or not, if not, jumping to the step (11), otherwise, selecting an unprocessed candidate optimization flow, and calculating the optimization quality of the candidate optimization flow, wherein the specific steps are as follows:
(4.1) representing the set of capabilities required for each optimization segment as
Figure BDA0002747686460000021
Each capability therein
Figure BDA0002747686460000022
Expressed as a triangular blur number:
Figure BDA0002747686460000023
again, the set of capabilities each optimizer has is represented as
Figure BDA0002747686460000024
Each capability therein
Figure BDA0002747686460000025
Expressed as a triangular fuzzy number:
Figure BDA0002747686460000026
(4.2) a certain capability required for an optimization part
Figure BDA0002747686460000027
Having the capability of corresponding to an optimizer
Figure BDA0002747686460000028
Degree of match therebetween
Figure BDA0002747686460000029
The calculation is performed using equation 1:
Figure BDA00027476864600000210
(4.3) A set of capabilities required by the optimization section
Figure BDA00027476864600000211
And optimizing the set of abilities possessed by the personnel
Figure BDA00027476864600000212
Degree of match therebetween
Figure BDA00027476864600000213
The calculation is performed using equation 2:
Figure BDA0002747686460000031
(4.4) the Quality of optimization (QoI) of the candidate optimization process represents the average value of the maximum matching degrees corresponding to all optimized parts in the candidate optimization process, and the average value can be used as a common value
Equation 3 is calculated:
Figure BDA0002747686460000032
wherein,
Figure BDA0002747686460000033
presentation selection and optimization component
Figure BDA0002747686460000034
Process optimizer with highest matching degree
Figure BDA0002747686460000035
Come to right
Figure BDA0002747686460000036
Optimizing;
(5) judging whether the optimization quality of the candidate optimization process is greater than a given value, if so, jumping to the step (6), otherwise, executing the step (4);
(6) calculating the longest execution time of the candidate optimization process, and the specific steps are as follows:
(6.1) extracting the basic path set ip ═ ip of the candidate optimization process1,ip2,…,ipMThat is, any path can be selected from the start node to the end node of the candidate optimization flow to traverse, but each path at least comprises a node or an edge which is not used by the defined path;
(6.2) each task node in the candidate optimization process can be executed by a plurality of workers, corresponding execution time and cost are recorded in a resource allocation table, the time spent by a certain worker s for executing the task node task is t, the cost is c, and the cost can be expressed as < s, task, t, c >;
(6.3) calculating the longest execution time of each basic path in the candidate optimization flow, and the steps are as follows:
(6.3.1) for each task node in the basic path, finding out the longest execution time of the task node in the resource allocation table;
(6.3.2) adding the longest execution time of all task nodes in the basic path to obtain the longest execution time of the basic path;
(6.4) selecting the longest execution time from all the basic paths of the candidate optimization flow to represent the longest execution time of the candidate optimization flow;
(7) judging whether the longest execution time of the candidate optimization process is smaller than a given value, if so, jumping to the step (8), otherwise, executing the step (4);
(8) calculating the time stability rate of the candidate optimization process, which comprises the following specific steps:
(8.1) for each basic path of the candidate optimization process, calculating different allocations of different workers to different task nodes for each basic path by using a dynamic programming algorithm according to a resource allocation table to obtain execution time T ═ T corresponding to N different resource allocations1,t2,…,tN}；
(8.2) the time stability rate Φ (T) of the candidate optimization procedure can be calculated using equation 4:
Figure BDA0002747686460000041
wherein avg (T) represents the average execution time of all elements in the execution time set T, and formula 4 represents that the time stability rate of the candidate flow is measured based on the fluctuation degree of the execution time of the independent path under different resource allocations relative to the average execution time;
(9) judging whether the time stability rate of the candidate optimization process is greater than a given value, if so, jumping to the step (10), otherwise, executing the step (4);
(10) adding the candidate optimization flow to the candidate set, and jumping to the step (4);
(11) selecting a candidate optimization flow with the minimum cost from the candidate set;
(12) and judging whether the cost of the candidate optimization flow with the minimum cost is less than that of the first visit flow and the second visit flow, if so, taking the candidate optimization flow with the minimum cost as the optimal optimization flow, and otherwise, failing to establish the optimization flow.
The technical conception of the invention is as follows: the hospital clinic flow is often provided with a plurality of different versions, difference detection is carried out on the clinic flows of the two different versions, different combinations of the same part and the difference part between the two versions are connected, a candidate optimization flow is constructed, and the existing flow resources are fully utilized. Three optimization indexes are designed: and optimizing quality, the longest execution time and the time stability rate, filtering the candidate optimization flows, selecting the flow with the minimum cost which is less than the cost of the two existing versions of flows from the rest candidate optimization flows as the optimal optimization flow, and automatically constructing the optimization flow, so that the method has wider application range and higher efficiency.
