TWI410083B - Method for evaluating the performance of an internal network in an enterprise by fuzzy logic - Google Patents

Abstract

A method for evaluating the performance of an internal network in an enterprise is applied in a software. The software includes a network model. The network model has a plurality of nodes, and each the node corresponds to one of the persons of the internal network. Each the person is tested in capacity to get a test result. The test result is described depending on a first fuzzy set to define a grade of the node. The grade is transformed into a probability distribution range. Then, the probability distribution ranges corresponding to all nodes are transformed into a system reliability. The system reliability is transformed into a linguistic performance index depending on a second fuzzy set, and then the linguistic performance index is shown in a display.

Description

Method of using fuzzy logic to evaluate enterprise internal network performance

The present invention relates to an enterprise internal network performance evaluation method, and more particularly to a method for utilizing fuzzy logic to evaluate an enterprise internal network performance.

Enterprise Resource Planning (ERP) is a system that uses a new management model to transform the management model of the old corporate intranet. It provides a more efficient deployment of financial, human, information, technology, and hardware resources. Use, in order to more effectively complete the business needs of the enterprise.

Most of the researches in the study focus on how to successfully import ERP systems, that is, to explore the adjustment methods that enterprises must implement in the process of importing ERP systems, such as organizational changes, process changes, role changes of project members, import costs, and project considerations. The impact of the process, the upgrade of the ERP system, etc., and other changes that have a significant impact on the original ecology of the enterprise, thus exploring a strategic approach to successful introduction of the ERP system.

However, after an enterprise imports an ERP system to manage the internal network of its employees, it must still evaluate the effectiveness of the internal network of the enterprise. However, most of the general network performance evaluation methods are used to evaluate the performance of a network composed of hardware such as machines or roads, such as the Internet or a transportation network, which cannot effectively evaluate the internal enterprise composed of personnel. The effectiveness of the network. Therefore, it is necessary to provide an effective evaluation method for the enterprise operator to estimate the workload that the ERP system manages the internal network of the enterprise in a unit time, and then estimate the benefits brought by the ERP system.

The purpose of the present invention is to provide an enterprise internal network performance evaluation party. Law, which can be used to assess the performance of a network of people.

Other objects and advantages of the present invention will become apparent from the technical features disclosed herein.

In order to achieve one or a part or all of the above or other purposes, an embodiment of the present invention provides an enterprise internal network performance evaluation method, which is applied to a software program to evaluate the performance of an enterprise internal network. The software program includes a network model, a probability function table, a system reliability function, a first fuzzy set, and a second fuzzy set. The above network model includes a plurality of nodes to correspond to a plurality of employees of the internal network of the enterprise.

The method steps of the above embodiment are as follows: a test procedure is performed on the employee corresponding to each node to obtain a test result; the test result of the employee is described by the first fuzzy set to define a capability level of the node; The capability level is converted into a probability distribution range, and the probability distribution range is the capability level of the corresponding node; the system reliability value is used to perform an operation on the probability distribution range corresponding to all nodes to obtain a system reliability value; The fuzzy set performs a fuzzy logic operation on the system reliability value to obtain a linguistic performance indicator with a text description; and presents the linguistic performance indicator on a display screen.

The software program described above is stored in a recording medium, which is included in a computer system such as an enterprise resource planning system or a customer relationship management system, and is executed by a microprocessor. The steps of the method are described in more detail as follows: a network model is provided in the software program, and the network model corresponds to the internal network of the enterprise, and includes a plurality of nodes corresponding to the employees and forming a plurality of File flow path; each node defined in the network model has a file capacity value, and each file flow path has a file The flow value, wherein the file capacity value has a first operational relationship with the file flow value, and the group formed by the file capacity values of all nodes is defined as a capacity vector, and the group formed by the file flow values of all file flow paths The group is defined as a flow vector; an expected value of the total flow of the network model is determined, and the expected value is stored in the recording medium; a test procedure is performed on the employee corresponding to each node to obtain a test result; The fuzzy set describes the test result of the employee to define a capability level of the node; the probability level is converted into a probability distribution range by the probability function table, and the probability distribution range is the capability level of the corresponding node, and is stored in the recording medium; querying the recording medium The probability distribution range in which to select the maximum file capacity value of each node; execute a software program to read the expected value and the maximum file capacity value in the recording medium, and obtain all possible traffic vectors according to the first operational relationship Calculate all possible capacitances based on all possible traffic vectors and the first operational relationship a vector; a minimum capacity vector is selected from all possible capacity vectors; a system reliability function is provided, including a probability distribution range and a second operational relationship between the minimum capacity vector and a system reliability value; according to the second operation Relationship, calculate the system reliability value when the total flow of the network model can be successfully processed is not less than the expected value; perform a fuzzy logic operation on the system reliability value, and convert to obtain a language performance indicator corresponding to the system reliability value, and The linguistic performance indicators are presented on a display screen.