The invention (1) optimizes the flow according to the difference of two aspects of structure and behavior; (2) the triangular fuzzy number is adopted to represent the capability level, and the matching degree is calculated based on the capability level required by the flow optimization part and the triangular fuzzy number corresponding to the capability level owned by the optimization personnel, so that the optimization personnel with the highest matching degree is selected to optimize the corresponding part; (3) extracting a basic path of the candidate optimization flow, and obtaining the longest execution time of the candidate optimization flow based on the longest execution time of the independent path; (4) and finding out all resource allocation conditions of the independent paths of the candidate optimization flows by using a dynamic programming algorithm, and further measuring the time stability rate of the candidate flows based on the fluctuation degree of the execution time of the independent paths under different resource allocation relative to the average execution time.
The invention has the advantages that: the method breaks through the limitations that the existing method manually or semi-automatically optimizes the process, is long in time consumption, high in cost and easy to make mistakes, utilizes the existing process resources to automatically optimize the process, greatly saves the time and cost for optimizing the process while ensuring the time for patients to see a doctor, and improves the actual usability.
Drawings
FIG. 1 is a general flow diagram of the present invention.
Fig. 2 a-2 b are flow charts of two different versions of a hospital visit procedure: wherein fig. 2a is a flow chart of a first visit procedure, and fig. 2b is a flow chart of a second visit procedure.
Fig. 3a to 3b are flow structure tree diagrams corresponding to two different versions of hospital visit flows, wherein fig. 3a is a TPST converted from the first visit flow, and fig. 3b is a TPST converted from the second visit flow.
Fig. 4 is a flowchart of a fourth candidate optimization procedure.
FIG. 5 is a flow chart of the optimization: and a second candidate optimization process.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
Example one
Reference is made to FIG. 1
A hospital clinic visit process optimization method based on difference detection comprises the following steps:
(1) two different versions of the procedure for hospital visits are entered: a first treatment flow and a second treatment flow;
(2) the difference detection is carried out on the first treatment process and the second treatment process, namely different parts in the first treatment process and the second treatment process are found out and are described by using a difference mode, and the steps specifically comprise the following steps:
(2.1) structural difference detection, namely finding out nodes which only appear in one process but not in the other process;
and (2.2) detecting the behavior difference, namely finding out the change of the execution relationship between the two nodes mapped in the first visit process and the second visit process, wherein the execution relationship comprises sequence, selection, parallel and circulation.
(3) If there are n differences between the first and second treatment procedures, one difference will correspond to 2 different parts in the two procedures, and 2 can be established by connecting the same parts and different parts at all positions of the first and second treatment proceduresn-2 candidate optimization procedures, wherein the 2 procedures subtracted are visit procedure one and visit procedure two;
(4) judging whether unprocessed candidate optimization flows exist or not, if not, jumping to the step (11), otherwise, selecting an unprocessed candidate optimization flow, and calculating the optimization quality of the candidate optimization flow, wherein the specific steps are as follows:
(4.1) representing the set of capabilities required for each optimized portion as
Figure BDA0002747686460000071
Each capability therein
Figure BDA0002747686460000072
Expressed as a triangular blur number:
Figure BDA0002747686460000073
again, the set of capabilities each optimizer has is represented as
Figure BDA0002747686460000074
Each capability therein
Figure BDA0002747686460000075
Expressed as a triangular fuzzy number:
Figure BDA0002747686460000076
(4.2) a certain capability required for an optimization part
Figure BDA0002747686460000077
Having the capability of corresponding to an optimizer
Figure BDA0002747686460000078
Degree of match therebetween
Figure BDA0002747686460000079
The calculation is performed using equation 1:
Figure BDA00027476864600000710
(4.3) A set of capabilities required by the optimization section
Figure BDA00027476864600000711
And optimizing the set of abilities possessed by the personnel
Figure BDA00027476864600000712
Degree of match therebetween
Figure BDA00027476864600000713
The calculation is performed using equation 2:
Figure BDA00027476864600000714
(4.4) the Quality of optimization (QoI) of the candidate optimization process represents the average value of the maximum matching degrees corresponding to all optimized parts in the candidate optimization process, and the average value can be used as a common value
Equation 3 is calculated:
Figure BDA00027476864600000715
wherein,
Figure BDA00027476864600000716
presentation selection and optimization component
Figure BDA00027476864600000717
Process optimizer with highest matching degree
Figure BDA00027476864600000718
Come to right
Figure BDA00027476864600000719
Optimizing;
(5) judging whether the optimization quality of the candidate optimization process is greater than a given value, if so, jumping to the step (6), otherwise, executing the step (4);
(6) calculating the longest execution time of the candidate optimization process, and the specific steps are as follows:
(6.1) extracting the basic path set ip ═ ip of the candidate optimization process1,ip2,…,ipMThat is, any path can be selected from the starting node to the ending node of the candidate optimization process to traverseBut each path should contain at least one node or edge that the defined path has not used;
(6.2) each task node in the candidate optimization process can be executed by a plurality of workers, corresponding execution time and cost are recorded in a resource allocation table, the time spent by a certain worker s for executing the task node task is t, the cost is c, and the cost can be expressed as < s, task, t, c >;
(6.3) calculating the longest execution time of each basic path in the candidate optimization flow, and the steps are as follows:
(6.3.1) for each task node in the basic path, finding out the longest execution time of the task node in the resource allocation table;
(6.3.2) adding the longest execution time of all task nodes in the basic path to obtain the longest execution time of the basic path;
(6.4) selecting the longest execution time from all the basic paths of the candidate optimization flow to represent the longest execution time of the candidate optimization flow;
(7) judging whether the longest execution time of the candidate optimization process is smaller than a given value, if so, jumping to the step (8), otherwise, executing the step (4);
(8) calculating the time stability rate of the candidate optimization process, which comprises the following specific steps:
(8.1) for each basic path of the candidate optimization process, calculating different allocations of different workers to different task nodes for each basic path by using a dynamic programming algorithm according to a resource allocation table to obtain execution time T ═ T corresponding to N different resource allocations1,t2,…,tN}；
(8.2) the time stability rate Φ (T) of the candidate optimization procedure can be calculated using equation 4:
Figure BDA0002747686460000081
wherein avg (T) represents the average execution time of all elements in the execution time set T, and formula 4 represents that the time stability rate of the candidate flow is measured based on the fluctuation degree of the execution time of the independent path under different resource allocations relative to the average execution time;
(9) judging whether the time stability rate of the candidate optimization process is greater than a given value, if so, jumping to the step (10), otherwise, executing the step (4);
(10) adding the candidate optimization flow to the candidate set, and jumping to the step (4);
(11) selecting a candidate optimization flow with the minimum cost from the candidate set;
(12) and judging whether the cost of the candidate optimization flow with the minimum cost is less than that of the first visit flow and the second visit flow, if so, taking the candidate optimization flow with the minimum cost as the optimal optimization flow, and otherwise, failing to establish the optimization flow.
Example two
Reference is made to figures 2, 3, 4, 5 and tables 1, 2, 3, 4
Figure 2 shows the flow of two different versions of the hospital visit flow: the first and second treatment flows, wherein the rectangular nodes represent task nodes, the circular nodes represent gateway nodes, and there are 6 types: "And-split", "And-join", "Xor-split", "Xor-join", "Loop-split", And "Loop-join". A hospital visit procedure contains 4 different types of control flow structures: sequential, selective, parallel, and cyclic. Each task node in the sequence structure only has one incoming edge and one outgoing edge; selecting a structure starting from 'Xor-split' and ending at 'Xor-join', the structure comprising a plurality of paths, wherein only one path can be executed; the parallel structure starts from 'Ant-split' And ends from 'Ant-join', And task nodes in the structure can be executed simultaneously; the Loop structure starts at "Loop-split" and ends at "Loop-join", and the task nodes in the structure can be executed multiple times. Therefore, in the first visit flow of fig. 2, task node J and task node M are executed in parallel; in the second visit process, the task node B and the task node D are executed circularly (the present invention specifies that their execution times is 2), and the task node H and the task node G are executed selectively, that is, only one of them will be executed.
FIG. 3 shows a Task-based Process Structure Tree (TPST) transformed from the two different versions of the hospital visit flow of FIG. 2. The leaf nodes in one TPST correspond to task nodes of one process, the intermediate nodes represent control flow structures in a process model, And the intermediate nodes have four types of sequences, ands, xors And Loop, And respectively correspond to sequences, parallels, selections And Loop structures, wherein the task nodes in the sequences And Loop structures are sequentially executed, And the task nodes in the parallels And Loop structures are unordered in execution.
FIG. 4 shows one possible candidate optimization procedure created based on the differences between two different versions of a hospital visit procedure: and fourthly, candidate optimization flow. Figure 5 shows the optimal optimization procedure created based on the difference between the first and second visit procedures.
In order to optimize the hospital visit process, the following steps are required:
(1) two different versions of the procedure for hospital visits are entered: a first treatment flow and a second treatment flow;
(2) the difference detection is carried out on the first treatment process and the second treatment process, namely different parts in the first treatment process and the second treatment process are found out and are described by using a difference mode, and the steps specifically comprise the following steps:
(2.1) structural difference detection, wherein no structural difference exists between the first treatment process and the second treatment process because no node exists between the first treatment process and the second treatment process, wherein the node only exists in one process and does not exist in the other process;