In a preferred embodiment, the step of performing a fuzzy logic operation on the system reliability value performs a fuzzy logic operation using an inverse function of the system reliability function to obtain a linguistic performance indicator.

In a preferred embodiment, the plurality of nodes includes a start node, a plurality of relay nodes, and a terminal node, and all file flow paths are from the start node. The point starts, passes through some relay nodes, and ends at the terminal node. The step of forming a plurality of file flow paths includes: defining at least one program sequence relationship between each node in the network model; and defining a plurality of file flow paths of the network model according to the program sequence relationship.

In a preferred embodiment, the operational relationship between the file capacity value and the file flow value includes: the sum of all file flow values of all file flow paths through the same node is not greater than the maximum file capacity value; and at the same time point, The sum of all file traffic values through all file flow paths of the same node is not greater than the file capacity value of the node at that time.

In a preferred embodiment, the step of performing a test procedure for each employee corresponding to each node includes: providing a plurality of test capacities for each employee to be tested; and recording the test results of the employee for each test capacity in the record In the medium; providing a plurality of member functions; establishing a probability function table in the software program, including the member functions, the test capacity, the capability levels, and the relationship between the probability distribution ranges; and, according to the test capacity and the capability level, The probability function table is queried to obtain a range of probability distributions corresponding to each node having different capability levels.

With the above embodiments, it is only necessary to know the file processing capability of each employee, and then the performance of the internal network of the enterprise can be evaluated, and a language performance indicator can be obtained.

The above and other technical contents, features and advantages of the present invention will be apparent from the following detailed description of the preferred embodiments.

The present invention relates to a method for evaluating effectiveness for evaluating an enterprise The effectiveness of the network, especially the enterprise internal network using Enterprise Resource Planning (ERP) or Customer Relationship Management (CRM). The internal network of the enterprise referred to in this paper consists of a number of employees and the program-by-sequence relationship between the employees and the documents.

The method of the present invention first classifies the employees in the internal network of the enterprise by their work ability level, for example, the words "excellent", "good", "normal", "bad", "failed" and the like are used to describe the employee's ability level. Input the relevant information of the employee's ability level into a computer system, and use the computer system to perform a fuzzy logic operation, and convert the relevant data of the employee's ability level into a Linguistic Performance Index with text narrative, for example Use "Declined", "Recertified", "Acceptable" and other words to assess the effectiveness of the company's internal network. Incidentally, the computer system for performing the performance evaluation method includes at least a software program stored in a recording medium and performing an operation thereof by a microprocessor.

Figure 1 shows the general process architecture of the performance evaluation method, which includes two data generation processes and a language performance indicator calculation process: 1. The process of generating a lower boundary point: this process must be based on the internal The network process establishes a network model and calculates the 1 boundary point of the network model based on the expected value d of the total traffic of the file that the operator can load on the internal network of the enterprise. .

2. The process of generating the employee's ability grade: After the relevant training for the employees, the employee test is conducted to obtain the work ability level of each employee in the internal network of the enterprise, which includes a probability distribution range, and its representative The employee can successfully handle all the possibilities of different file sizes.

3. The linguistic performance index calculation process: using the principle of displacement and fuzzy logic to calculate the bottom point and probability distribution range to generate linguistic performance indicators.

Figure 2A is a detailed flow chart for generating a bottom point, as detailed below.