(2.2) detecting the behavior difference, wherein 4 behavior differences are included between the first treatment process and the second treatment process, and correspond to 4 parts needing to be optimized: the first optimization part, the fourth optimization part, is highlighted by light blue, light green, light yellow and light purple respectively, and corresponds to a first difference corresponding part (difference one, difference five), a second difference corresponding part (difference two, difference six), a third difference corresponding part (difference three, difference seven) and a fourth difference corresponding part (difference four, difference eight) in the first visit procedure and the second visit procedure respectively. The optimization part I corresponds to the optimization part I (difference I, difference V), and indicates that the task node B and the task node D are sequentially executed in the first visit procedure and are sequentially executed in the second visit procedure; the second optimization part corresponds to (difference two, difference six), which means that the task node F is executed before the task node E in the first visit process, and the task node F is executed after the task node E in the second visit process; the optimization part III corresponds to the optimization part III (difference III, difference VII), and indicates that the task node H and the task node G are sequentially executed in the first treatment process and are selectively executed in the second treatment process; the optimization part four corresponds to (difference four, difference eight), which means that the task node J and the task node M are executed in parallel in the first visit procedure and are executed sequentially in the second visit procedure.
(3) The combination of connecting the same part of the first and second treatment procedures and the different parts at 4 positions can establish 24-2 ═ 14 candidate optimization flows: candidate optimization procedure one-candidate optimization procedure fourteen;
(4) judging whether unprocessed candidate optimization flows exist or not, if not, jumping to the step (11), otherwise, selecting an unprocessed candidate optimization flow, and calculating the optimization quality of the candidate optimization flow, wherein the specific steps are as follows:
(4.1) the capabilities required by each optimization part and the capabilities possessed by each optimization person in the first and second treatment processes are shown in table 1, the required capabilities are set as { analysis, calculation, design }, and each capability has 5 grades: "none", "low", "normal", "good", and "excellent", respectively, correspond to the triangular blur numbers "none" — (0,0,0.1), "low" — (0,0.3,0.5), "normal" — (0.3,0.5,0.7), "good" — (0.5, 0.7,0.9}, and "excellent" — (0.7, 0.9,1 };
(4.2) specific capabilities required for an optimization section
Figure BDA0002747686460000101
Having the capability of corresponding to an optimizer
Figure BDA0002747686460000111
The degree of matching md between them is calculated by equation 1:
Figure BDA0002747686460000112
as shown in table 1, taking the computing power of the first optimization part and the optimizer B as an example, the computing power requirement of the first optimization part is "good" ({ 0.5,0.7,0.9}, and the computing power possessed by the optimizer B is "excellent" ({ 0.7,0.9,1}, then the matching degree between the computing power of the first optimization part and the computing power of the optimizer B is:
Figure BDA0002747686460000113
the matching degree between the first optimized part and the design ability of the optimizer B is 1, and the matching degree between the first optimized part and the calculation ability of the optimizer A is 0, because the ability of the optimizer A does not reach the ability required by the first optimized part;
(4.3) A set of capabilities required by the optimization section
Figure BDA0002747686460000114
And optimizing the set of abilities possessed by the personnel
Figure BDA0002747686460000115
Degree of match therebetween
Figure BDA0002747686460000116
The calculation is performed using equation 2:
Figure BDA0002747686460000117
as shown in table 1, to optimize fraction I1For example, the ability matching degree with the optimizer B is the average value of the analysis, calculation and design ability matching degrees, namely
Figure BDA0002747686460000118
While optimizing part I1And optimizer AThe capability matching degree of (2) is 0, and the calculation result of the capability matching degree of each optimization part and each optimization person is shown in table 2;
TABLE 1 capability required by the optimization part of the process and capability tables owned by the optimizer
Figure BDA0002747686460000119
TABLE 2 Table of capability matchings between Process optimization and optimization personnel
Figure BDA00027476864600001110
Figure BDA0002747686460000121
(4.4) the Quality of optimization (QoI) of the candidate optimization process represents the average value of the maximum matching degrees corresponding to all optimized parts in the candidate optimization process, and the average value can be used as a common value
Equation 3 is calculated:
Figure BDA0002747686460000122
wherein,
Figure BDA0002747686460000123
presentation and optimization section
Figure BDA0002747686460000124
Process optimizer with highest matching degree
Figure BDA0002747686460000125
Selected to optimize the part
Figure BDA0002747686460000126
Optimizing;
taking the candidate optimization flow four in fig. 4 as an example, the optimization part related thereto includes an optimization part one and an optimization part three, so that the optimization person with the highest matching degree with the optimization part one and the optimization part three is selected to optimize the optimization person, as shown in table 2, the matching degrees are respectively task node D and task node C, the matching degrees are respectively 1 and 0.94, so that the optimization quality of the candidate optimization flow four is the average value of the two matching degrees, i.e. 0.97, and the optimization qualities of the remaining 13 candidate optimization flows are shown in table 4;
(5) judging whether the optimization quality of the candidate optimization process is greater than a given value, if so, jumping to the step (6), otherwise, executing the step (4), and if the given optimization quality is 0.93, filtering the candidate optimization process III and the candidate optimization process IV;
(6) calculating the longest execution time of the candidate optimization process, and the specific steps are as follows:
(6.1) extracting the basic path set ip ═ ip of the candidate optimization process1,ip2,…,ipMThe method comprises the following steps of (1) selecting any path from a start node to an end node of a candidate optimization flow to traverse, but each path at least comprises a node or an edge which is not used by a defined path, taking the candidate optimization flow four shown in fig. 4 as an example, the basic path set is { ABDBDFEHIJMP, ABDBDFEGIJMP };
(6.2) each task node in the candidate optimization process can be executed by a plurality of workers, and the corresponding execution time and cost are recorded in the resource allocation table, as shown in table 3, wherein the first row represents the task node, the first column represents the workers, and two numbers in each cell respectively represent the execution cost and time, and the unit is "minute" and "element";
(6.3) calculating each basic path ip in the candidate optimization flowiTaking the candidate optimization flow four shown in FIG. 4 as an example, the longest execution time of (1 ≦ i ≦ M) is as follows:
(6.3.1) for each task node in the first basic path "ABDBDFEHIJMP" of the candidate optimization procedure four shown in FIG. 4, find its longest execution time in the resource allocation table: task node A executes for 5 minutes, task node B executes for 1 minute, task node D executes for 1 minute, task node F executes for 2 minutes, task node E executes for 10 minutes, task node H executes for 20 minutes, task node I executes for 20 minutes, task node J executes for 10 minutes, task node M executes for 16 minutes, and task node P executes for 15 minutes;
table 3 resource allocation table
Figure BDA0002747686460000131
(6.3.2) adding the longest execution time of all task nodes in the first basic path of the candidate optimization flow four shown in fig. 4 to obtain the longest execution time of the basic path: 102 minutes, and likewise, the longest execution time of the second basic path is 95 minutes;
(6.4) taking max (102,95) as the longest execution time of the fourth candidate optimization procedure shown in fig. 4 as 102 minutes, and the longest execution times of the remaining 13 candidate optimization procedures are shown in table 4;
(7) judging whether the longest execution time of the candidate optimization process is smaller than a given value, if so, jumping to the step (8), otherwise, executing the step (4), and if the given longest execution time is 100 minutes, leaving the candidate optimization process I, the candidate optimization process II and the candidate optimization process fourteen, and filtering the rest candidate processes;
(8) calculating the time stability rate of the candidate optimization process, which comprises the following specific steps:
(8.1) for each basic path of the candidate optimization process, calculating different allocations of different workers to different task nodes by using a dynamic programming algorithm according to an allocation table to obtain N different execution times T ═ T1,t2,…,tN}；
Taking the fourth candidate optimization procedure shown in fig. 4 as an example, the first basic path has 8 resource allocation methods: "aaaacbbddbe", "aaaaaacbbbe", "aaaaacacbbe", "aaaaacacbdbbe", "aaaaaacbcbe", "aaaaacacbdbbe", "aaaaacacddbbe", "aaaaacaccdbe", and "aaaaaaacaccdebe", corresponding to an execution time of {97,92,102,97,92,87,97,92}, the second basic path has 4 resource allocation patterns in common: "aaaaacbbddbe", "aaaaacbdbe", "aaaaaccbbe", "aaaaaaccbdbbe", "aaaaaaccbbe", and the corresponding execution time "two is {95,90,90,85 };
(8.2)Pcthe time stabilization rate Φ (T) of (c) can be calculated by equation 4:
Figure BDA0002747686460000141
wherein avg (T) represents the average execution time of all elements in the execution time set T, and formula 4 represents that the time stability rate of the candidate flow is measured based on the fluctuation degree of the execution time of the independent path under different resource allocations relative to the average execution time;
taking the candidate process four shown in fig. 4 as an example, the average execution time of 12 resource allocation manners corresponding to the two basic paths is 93 minutes, according to formula 4, the time stability rate of the candidate process four is 0.945, and the time stability rates of the remaining 13 candidate optimization processes are shown in table 4;
(9) judging whether the time stability rate of the candidate process is greater than a given value, if so, jumping to the step (10), otherwise, executing the step (4), and if the given time stability rate is 0.9, leaving the candidate process I, the candidate process II and the candidate process fourteen;
(10) adding the candidate flow I, the candidate flow II and the candidate flow fourteen into the candidate set, and jumping to the step (4);
(11) selecting a candidate optimization flow with the lowest cost in the candidate set, wherein any flow can be selected as the costs of the candidate flow I, the candidate flow II and the candidate flow fourteen are the same;
(12) the cost of the first visit flow and the second visit flow is 256.5 and 219, respectively, and the cost of the first candidate flow, the second candidate flow and the fourteenth candidate flow is less than the cost of the first visit flow and the second visit flow, so any one of the first candidate flow, the second candidate flow and the fourteenth candidate flow can be selected as the best optimization flow, assuming that the second candidate flow is selected, as shown in fig. 5.
TABLE 4 index calculation tables corresponding to all candidate optimization processes
Figure BDA0002747686460000161