First, the network model is defined in the software program to have a plurality of nodes b _i to simulate an enterprise internal process network, and corresponding to a plurality of employees (S201), and then define each two nodes in the network model. The possible program sequence relationship between b _i (S202), the sequence relationship of these programs may determine a plurality of file flow paths P _j (S203), after which the network model of Figure 3 can be obtained, and the file will be These file flow paths P _j are transferred for processing.

After the file flow path P _j is determined, the expected value d of the total flow of the enterprise process network set by the operator is assigned to the file flow paths P _j (S204), and each file flow path P _j is allocated. File traffic values are represented by f _j . That is, it is preferable to find the file flow value f _j corresponding to all the file flow paths P _j , and it is preferable to determine the file flow value corresponding to each file flow path P _j according to a maximum file capacity value m _i . f _j and its composed flow vector F (S205).

Next, the file flow value f _{j is} assigned to each node in each file flow path P _j , and various changes in the file capacity value x _i that each node must handle at different points in the network model are derived. State, and define it as capacity vector X (S206). Since multiple file flow paths P _j may pass through the same node, the file capacity value x _i at the node should be at least all file flow paths P _j through the node. The sum of all file traffic values f _j .

Preferably, the operation relationship between the file capacity value and the file flow value is as follows: 1. The sum of all file flow values of all file flow paths through the same node is not greater than the maximum file capacity value; At the point, the sum of all file traffic values through all file flow paths of the same node is not greater than the current capacity of the node at that time.

Then, from the various change states of the file capacity value x _i that each employee must process, an optimal set (Ω min ) is found (S207), and each capacity change state represented by the optimal set (Ω min ) represents The minimum human resource required to complete the expected value d of the total flow of documents expected by the operator is referred to herein as the lower boundary point.

Figure 2B is a detailed flow chart for generating employee competency levels, as detailed below. First, an education training and test for all employees' work-related businesses in the internal network of the enterprise to detect the ability of each employee in the internal network of the enterprise to handle different file capacities in the unit time (S211). A first fuzzy set is defined in the software program, and includes a text description of a plurality of capability levels (S212), for example: "excellent", "good", "normal", "bad", "failed", etc. Can be used to describe the employee's test results. Providing a probability function table, comprising a membership function, selecting an appropriate member function μ _i corresponding to each node b _i by the probability function table, and performing operations according to the capability level and file capacity value of the node (S213) ), because the capability level and the file capacity value of the same node are a variable, the result of the operation can obtain the probability distribution range (S214) of each node b _i in the network model at different capability levels, which includes A node b _i with a certain capability level can successfully handle the probability of different file capacity values x _i . Incidentally, the probability distribution range can be used to determine the maximum file capacity value m _i that each node can complete in the unit time.

Figure 2C is a flow chart for calculating the linguistic performance indicators, as detailed below. A system reliability function (S221) is first provided in the software program, which includes a relationship between the bottom point and the probability distribution range as a system reliability value. According to the bottom limit point and the probability distribution range obtained by the flow of FIG. 2A and FIG. 2B, the system reliability value when the total flow of the file that can be completed in the network model is greater than the expected value d can be calculated ( S222), and using the inverse function of the system reliability function to convert the system reliability value into a linguistic performance indicator (S223), wherein the inverse function operation of the system reliability function includes a second fuzzy set, for example: The terms "Declined", "Recertified", "Acceptable", etc., which evaluate the internal network performance of the enterprise, that is, the above-mentioned "linguistic performance indicators" are presented on a display screen (S224).

Figure 3 shows a simple network model. The nodes b ₁ , b ₂ , b ₃ , and b ₄ of the network model represent employees in the internal network of the enterprise, s is the source node of the network model, t is the destination node, and the rest is Relay node (succeeding node). The arrows a ₁ , a ₂ , a ₃ , a ₄ , a ₅ , a ₆ represent the process precedence relationship between the two nodes b ₁ , b ₂ , b ₃ , b ₄ and the direction in which the file flows. Nodes b ₁ , b ₂ , b ₃ , b ₄ and their program sequence can combine multiple file flow paths. For example, all file flow paths in Figure 3 are P ₁ ={ b ₁ , b ₂ , b ₄ }, P ₂ ={ b ₁ , b ₂ , b ₃ , b ₄ }, P ₃ ={ b ₁ , b ₃ , b ₄ }, P ₄ ={ b ₁ , b ₃ , b ₂ , b ₄ } . In this embodiment, all file flow paths P _i start from the same starting node s, pass through the relay nodes, and end at the same terminal node t. The file flow path P _i can be defined in a manner of a minimal path technique.