Claims (2)

1. A hospital clinic visit process optimization method based on difference detection comprises the following steps:
(1) two different versions of the procedure for hospital visits are entered: a first treatment flow and a second treatment flow;
(2) the difference detection is carried out on the first treatment process and the second treatment process, namely different parts in the first treatment process and the second treatment process are found out and are described by using a difference mode, and the steps specifically comprise the following steps:
(2.1) structural difference detection, namely finding out nodes which only appear in one process but not in the other process;
and (2.2) detecting the behavior difference, namely finding out the change of the execution relationship between the two nodes mapped in the first visit process and the second visit process, wherein the execution relationship comprises sequence, selection, parallel and circulation.
(3) If there are n differences between the first and second treatment procedures, one difference will correspond to 2 different parts in the two procedures, and 2 can be established by connecting the same parts and different parts at all positions of the first and second treatment proceduresn-2 candidate optimization procedures, wherein the 2 procedures subtracted are visit procedure one and visit procedure two;
(4) judging whether unprocessed candidate optimization flows exist or not, if not, jumping to the step (11), otherwise, selecting an unprocessed candidate optimization flow, and calculating the optimization quality of the candidate optimization flow, wherein the specific steps are as follows:
(4.1) representing the set of capabilities required for each optimization segment as
Figure FDA0002747686450000011
Figure FDA0002747686450000012
Each capability therein
Figure FDA0002747686450000013
Expressed as a triangular blur number:
Figure FDA0002747686450000014
again, the set of capabilities each optimizer has is represented as
Figure FDA0002747686450000015
Each capability therein
Figure FDA0002747686450000016
Expressed as a triangular fuzzy number:
Figure FDA0002747686450000017
(4.2) a certain capability required for an optimization part
Figure FDA0002747686450000018
Having the capability of corresponding to an optimizer
Figure FDA0002747686450000019
Degree of match therebetween
Figure FDA00027476864500000110
The calculation is performed using equation 1:
Figure FDA0002747686450000021
(4.3) A set of capabilities required by the optimization section
Figure FDA0002747686450000022
And optimizing the set of abilities possessed by the personnel
Figure FDA0002747686450000023
Degree of match therebetween
Figure FDA0002747686450000024
The calculation is performed using equation 2:
Figure FDA0002747686450000025
(4.4) the Quality of optimization (QoI) of the candidate optimization process represents the average of the maximum matching degrees corresponding to all optimized parts in the candidate optimization process, and can be calculated by using formula 3:
Figure FDA0002747686450000026
wherein,
Figure FDA0002747686450000027
presentation selection and optimization component
Figure FDA0002747686450000028
Process optimizer with highest matching degree
Figure FDA0002747686450000029
Come to right
Figure FDA00027476864500000210
Optimizing;
(5) judging whether the optimization quality of the candidate optimization process is greater than a given value, if so, jumping to the step (6), otherwise, executing the step (4);
(6) calculating the longest execution time of the candidate optimization process, and the specific steps are as follows:
(6.1) extracting the basic path set ip ═ ip of the candidate optimization process1,ip2,…,ipMI.e. from the candidateAny path can be selected from the starting node to the ending node of the optimization process for traversal, but each path at least comprises a node or an edge which is not used by the defined path;
(6.2) each task node in the candidate optimization process can be executed by a plurality of workers, corresponding execution time and cost are recorded in a resource allocation table, the time spent by a certain worker s for executing the task node task is t, the cost is c, and the cost can be expressed as < s, task, t, c >;
(6.3) calculating the longest execution time of each basic path in the candidate optimization flow, and the steps are as follows:
(6.3.1) for each task node in the basic path, finding out the longest execution time of the task node in the resource allocation table;
(6.3.2) adding the longest execution time of all task nodes in the basic path to obtain the longest execution time of the basic path;
(6.4) selecting the longest execution time from all the basic paths of the candidate optimization flow to represent the longest execution time of the candidate optimization flow;
(7) judging whether the longest execution time of the candidate optimization process is smaller than a given value, if so, jumping to the step (8), otherwise, executing the step (4);
(8) calculating the time stability rate of the candidate optimization process, which comprises the following specific steps:
(8.1) for each basic path of the candidate optimization process, calculating different allocations of different workers to different task nodes for each basic path by using a dynamic programming algorithm according to a resource allocation table to obtain execution time T ═ T corresponding to N different resource allocations1,t2,…,tN}；
(8.2) the time stability rate Φ (T) of the candidate optimization procedure can be calculated using equation 4:
Figure FDA0002747686450000031
wherein avg (T) represents the average execution time of all elements in the execution time set T, and formula 4 represents that the time stability rate of the candidate flow is measured based on the fluctuation degree of the execution time of the independent path under different resource allocations relative to the average execution time;
(9) judging whether the time stability rate of the candidate optimization process is greater than a given value, if so, jumping to the step (10), otherwise, executing the step (4);
(10) adding the candidate optimization flow to the candidate set, and jumping to the step (4);
(11) selecting a candidate optimization flow with the minimum cost from the candidate set;
(12) and judging whether the cost of the candidate optimization flow with the minimum cost is less than that of the first visit flow and the second visit flow, if so, taking the candidate optimization flow with the minimum cost as the optimal optimization flow, and otherwise, failing to establish the optimization flow.
2. The hospital visit procedure optimization method based on difference detection as claimed in claim 1, wherein: (1) optimizing the flow according to the difference of two aspects of the structure and the behavior; (2) and calculating the matching degree based on the capacity level required by the flow optimization part and the triangular fuzzy number corresponding to the capacity level owned by the optimization personnel, thereby selecting the optimization personnel with the highest matching degree to optimize the corresponding part. (3) Extracting a basic path of the candidate optimization flow, and obtaining the longest execution time of the candidate optimization flow based on the longest execution time of the independent path; (4) and finding out all resource allocation conditions of the independent paths of the candidate optimization flows by using a dynamic programming algorithm, and further measuring the time stability rate of the candidate flows based on the fluctuation degree of the execution time of the independent paths under different resource allocation relative to the average execution time.
CN202011172348.0A 2020-10-28 2020-10-28 Hospital visit flow optimization method based on difference detection Active CN112382375B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011172348.0A CN112382375B (en) 2020-10-28 2020-10-28 Hospital visit flow optimization method based on difference detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011172348.0A CN112382375B (en) 2020-10-28 2020-10-28 Hospital visit flow optimization method based on difference detection

Publications (2)

Publication Number Publication Date
CN112382375A true CN112382375A (en) 2021-02-19
CN112382375B CN112382375B (en) 2022-10-11

Family

ID=74577319

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011172348.0A Active CN112382375B (en) 2020-10-28 2020-10-28 Hospital visit flow optimization method based on difference detection

Country Status (1)

Country Link
CN (1) CN112382375B (en)

Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2542173A1 (en) * 2006-04-06 2007-10-06 Ibm Canada Limited - Ibm Canada Limitee System and method for application balanced scorecard optimizer
US20090319653A1 (en) * 2008-06-20 2009-12-24 International Business Machines Corporation Server configuration management method
CN103969525A (en) * 2014-04-23 2014-08-06 国家电网公司 Flexible net-in quality detection system and method for electrical equipment
CN104463354A (en) * 2014-11-21 2015-03-25 山东航天信息有限公司 Improving method of distributed inventory dispatching
CN106599545A (en) * 2016-11-21 2017-04-26 深圳市医诺智能科技发展有限公司 Dynamic management method of radiation therapy processes
AU2017202982B1 (en) * 2016-01-15 2017-06-01 Hro Holdings Pty Ltd A patient procedure schedule throughput optimiser supervised machine learning system
CN108983719A (en) * 2018-07-24 2018-12-11 昆明理工大学 The Optimization Scheduling of panel beating process in a kind of manufacture of computer cabinet
CN109146123A (en) * 2018-06-29 2019-01-04 中国电力科学研究院有限公司 A kind of multiple-energy-source comprehensive coordination effect evaluation method and system
CN109659013A (en) * 2018-11-28 2019-04-19 平安科技（深圳）有限公司 Illness point is examined and method for optimizing route, device, equipment and storage medium
EP3477652A1 (en) * 2017-10-30 2019-05-01 Koninklijke Philips N.V. Matching a subject to resources
CN109711658A (en) * 2018-11-09 2019-05-03 成都数之联科技有限公司 A kind of industrial production optimizing detection system and method
CN109919578A (en) * 2019-02-26 2019-06-21 浪潮金融信息技术有限公司 A kind of business processing flow optimization method based on Hybrid
CN110232972A (en) * 2019-06-18 2019-09-13 中国人民解放军海军军医大学 Navy hospital's Office visits system
AU2019101317A4 (en) * 2019-10-30 2019-12-05 Southeast University A Bi-level Game-Based Planning Framework for Distribution Networks with multiple Micro-girds
BR102018017070A2 (en) * 2018-08-21 2020-03-10 Paulo Coelho Vieira SYSTEM FOR PRODUCTION, DISTRIBUTION AND STORAGE OF NETWORK INFORMATION
TW202013305A (en) * 2018-09-18 2020-04-01 大陸商北京嘀嘀無限科技發展有限公司 Artificial intelligent systems and methods for predicting traffic accident locations
CN111564194A (en) * 2020-05-09 2020-08-21 广州仟金方网络科技有限公司 Method for optimizing treatment process