Example path P defined in the network-based model of the present embodiment of the document flow ₁ is f _1, the file path P ₂ flow rate value f _2, the file path P _3, the flow rate is f _{3, 4} of the flow path of the file P The value is f ₄ and the network model is set to an expected value d = 5 (acceptable level) of the required document flow to be processed in a unit time, and the expected value d is stored in the recording medium.

In order to understand the maximum file capacity m _i that each node b ₁ , b ₂ , b ₃ , b ₄ in the network model may complete in a unit of time, it is necessary for each node b ₁ , b ₂ , b ₃ , b ₄ represent the education and training of the staff-related business and conduct a test procedure to detect the ability of each employee in the company's internal network to handle different amounts of documents in the unit time.

The steps of the test procedure are as follows: a plurality of different test capacities are provided to test each employee to know the ability of the employee to handle different file capacities; and then, according to the test result of each employee, a capability level is respectively given. To distinguish the file processing capabilities of each employee. In other words, the competency level can represent the employee's test results.

In this embodiment, a first fuzzy set represents different capability levels, and the first fuzzy set is described by capability levels such as "excellent", "good", "normal", "bad", and "failed". A group or collection of words that corresponds to a real interval, such as [0, 1], to facilitate computation by the computer system. Record the test result of each test capacity of the employee on the recording medium of the computer system; and then establish a probability function table (shown in Table 2) in the software program, the probability function table includes the member function μ _i (membership functions), File capacity value (or test capacity) x _i , the capability level contained in the first fuzzy set, and the correspondence of a probability distribution range; then the member function μ _i is executed to obtain a different capability level for each node. , can successfully handle a range of probability distribution of different file capacity (as shown in Table 2), and store the probability distribution range on the recording medium.

Table 1 lists the test results of nodes b ₁ , b ₂ , b ₃ , and b ₄ and their corresponding member functions μ ₁ , μ ₂ , μ ₃ , and μ ₄ . For example, the test result of node b ₁ is good, which corresponds to The member function is μ ₁ (good); the test result of node b ₂ is failed, its corresponding member function is μ ₂ (failed)... and so on.

Table 2 lists the probability distribution values of the employees of each node b ₁ , b ₂ , b ₃ , and b ₄ that deal with a file capacity value x _i at different capability levels (excellent, good..), and the corresponding file capacity value x _{When the} time is different, the probability values are also different. The numbers 0~6 in the first column above Table 2 represent different file capacity values x _i , and the left first behavior member function μ _i .

According to the test results of Table ₁ , the capability level of node b ₁ is "good", the capability level of b ₂ is "failed", the capability level of b ₃ is "good", and the capability level of b ₄ is "excellent". According to this, the probability values of four columns of μ ₁ (good), μ ₂ (failed), μ ₃ (good), and μ ₄ (excellent) in Table 2 are queried. For example, the node b ₁ obtains the capability level "good" in the test, so the member function is μ ₁ (good), and when the file capacity value x _l is 6, the table 2 is obtained by the node b ₁ in the unit time. The probability of being able to process 6 documents is 0.200. Since the test capacity is up to 6 files in this test, the probability value is not yet 0, which represents the maximum file capacity value m ₁ =6 that node b ₁ can handle in this unit time.

And so on, node b ₂ obtains the capability level "failed" in the test, so the member function is taken as μ ₁ (failed), and when the file capacity value x ₁ is 6, the check table 2 gets the node b ₂ successfully processed. The probability of 6 copies of the file is 0.000, which means that node b ₂ cannot process 6 copies of the file in the unit time; and when the file capacity value x ₂ is 5, the table b ₂ can successfully process 5 copies of the file. The probability of this is 0.004, which represents the maximum file capacity value m ₂ =5 that node b ₂ can handle in this unit of time.

From the results of the wheel surveys compiled in Tables 1 and 2, it can be known that the maximum file capacity value m _{i that} each employee of each node b ₁ , b ₂ , b ₃ , b ₄ can handle in this unit time is m ₁ =6 m ₂ =5, m ₃ =6, m ₄ =6. Thereby, the software program is executed to read the expected value d and the maximum file capacity value m _i in the recording medium, and according to the operation relationship between the file capacity value x _i and the file flow value f _j , the four file flow values f ₁ , f ₂ , f ₃ , f ₄ can be derived:

Where f _j : the file flow value on each file flow path P _j ; j : the number of the file flow path; F : the group formed by the file flow value f _i of all file flow paths P _j , expressed as a vector, Is a flow vector; F : a set of flow vectors F.

The file flow value f _j on each file flow path P _j is different at different time points, so the set F of the flow vector F has the following result: {(0,0,0,5),(0 ,0,1,4),(0,0,2,3),(0,0,3,2),(0,0,4,1),(0,0,5,0),(0 ,1,0,4),(0,1,1,3),(0,1,2,2),(0,1,3,1),(0,1.4,0),(0,2 ,0,3),(0,2,1,2),(0,2,2,1),(0,2,3,0),(0,3,0,2),(0,3 ,1,1),(0,3,2,0),(0,4,0,1),(0,4,1,0),(0,5,0,0),(1,0 ,0,4),(1,0,1,3),(1,0,2,2),(1,0,3,1),(1,0,4,0),(1,1 ,0.3),(1,1,1,2),(1.1,2,1),(1,1,3,0),(1,2,0,2),(1,2,1,1 ),(1,2,2,0).(1,3,0,1),(1,3,1,0),(1,4,0,0),(2,0,0,3 ),(2,0,1,2),(2,0,2,1),(2,0,3,0),(2,1,0,2),(2,1,1,1 ), (2,1,2,0), (2,2.0,1),(2,2,1,0), (2,3,0,0), (3,0,0,2), (3,0,1,1),(3,0,2,0).(3,1,0,1),(3,1,1,0),(3,2,0,0), (4,0,0,1), (4,0,1,0), (4,1,0,0), (5,0,0,0)}.

Since the employees of each node b ₁ , b ₂ , b ₃ , b ₄ may have to process files from multiple file flow paths P _j , the file capacity value x _i that each employee needs to process should be all the files that are handled. The flow value f _j sums up. The group formed by the file capacity values x _i of all the nodes b _{i is} represented by a capacity vector X = ( x ₁ , x ₂ , x ₃ , x ₄ ) (capacity vector). The capacity vector X = ( x ₁ , x ₂ , x ₃ , x ₄ ) can be known using all of the possible flow vectors F described above and the following operational relationships:

The change of the file capacity value x _i of the employees of each node b ₁ , b ₂ , b ₃ , b ₄ at different time points is different, so the various capacity vectors X underneath can be obtained: { X ₁ = (5,5,0,5), X ₂ =(5,5,1,5), X ₃ =(5,4,1,5), X ₄ =(5,5,2,5), X ₅ =(5,4,2,5), X ₆ =(5,3,2,5), X ₇ =(5,5,3,5), X ₈ =(5,4,3,5) , X ₉ =(5,3,3,5), X ₁₀ =(5,2,3,5), X ₁₁ =(5,5,4,5), X ₁₂ =(5,4,4, _{5), X 13 = (5,3,4,5} ), X 14 = (5,2,4,5), X 15 = (5,1,4,5), X 16 = (5,5, 5,5), X ₁₇ =(5,4,5,5), X ₁₈ =(5,3,5,5), X ₁₉ =(5,2,5,5), X ₂₀ =(5, 1,5,5), X ₂₁ =(5,0,5,5)).

Next, among the various capacity change states represented by the capacity vector X , a minimum capacity vector X is selected to form an optimal set (Ω _min ) as: { X ₁ =(5,5,0, _{5), X 3 = (5,4,1,5} ), X 0 = (5,3,2,5), X 10 = (5,2,3,5), X 15 = (5,1, 4,5), X ₂₁ =(5,0,5,5)}.

The above operation process is an operation research method, and each of the minimum capacity vectors X in the optimal set (Ω _min ) is used as one of the lower boundary points of the work research method. Each of the smallest capacity vectors X contained in the optimal set (Ω _min ) represents the minimum human resources required to complete the expected value d of the total file traffic expected by the operator from the internal network of the enterprise, with the least amount of manpower. Resources will be able to complete tasks with the highest performance.

Then, according to the optimal set (Ω _min ) composed of the probability distribution range of Table 2 and the minimum capacity vector X , the probability value when the total flow rate successfully processed by the network model is not less than the expected value is calculated, and the probability value is calculated. Presented on a display screen. That is, using the probability theory (discharge principle) to obtain the possible system reliability value when the enterprise internal network processes more than five files per unit time (the expected total flow rate is expected to be d = 5) R ₅ ) is 0.6693, as shown in the calculation below:

Where B _i represents the following lower boundary point:

Μ μ then the inverse function of the corresponding language available ^-1 effectiveness index (Linguistic Performance Index), = (0.6693)=Acceptable.

In summary, the corresponding relationship between system reliability value, real interval and linguistic performance index is shown in Figure 4. The horizontal axis represents the system reliability value, and the middle polyline divides the system reliability value into several regions, each The region is given different linguistic performance indicators; the vertical axis represents the real interval [0, 1] corresponding to the linguistic performance indicator.

The above is only the preferred embodiment of the present invention, and the scope of the invention is not limited thereto, that is, the simple equivalent changes and modifications made by the scope of the invention and the description of the invention are All remain within the scope of the invention patent. In addition, any embodiment or patent application of the present invention The scope of the invention is not intended to be exhaustive or to be construed. In addition, the abstract sections and headings are only used to assist in the search of patent documents and are not intended to limit the scope of the invention.

The starting node of the corporate intranet ‧‧‧s

The endpoint of the corporate intranet ‧‧‧t

Program relationship ‧‧‧a1, a2, a3, a4, a5, a6

Node ‧‧‧b1, b2, b3, b4

FIG. 1 is a schematic diagram of an operation flow of an internal network performance evaluation method of an enterprise according to an embodiment of the present invention.

FIG. 2A is a schematic diagram showing a detailed process of generating a bottom point according to an embodiment of the invention.

FIG. 2B is a schematic diagram showing a detailed process of generating an employee's ability level according to an embodiment of the present invention.

FIG. 2C is a schematic diagram of a calculation process of a language performance indicator according to an embodiment of the invention.

FIG. 3 is a schematic diagram of a network model according to an embodiment of the present invention.

FIG. 4 is a schematic diagram of language performance indicators based on fuzzy logic operation results according to an embodiment of the invention.

Claims (15)

An enterprise internal network performance evaluation method is applied to a software program to evaluate the performance of an enterprise internal network. The software program includes a network model, a probability function table, a system reliability function, and a first a fuzzy set and a second fuzzy set, the network model comprising a plurality of nodes to correspond to a plurality of employees of the internal network of the enterprise, the method comprising: performing a test procedure on the employee corresponding to each of the nodes, Obtaining a test result; describing the test result of the employee by the first fuzzy set to define a capability level of the node; converting the capability level to a probability distribution range by using the probability function table; and using the system reliability function Performing an operation on all the probability distribution ranges corresponding to the node to obtain a system reliability value; performing a fuzzy logic operation on the system reliability value according to the second fuzzy set to obtain a text description a linguistic performance indicator; and presenting the linguistic performance indicator on a display screen. The method for evaluating an internal network performance of an enterprise according to claim 1, wherein the plurality of nodes of the network model include a start node, a plurality of relay nodes, and a terminal node, and each of the two nodes There is a program sequence relationship, and a plurality of file flow paths are defined according to the program sequence relationship, and all of the file flow paths start from the start node, and end through the relay nodes and end at the terminal node. The method for evaluating an internal network performance of an enterprise according to claim 1, wherein the system reliability value is performed according to the second fuzzy set. The step of the fuzzy logic operation performs the fuzzy logic operation using the inverse function of the system reliability function. The method for evaluating an internal network performance of an enterprise according to claim 1, wherein the step of performing the test procedure for each of the employees corresponding to the node comprises: providing a plurality of test capacities for each The employee performs a test; records the test result of the employee for each test capacity on the record medium; provides a plural member function; and establishes the probability function table in the software program, the probability function table includes the member functions, Corresponding relationship between the test capacity, the capability levels, and the probability distribution range; and querying the probability function table according to the test capacity and the capability level to obtain each of the nodes having different capability levels, respectively corresponding to The probability distribution range. For example, the method for evaluating the internal network performance of the enterprise described in the first application of the patent scope, wherein the software program is included in an enterprise resource planning system. For example, the method for evaluating the internal network performance of the enterprise described in claim 1, wherein the software program is included in a customer relationship management system. An enterprise internal network performance evaluation method is applied to a computer system for evaluating the performance of an enterprise internal network, wherein the computer system includes a software program stored in a recording medium and processed by a micro processing To perform its operations, and the internal network of the enterprise includes a plurality of employees, including: Providing a network model in the software program, the network model corresponding to the enterprise internal network, comprising a plurality of nodes corresponding to the employees, and forming a plurality of file flow paths; Each of the nodes defined in the network model has a file capacity value, and each file flow path has a file flow value, wherein the file capacity value has a first operational relationship with the file flow value, and all files of the node are The group formed by the capacity value is defined as a capacity vector, and the group formed by the file flow value of all the file flow paths is defined as a flow vector; an expected value of the total flow of the network model is determined. And storing the expected value in the recording medium; performing a test procedure on the employee corresponding to each node to obtain a test result; describing the test result of the employee in a fuzzy set to define the node a capability level; converting the capability level into a probability distribution range by a probability function table, and storing in the recording medium; querying the recording medium The probability function table, and selecting a maximum file capacity value from the file capacity values corresponding to the probability distribution range of each node; executing the software program to read the expected value in the recording medium and the a maximum file capacity value, and according to the first operational relationship to obtain all possible traffic vectors; and according to all possible traffic vectors and the first operational relationship, the calculation It is possible that the capacity vector is selected from all possible capacity vectors; a system reliability function is provided, which includes the probability distribution range and the minimum capacity vector and a system reliability value. An operation relationship; according to the second operation relationship, calculating a system reliability value when the total flow rate successfully processed by the network model is not less than the expected value; performing a fuzzy logic operation on the system reliability value, and converting the system A linguistic performance indicator corresponding to the reliability value, and the linguistic performance indicator is presented on a display screen. The method for evaluating an internal network performance of an enterprise according to claim 7, wherein the step of performing a fuzzy logic operation on the reliability value of the system performs the fuzzy logic operation by using an inverse function of the system reliability function to obtain the Linguistic performance indicators. The method for evaluating an internal network performance of an enterprise according to claim 7, wherein the plurality of nodes includes a start node, a plurality of relay nodes, and a terminal node, and all the file flow paths are from the start node. Start, pass through the relay nodes, and end at the terminal node. The method for evaluating an internal network performance of an enterprise according to claim 7, wherein the step of forming the plurality of file flow paths includes: defining at least one program sequence relationship between each of the two nodes in the network model; According to the sequence relationship of the programs, a plurality of file flow paths of the network model are defined. The method for evaluating an internal network performance of an enterprise according to claim 7, wherein the first operational relationship comprises: a total sum of all the file traffic values of all the file flow paths through the same node is not greater than the maximum file. Capacity value. The method for evaluating an internal network performance of an enterprise according to claim 7, wherein the first operational relationship includes: at the same time point, all the file traffic values of all the file flow paths of the same node at the same time point The value is not greater than the file capacity value of the node at the time. The method for evaluating an internal network performance of an enterprise according to claim 7, wherein the step of performing the test procedure for each of the employees corresponding to the node comprises: providing a plurality of test capacities for each of the employees a test; recording, by the employee, a test result of each test capacity in the recording medium; providing a complex member function; establishing the probability function table in the software program, the probability function table including the member functions, and the quizzes Corresponding relationship between the capacity, the capability levels, and the probability distribution range; and querying the probability function table according to the test capacity and the capability level to obtain a probability corresponding to each of the nodes having different capability levels distribution range. For example, the method for evaluating the internal network effectiveness of the enterprise described in claim 7 is wherein the computer system is an enterprise resource planning system. Such as the internal network performance evaluation of the enterprise mentioned in the scope of patent application The estimation method, wherein the computer system is a customer relationship management system.