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2542173A1 (en) * 2006-04-06 2007-10-06 Ibm Canada Limited - Ibm Canada Limitee System and method for application balanced scorecard optimizer
US20090319653A1 (en) * 2008-06-20 2009-12-24 International Business Machines Corporation Server configuration management method
CN103969525A (en) * 2014-04-23 2014-08-06 国家电网公司 Flexible net-in quality detection system and method for electrical equipment
CN104463354A (en) * 2014-11-21 2015-03-25 山东航天信息有限公司 Improving method of distributed inventory dispatching
AU2017202982B1 (en) * 2016-01-15 2017-06-01 Hro Holdings Pty Ltd A patient procedure schedule throughput optimiser supervised machine learning system
CN106599545A (en) * 2016-11-21 2017-04-26 深圳市医诺智能科技发展有限公司 Dynamic management method of radiation therapy processes
EP3477652A1 (en) * 2017-10-30 2019-05-01 Koninklijke Philips N.V. Matching a subject to resources
CN109146123A (en) * 2018-06-29 2019-01-04 中国电力科学研究院有限公司 A kind of multiple-energy-source comprehensive coordination effect evaluation method and system
CN108983719A (en) * 2018-07-24 2018-12-11 昆明理工大学 The Optimization Scheduling of panel beating process in a kind of manufacture of computer cabinet
BR102018017070A2 (en) * 2018-08-21 2020-03-10 Paulo Coelho Vieira SYSTEM FOR PRODUCTION, DISTRIBUTION AND STORAGE OF NETWORK INFORMATION
TW202013305A (en) * 2018-09-18 2020-04-01 大陸商北京嘀嘀無限科技發展有限公司 Artificial intelligent systems and methods for predicting traffic accident locations
CN109711658A (en) * 2018-11-09 2019-05-03 成都数之联科技有限公司 A kind of industrial production optimizing detection system and method
CN109659013A (en) * 2018-11-28 2019-04-19 平安科技（深圳）有限公司 Illness point is examined and method for optimizing route, device, equipment and storage medium
CN109919578A (en) * 2019-02-26 2019-06-21 浪潮金融信息技术有限公司 A kind of business processing flow optimization method based on Hybrid
CN110232972A (en) * 2019-06-18 2019-09-13 中国人民解放军海军军医大学 Navy hospital's Office visits system
AU2019101317A4 (en) * 2019-10-30 2019-12-05 Southeast University A Bi-level Game-Based Planning Framework for Distribution Networks with multiple Micro-girds
CN111564194A (en) * 2020-05-09 2020-08-21 广州仟金方网络科技有限公司 Method for optimizing treatment process

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JING FAN 等: ""Detecting Difference between Process Models Based on the Refined Process Structure Tree"", 《MOBILE INFORMATION SYSTEMS》 *

Also Published As

Publication number Publication date
CN112382375B (en) 2022-10-11

Similar Documents

Publication Publication Date Title
CN109800936B (en) Scheduling method based on tree search and electronic device using the same
JP2018524733A (en) Data processing method, apparatus and system, program, and recording medium
RU2005136998A (en) EFFICIENT AND FLEXIBLE BUSINESS MODELING BASED ON STRUCTURED BUSINESS CHARACTERISTICS
Veerappa et al. Understanding clusters of optimal solutions in multi-objective decision problems
CN110515843B (en) Test case priority ordering method based on defect set and inverted index
Naderifar et al. A review on conformance checking technique for the evaluation of process mining algorithms
CN110390496A (en) A kind of workflow design method that adaptation business is complicated and changeable with tissue
CN111258735A (en) Deep learning task scheduling method supporting QoS (quality of service) perception of user
CN112382375B (en) Hospital visit flow optimization method based on difference detection
CN109146172B (en) Multi-subject modeling method for climate change coping strategy
CN111553610B (en) Scheduling method and system based on learning-forgetting effect
JP5545087B2 (en) Distributed control system test execution management device
CN113377683B (en) Software test case generation method, system, device, terminal, medium and application
In et al. From requirements negotiation to software architectural decisions
JP2003256206A (en) Test plan preparation support method for software system and test plan preparation support program
CN109039698A (en) Industry internet intelligent Service processing method, readable storage medium storing program for executing, terminal
López et al. Modeling life: a conceptual schema-centric approach to understand the genome
CN105049475A (en) High-efficiency data storage and optimization method and system for large-scale community
CN107909181A (en) Household electrical appliances after-sale service operation flow node analysis method based on big data
CN110659034B (en) Combined optimization deployment method, system and storage medium of cloud-edge hybrid computing service
CN104994136B (en) The data store optimization method and system of extensive community
CN106846136A (en) A kind of data comparison method and equipment
CN114095336B (en) End-to-end problem diagnosis method and system
CN114462900B (en) Method, device and equipment for splitting service active node
CN109240930A (en) A kind of collaboration test method of web application

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant