CN102622668A - Risk early warning method based on technological processes - Google Patents

Abstract

The invention discloses a risk early warning method based on technological processes, which belongs to the field of risk early warning. The risk early warning method includes: A, obtaining information of each business processing joint of a business processing technological process of an operation control system, finding out various business processing joints with safety risks according to the information, and correspondingly setting up various technological process risk points; B, finding out various relevant risk points capable of triggering monitored potential unsafe events from the technological process risk points, building a risk conduction path model according to the relevant risk points and the potential unsafe events, and obtaining severity level of the potential unsafe events and gathered data information of unsafe information of the relevant risk points; C, estimating the relevant risk points to determine severity level according to the risk conduction path model, the severity level of the potential unsafe events and the gathered data information; and D, early warning via the operation control system when the severity degree of the risk points reaches or exceeds preset severity level. The method can timely early warn according to the severity degree of the risk points in the determined technological processes.

Description

Method for prewarning risk based on flow process

Technical field

The present invention relates to the Risk-warning field, relate in particular to a kind of method for prewarning risk based on flow process.

Background technology

In the present air line, operation control center is that flight production ensures department, and groundwork is that the flight operation information is compiled, processes, analyzes, makes a strategic decision and issues.Operation control system by operation control center is handled the key business flow process of flight, like dispatch clearance flow process etc.But in the handled flow process of operation control system, have strong relevance between the business operation of each service processing node, the risk conducting problem is more outstanding, and each service processing node all can become a risk point, thus potential dangerous incident after touch.And present operation control system does not have and in the processing operation flow, has the some of risk or certain several service processing node to the risk of a certain potential dangerous incident when higher; When promptly be prone to triggering this potential dangerous incident; In time carry out early warning; Thereby avoid causing this dangerous incident, avoid accident to take place.

Summary of the invention

The purpose of embodiment of the present invention provides a kind of method for prewarning risk based on flow process; Solve the operation control system that uses in the present air line; Can't find out the risky service processing node that possibly cause follow-up dangerous incident in its business processing flow and in time carry out early warning, cause the problem of follow-up easy initiation serious accident.

The present invention realizes through following technical proposals:

Embodiment of the present invention provides a kind of method for prewarning risk based on flow process, may further comprise the steps:

Steps A; Obtain the information of each included service processing node of business processing flow that operation control system carries out; Find out each service processing node that has security risk in the said information, set up risk point as the flow process risk point corresponding to each service processing node of finding out;

Step B; From the corresponding potential dangerous incident of said business processing flow; Selected potential dangerous incident of being monitored; From the flow process risk point, find out the relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model;

Step C assesses each relevant risk point according to the serious grade and the said combined data information of said risk conducting path model of setting up and the said potential dangerous incident obtained, confirms the order of severity of each relevant risk point;

Step D if the order of severity of relevant risk point meets or exceeds preset serious grade, then to said operation control system output early warning information, carries out early warning by said operation control system according to said early warning information.

Embodiment of the present invention also provides a kind of method for prewarning risk based on flow process, may further comprise the steps:

Steps A; Obtain the information of each included service processing node of business processing flow that operation control system carries out; Find out each service processing node that has security risk in the said information, set up risk point as the flow process risk point corresponding to each service processing node of finding out;

Step B; From the corresponding potential dangerous incident of said business processing flow; Selected a kind of potential dangerous incident; From said flow process risk point, find out each relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model;

Step C assesses each relevant risk point according to the serious grade and the said combined data information of said risk conducting path model of setting up and the said potential dangerous incident obtained, confirms the comprehensive order of severity of each relevant risk point;

Step D; Repeat above-mentioned steps B, the comprehensive order of severity of C all risk points in drawing said flow process risk point under the different potential dangerous incident of same serious grade, calculate the comprehensive order of severity of the overall situation of each risk point based on the order of severity of each risk point;

Step e if the comprehensive order of severity of the overall situation of risk point meets or exceeds the preset order of severity, then to said operation control system output early warning information, is carried out early warning by said operation control system according to said early warning information.

Technical scheme by the invention described above provides can be found out; The method for prewarning risk that embodiment of the present invention provides; The business processing flow performed according to operation control system; Find out the flow process risk point in each service processing node in the business processing flow, according to the potential dangerous incident of monitoring, from the flow process risk point, obtain each relevant risk point relevant and set up risk conducting path model again with this potential dangerous incident; And obtain in the certain hour section with the combined data of the non-safety information of each relevant risk spot correlation and combine the serious grade of potential dangerous incident; Serious grade and combined data information through risk conducting path model of setting up and the potential dangerous incident of obtaining are assessed each relevant risk point, confirm the order of severity of each relevant risk point, thereby according to the order of severity of each relevant risk point; In time, carry out early warning according to said early warning information by operation control system to said operation control system output early warning information.The order of severity of each the relevant risk point under the serious grade of the potential dangerous incident that this method is monitored through being evaluated at; Which risk point of early warning accurately has the greater risk of this potential dangerous incident of triggering; Thereby utilize operation control system to carry out timely early warning; Effectively avoid the higher risk point of serious grade to cause dangerous incident, caused the problem of security incident.

Description of drawings

In order to be illustrated more clearly in the technical scheme of the embodiment of the invention; The accompanying drawing of required use is done to introduce simply in will describing embodiment below; Obviously, the accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skill in the art; Under the prerequisite of not paying creative work, can also obtain other accompanying drawings according to these accompanying drawings.

The method for prewarning risk process flow diagram that Fig. 1 provides for the embodiment of the invention;

Another process flow diagram of method for prewarning risk that Fig. 2 provides for the embodiment of the invention;

The overall flow figure of the operation control system that Fig. 3 provides for the embodiment of the invention;

The risk conduction network diagram that Fig. 4 provides for the embodiment of the invention to low on fuel;

The adjacency matrix synoptic diagram that Fig. 5 provides for the embodiment of the invention;

Fig. 6 analyzes synoptic diagram for the risk conduction of the risk point that the embodiment of the invention provides;

The conduction mode that Fig. 7 provides for the embodiment of the invention is an individual paths series model synoptic diagram;

The synoptic diagram of the risk conduction mode situation 1 that Fig. 8 provides for the embodiment of the invention;

The risk conduction mode that Fig. 9 provides for the embodiment of the invention is another synoptic diagram of individual paths series model;

The risk conduction mode that Figure 10 provides for the embodiment of the invention is a multipath conduction mode synoptic diagram;

The risk conduction mode that Figure 11 provides for the embodiment of the invention is the conduction mode synoptic diagram that contains common path;

The risk conduction mode that Figure 12 provides for the embodiment of the invention is another conduction mode synoptic diagram that contains common path;

The risk conduction mode that Figure 13 provides for the embodiment of the invention is the another conduction mode synoptic diagram that contains common path;

The non-safety information frequency statistical graph of low on fuel in the method for prewarning risk that Figure 14 provides for the embodiment of the invention;

Figure 15 is the synoptic diagram of individual paths series model for the method for prewarning risk risk conduction mode that the embodiment of the invention provides.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is carried out clear, intactly description, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on embodiments of the invention, those of ordinary skills belong to protection scope of the present invention not making the every other embodiment that is obtained under the creative work prerequisite.

To combine accompanying drawing that the embodiment of the invention is done to describe in detail further below.

First embodiment

The embodiment of the invention provides a kind of method for prewarning risk based on flow process; Can be used in the operation control system; Can carry out timely Risk-warning to the risk point of the corresponding easy initiation security incident of the business processing flow of operation control; Avoid security incident to take place, as shown in Figure 1, this method may further comprise the steps:

Steps A; Obtain the information of each included service processing node of business processing flow that operation control system carries out; Find out each service processing node that has security risk in the said information, set up risk point as the flow process risk point corresponding to each service processing node of finding out;

Above-mentioned steps A specifically may further comprise the steps:

Steps A 1 is obtained the information of each included service processing node of business processing flow that operation control system carries out;

Steps A 2 is found out each service processing node that comprises non-safety information in the said information, sets up risk point as the flow process risk point corresponding to each service processing node of finding out.

Step B; From the corresponding potential dangerous incident of said business processing flow; Selected potential dangerous incident of being monitored; From the flow process risk point, find out the relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model;

Among the above-mentioned steps B, specifically comprise according to each relevant risk point and said potential dangerous event establishment risk conducting path model:

Step B1 confirms pairing each service processing node of each relevant risk point;

Step B2 finds out each corresponding service processing node of each relevant risk point conducts to said potential dangerous incident in business processing flow incidence relation;

Step B3 is connected to each relevant risk point by said incidence relation and forms risk conducting path model after the said potential dangerous incident.

Among the above-mentioned steps B; The combined data information of obtaining the non-safety information of each relevant risk point of said risk conducting path model in the certain hour section specifically comprises: each relevant risk point in the certain hour section during as the risk source of triggering said potential dangerous incident the frequency of its non-safety information and each relevant risk point it conducts discovery number of times of next non-safety information to other risk point during as middle the risk point that triggers said potential dangerous incident in the certain hour section.This combined data information also comprises:

The frequency in the frequency of each relevant risk point non-safety information relevant with said potential dangerous incident in the certain hour section, each relevant risk point risk source of the said potential dangerous incident of conduct triggering in the certain hour section and each relevant risk point are as the frequency of the middle risk point in each risk source of triggering said potential dangerous incident.

Step C assesses each relevant risk point according to the serious grade and the said combined data information of said risk conducting path model of setting up and the said potential dangerous incident obtained, confirms the order of severity of each relevant risk point;

Above-mentioned steps C specifically comprises:

Step C1 from each relevant risk point of the said risk conducting path model set up, finds out the conducting path that conducts to said potential dangerous incident as the risk point in risk source through middle risk point; And the frequency that obtains the non-safety information in risk source in the said combined data information with each in the middle of the discovery number of times of risk point non-safety information that other risk point conduction is come;

Step C2; The frequency of the conducting path that finds according to step C1, the non-safety information in risk source and in the middle of each risk point to the discovery number of times of the non-safety information of other risk point conduction; And the serious grade of said potential dangerous incident; The independent order of severity of risk point in the middle of calculating through following assessment models one, assessment models one is:

$\underset{A_i^k}{\min }\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}I(i,j){(A_i^k-A_j^k)}^2$

In the above-mentioned model; δ is the adjustment parameter; The minimal difference of representing two adjacent middle risk point degrees of risk; The partial ordering relation that risk point satisfies in the middle of formula (1) expression, all middle risk point contribution degree sums equal the constraint condition of the serious grade of potential dangerous incident on conducting path of formula (2) expression, and A is the serious grade of said potential dangerous incident;

Step C3, the independent order of severity of the middle risk point that obtains according to step C2 calculates the independent order of severity B in risk source through following assessment models two _k, assessment models two is:

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\·\·\·+\frac{h_i^k}{n_k}{H}_i^k+\·\·\·+\frac{h_r}{n_k}{H}_r^k$

$=\frac{h_1^k}{n_k}*0+\·\·\·+\frac{h_i^k}{n_k}*(A_1^k+\·\·\·+A_{i-1}^k)+\·\·\·+\frac{h_r^k}{n_k}A$

In the above-mentioned model two, n _kFor the frequency of the non-safety information of risk source k, promptly with the non-safety information frequency of risk point k as the risk source;

Be the number of times that non-safety information is found by risk point 1 place operation, the actual generation of risk source k this moment, the degree of risk of risk source k does $H_1^k=0;$

Be the number of times that non-safety information is found by the operation of risk point i place, this moment, the degree of risk of risk source k did $H_i^k=A_1^k+\·\·\·+A_{i-1}^k;$

spread out of operation control system for non-safety information; Found beyond the operation control system or the actual generation of final potential dangerous incident; This moment risk source k degree of risk is

in this case; This moment, the order of severity of risk source k equaled A, and A is the serious grade of said potential dangerous incident;

n _kWith

Between satisfy relation:

Step C4, the independent order of severity that above-mentioned assessment is obtained is as the order of severity of each self-corresponding each relevant risk point.

Among the above-mentioned steps C, after step C2, also comprise:

Step C20 based on the independent order of severity of each relevant risk point being assessed each the relevant risk point that obtains, carries out whole order of severity assessment to each relevant risk point, obtains the whole order of severity of each relevant risk point.Wherein, each relevant risk point being carried out whole order of severity assessment specifically comprises:

Step C201; Each relevant risk point is carried out whole order of severity assessment one by one; At first select a certain relevant risk point, from said combined data information, obtain the frequency of the middle risk point in the frequency of this relevant risk point non-safety information relevant in the certain hour section, the frequency in this relevant risk point risk source of the said potential dangerous incident of conduct triggering in the certain hour section and each risk source that the conduct of this relevant risk point triggers said potential dangerous incident with said potential dangerous incident;

Step C202; The frequency of this relevant risk point non-safety information relevant with said potential dangerous incident in the certain hour section that obtains according to step C201, this relevant risk point as the frequency in the risk source of triggering said potential dangerous incident and this relevant risk point frequency as the middle risk point in each risk source of triggering said potential dangerous incident, draw the whole order of severity r of this relevant risk point through following assessment models three calculating in the certain hour section _i, assessment models three is:

$r_i={\mathrm{\β}}_i{B}_i+\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}_k{A}_i^k$

β _i＝n _i/N _i

${\mathrm{\α}}_k=n_i^k/N_i$

$N_i=n_i+\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{n}_i^k$

In the above-mentioned assessment models three, N _iFrequency for risk point i non-safety information relevant in the certain hour section with said potential dangerous incident;

n _iBe risk point i number of times as the generation of risk source in the certain hour section;

is the frequency of risk point i as the middle risk point of risk source k;

M is illustrated in has m risk source to take place to comprise risk point i in its conducting path of back in the certain hour section;

Step 203 repeats above-mentioned steps 201,202 until the whole order of severity of obtaining each relevant risk point.

Among the above-mentioned steps C, further comprising the steps of after step C20:

Step C21 according to the independent order of severity and the whole order of severity of each relevant risk point being assessed each the relevant risk point that obtains, carries out comprehensive order of severity assessment to each relevant risk point, obtains the comprehensive order of severity of each relevant risk point.

Saidly each relevant risk point carried out the assessment of the comprehensive order of severity specifically comprise:

Step C211; Each relevant risk point is carried out comprehensive order of severity assessment one by one; A certain relevant risk point is selected in the road, from said combined data information, obtain this relevant risk point in the certain hour section in the frequency relevant, the certain hour section with said potential dangerous incident with said potential dangerous incident the frequency of all relevant risk points;

Step C212; This relevant risk point that obtains according to step C211 in the certain hour section in the number of times relevant, the certain hour section with said potential dangerous incident with said potential dangerous incident the frequency of all relevant risk points and the whole order of severity of this risk point, calculate the comprehensive order of severity R of this relevant risk point through following assessment models four _i, assessment models four is:

$R_i={\mathrm{\η}}_i{r}_i=(N_i/\underset{i}{\mathrm{\Σ}}{N}_i)r_i$

In the above-mentioned assessment models four, N _iBe the number of times relevant of risk point i generation in the certain hour section with said latent consequences, Frequency for all relevant in certain hour section risk points with said latent consequences;

Step C213 repeats above-mentioned steps C211, C212 until the comprehensive order of severity of obtaining each relevant risk point.

Step D if the order of severity of relevant risk point meets or exceeds preset serious grade, then to said operation control system output early warning information, carries out early warning by said operation control system according to said early warning information.

Second embodiment

The embodiment of the invention also provides a kind of method for prewarning risk based on flow process; Different with said method is in this method after obtaining the comprehensive order of severity of each risk point; Confirm the comprehensive order of severity of the overall situation of each risk point again; Carry out Risk-warning according to the comprehensive order of severity of the risk point overall situation, as shown in Figure 2, this method may further comprise the steps:

Steps A is obtained the information of each included service processing node of business processing flow that operation control system carries out, finds out each service processing node that has security risk in the said information, sets up each flow process risk point corresponding to each service processing node of finding out;

Step B; From the corresponding potential dangerous incident of said business processing flow; Selected a kind of potential dangerous incident; From said flow process risk point, find out each relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model;

Step C assesses each relevant risk point according to the serious grade and the said combined data information of said risk conducting path model of setting up and the said potential dangerous incident obtained, confirms the comprehensive order of severity of each relevant risk point;

Step D; Repeat above-mentioned steps B, the comprehensive order of severity of C all risk points in drawing said flow process risk point under the different potential dangerous incident of same serious grade, calculate the comprehensive order of severity of the overall situation of each risk point based on the comprehensive order of severity of each risk point;

Step e if the comprehensive order of severity of the overall situation of risk point meets or exceeds the preset order of severity, then to said operation control system output early warning information, is carried out early warning by said operation control system according to said early warning information.

In the said method, step C specifically may further comprise the steps:

Step C1 from each relevant risk point of the said risk conducting path model set up, finds out the conducting path that conducts to said potential dangerous incident as the risk point in risk source through middle risk point; And the frequency that obtains the non-safety information in risk source in the said combined data information with each in the middle of the discovery number of times of risk point non-safety information that other risk point conduction is come;

Step C2; The frequency of the conducting path that finds according to step C1, the non-safety information in risk source and in the middle of each risk point to the discovery number of times of the non-safety information of other risk point conduction; And the serious grade of said potential dangerous incident; The independent order of severity of risk point in the middle of calculating through following first assessment models, first assessment models is:

$\underset{A_i^k}{\min }\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}I(i,j){(A_i^k-A_j^k)}^2$

In above-mentioned first assessment models; δ is the adjustment parameter; Be the inverse of all risk point sums, represent the minimal difference of two adjacent middle risk point degrees of risk, the partial ordering relation that risk point satisfies in the middle of formula (1) expression; All middle risk point contribution degree sums equal the constraint condition of the serious grade of potential dangerous incident on conducting path of formula (2) expression, and A is the serious grade of said potential dangerous incident;

Step C3, the independent order of severity of the middle risk point that obtains according to step C2 calculates the independent order of severity B in risk source through following second assessment models _k, second assessment models is:

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\·\·\·+\frac{h_i^k}{n_k}{H}_i^k+\·\·\·+\frac{h_r}{n_k}{H}_r^k$

$=\frac{h_1^k}{n_k}*0+\·\·\·+\frac{h_i^k}{n_k}*(A_1^k+\·\·\·+A_{i-1}^k)+\·\·\·+\frac{h_r^k}{n_k}A$

In above-mentioned second assessment models, n _kFor the frequency of the non-safety information of risk source k, promptly with the non-safety information frequency of risk point k as the risk source;

Be the number of times that non-safety information is found by risk point 1 place operation, the actual generation of risk source k this moment, the degree of risk of risk source k does $H_1^k=0;$

Be the number of times that non-safety information is found by the operation of risk point i place, this moment, the degree of risk of risk source k did $H_i^k=A_1^k+\·\·\·+A_{i-1}^k;$

spread out of operation control system for non-safety information; Found beyond the operation control system or the actual generation of final potential dangerous incident; This moment risk source k degree of risk is

in this case; This moment, the order of severity of risk source k equaled A, and A is the serious grade of said potential dangerous incident;

n _kWith

Between satisfy relation:

Step C4; Each relevant risk point is carried out whole order of severity assessment one by one; At first select a certain relevant risk point, from said combined data information, obtain the frequency of the middle risk point in the frequency of this relevant risk point non-safety information relevant in the certain hour section, the frequency in this relevant risk point risk source of the said potential dangerous incident of conduct triggering in the certain hour section and each risk source that the conduct of this relevant risk point triggers said potential dangerous incident with said potential dangerous incident;

Step C5; The frequency of this relevant risk point non-safety information relevant with said potential dangerous incident in the certain hour section that obtains according to step C5, this relevant risk point as the frequency in the risk source of triggering said potential dangerous incident and this relevant risk point frequency as the middle risk point in each risk source of the said potential dangerous incident of triggering, calculate the whole order of severity r of this relevant risk point through following the 3rd assessment models in the certain hour section _i, the 3rd assessment models is:

$r_i={\mathrm{\β}}_i{B}_i+\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}_k{A}_i^k$

β _i＝n _i/N _i

${\mathrm{\α}}_k=n_i^k/N_i$

$N_i=n_i+\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{n}_i^k$

In above-mentioned the 3rd assessment models, N _iFrequency for risk point i non-safety information relevant in the certain hour section with said potential dangerous incident;

n _iBe risk point i number of times as the generation of risk source in the certain hour section;

is the frequency of risk point i as the middle risk point of risk source k;

Its conducting path comprised risk point i after m was illustrated in and has m risk source to take place in the certain hour section;

Step C6 repeats above-mentioned steps C4, C5 until the whole order of severity of obtaining each relevant risk point;

Step C7; Each relevant risk point is carried out comprehensive order of severity assessment one by one; Selected a certain relevant risk point obtains the frequency of all risk points that this relevant risk point is correlated with said potential dangerous incident in the number of times relevant with said potential dangerous incident, the certain hour section in the certain hour section from said combined data information;

Step C8; This relevant risk point that obtains according to step C7 in the certain hour section in the number of times relevant, the certain hour section with said potential dangerous incident with said potential dangerous incident the frequency of all relevant risk points and the whole order of severity of this risk point, calculate the comprehensive order of severity R of this relevant risk point through following the 4th assessment models _i, the 4th assessment models is:

$R_i={\mathrm{\η}}_i{r}_i=(N_i/\underset{i}{\mathrm{\Σ}}{N}_i)r_i$

In above-mentioned the 4th assessment models, N _iBe the number of times relevant of risk point i generation in the certain hour section with said latent consequences,

is the frequency of all risk points relevant with said latent consequences in the certain hour section;

Step C9 repeats above-mentioned steps C7, C8 until the comprehensive order of severity of obtaining each relevant risk point.

Among the step D of said method, the comprehensive order of severity of the overall situation that calculates each risk point according to the comprehensive order of severity of each risk point specifically may further comprise the steps:

Step D1; Each relevant risk point is carried out comprehensive order of severity assessment one by one; Selected a certain risk point; According to the frequency of the relevant non-safety information of the different potential dangerous incident of this risk point that obtains under the comprehensive order of severity under the different potential dangerous incident of same serious grade and this same serious grade of obtaining, calculate the comprehensive order of severity E of the overall situation of this risk point through following the 5th assessment models _{I, x}, the 5th assessment models is:

$E_{i,x}=\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\π}}_j{R}_{i,j},$ Wherein ${\mathrm{\π}}_j={\mathrm{Num}}_j/\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{Num}}_k$

In above-mentioned the 5th assessment models, R _{I, j}Be the potential dangerous incident of the j class under this severity level, the comprehensive order of severity of risk point i; X is the grade of said dangerous incident; Num _jBe the frequency of non-safety information relevant with the potential dangerous incident of j class under this serious grade, m is the quantity of the potential dangerous incident that comprises under this serious grade.

Further specify below in conjunction with the method for prewarning risk of the specific embodiment of in the operation control system of aviation, using above-mentioned method for prewarning risk the embodiment of the invention.

Operation control system is that flight production ensures department, is the system that the flight operation information is compiled, processes, analyzes, makes a strategic decision and issues.In the operation control system of the airline that does not have the Risk-warning function, realize Risk-warning; Can be through existing all kinds of non-safety informations, incident, abnormal conditions and system defect in the business processing flow of operation control system; Find potential safety hazard that exists in the flow process and the risk point of being correlated with and paying close attention to; The risk point that the order of severity is reached respective degrees (can be preset with the order of severity contrast) carries out Risk-warning, avoids causing potential dangerous incident.

As being example as potential dangerous incident with " with low on fuel " in the dispatch clearance flow process of operation control system core business treatment scheme, the method for prewarning risk of the embodiment of the invention is described, specifically undertaken by following step:

Step 1 is confirmed the risk point that exists in dispatch clearance workflow and the flow process;

Step 2; According to from possible potential dangerous incident; The potential dangerous incident (low on fuel) of selected preparation monitoring; From risk point, find out the relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model; As collect the corresponding non-safety information of each risk point, according to information analysis non-safety information of collecting and risk point frequency etc.;

Step 3 to the relative order of severity assessment of each risk point, is confirmed the order of severity of each risk point;

Step 4 is carried out Risk-warning according to the order of severity of risk point.

In the above-mentioned steps 1; Can go out each business processing flow according to the dispatch flow process combing of letting pass; Combing goes out detailed dispatch clearance flow process the rules and regulations that mainly to be the routine work flow process of letting pass from dispatch and CAAC, CA let pass to dispatch, and concrete flow process is as shown in Figure 3.

From the dispatch clearance flow process that obtains, find out each risk point; Confirming of risk point can be to combine dispatch clearance flow process; Confirm which operation link is a weak link in the dispatch clearance flow process; Which kind of unsafe condition (also can with reference to dispatch staff related work experience in the past) can take place, and risk point is to the mistake that possibly the occur description of (comprising that human error and other external factor cause mistake).

From dispatch clearance flow process shown in Figure 3, confirmed 42 the flow process risk points (seeing table 1) in the dispatch clearance process, wherein, each flow process risk point has following attribute:

1. the risk point numbering is given unique numbering to each risk point, to distinguish other risk points.

2. risk point type: the dispatch flow process of letting pass is divided into nine aspects, is respectively the unit information processing, the seaworthiness information processing; Weather information is handled, and navigational intelligence is handled, and Computer flight plan is made; Dispatch is let pass, the explanation of letting pass, the monitoring of letting pass; Installations and facilities and working environment are to confirm according to the phase process of dispatch clearance flow process, thus these risk points corresponding corresponding dispatch clearance flow process; Risk point shows in system just can have two kinds of forms: first kind of form is with tabular form, and with all risk point centralized displaying, second kind of form is that risk point is presented at the operational phase corresponding on the flow process.

3. risk point content: specifically describe the contingent error message of this risk point.

4. possibly trigger main body: promptly which department or post can cause this risk point to take place.

5. risk point is explained: risk point is described content and triggered the explanation that main body triggers risk point mechanism, and if found the consequence that possibly cause by later operation link after the risk point generation.

Each flow process risk point to confirming can be safeguarded according to different time sections; The content that is risk point can change because of the change of the adjustment of flow process or working environment to some extent; Therefore; Maintenance to risk point mainly is particular content and an association attributes of revising risk point, adds new flow process risk point.

Each flow process risk point of table 1 dispatch clearance flow process

Confirm each flow process risk point of dispatch clearance flow process in step 1 after; Carry out step 2; Confirm that at first the potential dangerous incident of preparing to monitor (is a latent consequences; For describing conveniently, the back all is called latent consequences with potential dangerous incident), every kind of latent consequences all has corresponding serious grade (seeing Table 2);

Table 2 CA aviation safety management system handbook risk causes consequence seriousness evaluation criterion

Specifically, the corresponding serious grade of latent consequences mainly shows following four aspects:

(1) the general safety level of operation control system can be described with possibility and severity of consequence that limited type of dangerous incident takes place.

(2) for the dangerous incident that consequence takes place; Has definite consequence; Though playing concrete non-safety information consequence for each can not be identical, the actual order of severity or the loss that brings are also variant, and the non-safety information that can have similar consequence is classified as one type; Like " low on fuel ", and confirm its unique seriousness grade.This seriousness grade that just requires to confirm can contain this type of consequence takes place.

(3) in time found by subsequent operation for risk; Do not cause the non-safety information of serious consequence; Can't directly go to estimate this incident and can cause much influences, if therefore need do not found the possible consequence that to bring from such risk by subsequent operation to the normal operation of flight from incident itself; And these type of wrong possibility two aspects of subsequent operation discovery are estimated; Just the seriousness with latent consequences is standard, weighs " distance " of this non-safety information to the generation of latent consequences, confirms its real order of severity.In the actual motion of flight, really take place and all kinds of dangerous incident that the has serious consequences part that only occupies the minority, it is a large amount of that what exist is to produce risk; But by the timely non-safety information of finding of subsequent operation; Because this type of information does not produce actual loss, thereby does not often cause enough attention, yet; According to " iceberg is theoretical " in the safety engineering, the dangerous incident that floats on the iceberg is the external manifestation of a large amount of potential safety hazards (not producing the non-safety information of consequence).Because this type of non-safety information produces on dispatch clearance flow process; Therefore can judge under certain precondition according to the internal logic of flow process; This non-safety information is not if can be caused the incident generation of which kind of type and the serious grade of latent consequences by timely discovery.

Above-mentioned explanation on the base confirms that the corresponding serious grade of preparing to monitor of latent consequences " low on fuel " is 5 grades (specifically seeing Table 3), and other latent consequences can be with reference to above-mentioned explanation corresponding to different serious grades.

The serious grade that table 3 latent consequences and latent consequences are corresponding

Sequence number	Title	Latent consequences is described	Serious grade
				Consequence
1	Low on fuel	Aircraft is less than in oil mass under the situation of regulation and takes off	5 grades
				Consequence
2	......	......	......
					......	......	......
Consequence n	......	......	......

From the flow process risk point that above-mentioned steps 1 obtains; Definite each risk point that can cause latent consequences for " low on fuel " forms the risk conducting path network corresponding to the latent consequences of " low on fuel " according to each the flow process risk point that obtains, and its risk conducting path is as shown in Figure 4; Elliptical section is divided the expression risk point among Fig. 4; Numerical portion is represented corresponding risk point numbering, the rectangle part consequence that expresses possibility, and each single sub path among Fig. 4 is all represented one by risk source (risk point that promptly triggers first) the risk conducting path to possible consequence; Square frame partly is the explanation to the path among Fig. 4, is the note to the path mechanism.Because each risk point all has and a plurality ofly possibly trigger main body; But can find out as long as after confirming the risk source and finding risk point from Fig. 4; Just can confirm from this risk source the risk conducting path of risk point to the end, for example the risk source is that risk point 8 (meteorological data is inaccurate, imperfect or fail in time to transmit) discovery risk point is a risk point 14, and so final intercurrent risk point must be risk point 9 (analysis to meteorological data is inaccurate) and risk point 10 (failing to make correct clearance makes a strategic decision); Risk point in the middle of just can confirming according to risk source and discovery risk point like this; Thereby when information gathering, can reduce input information as far as possible, thereby raise the efficiency.

Can confirm that from the risk conducting path that Fig. 4 provides this path network is actually a directed acyclic graph, therefore can utilize adjacency matrix to represent corresponding to " low on fuel " latent consequences.

Adjacency matrix is a kind of method that in computing machine, concerns between the statement figure summit, and the essence of adjacency matrix is to utilize bi-values 0 or 1 to represent whether to exist between two nodes the forward-backward correlation relation.As establish adjacency matrix G=<v, E>Be digraph, wherein a V=<v ₁, V ₂..., V _n>, being the digraph vertex set, the adjacency matrix of G is a n * n matrix with following relation:

$a_{\mathrm{Ij}}=\left\{\begin{array}{cc}1& <v_i,v_j>\&Element;E\left(G\right)\\ 0& \mathrm{Else}\end{array}\right.,$ Wherein<v _i, v _j>Expression is the i directed edge of j to the limit from the summit.A digraph with shown in Figure 5 is an example, and its corresponding adjacency matrix is shown in the following formula (3.1).As the adjacency matrix that can the digraph that the risk conducting path of " low on fuel " latent consequences is corresponding converts to is expressed as being similar to the sample table of expression.

$\left[\begin{array}{cccc}0& 1& 1& 0\\ 0& 0& 0& 1\\ 0& 0& 0& 1\\ 0& 0& 0& 0\end{array}\right]---\left(3.1\right)$

Table 4 adjacency matrix is corresponding to each risk point sample table

As seen from the above description, the risk conducting path network of every kind of latent consequences all corresponding corresponding adjacency matrix.

Confirmed to be each risk point of " low on fuel " through above-mentioned steps 2, and the risk conduction network that forms of each risk point, and can represent the risk conducting path confirmed through the form of adjacency matrix corresponding to latent consequences.

After above-mentioned steps 2 is confirmed the risk conducting path to the latent consequences of " low on fuel "; Carry out the processing of step 3; Each risk point order of severity is assessed; Can be according to investigating the non-safety information of collecting in the time period, to the latent consequences of being monitored, draw the relative order of severity (order of severity of risk point is meant the mistake that produces on certain operating process influence degree to dispatch clearance flow process) of each risk point through corresponding calculated with mathematical model.

Concrete assessment can be undertaken by following step:

(1) with the risk conducting path network of the latent consequences of " low on fuel ", representes with adjacency matrix;

(2) will investigate in the time period non-safety information and gather and obtain combined data, wherein every non-safety information comprises following essential information: 1. risk source, trigger the risk point that this non-safety information takes place; 2. find risk point, because of in time finding risk, and avoid the risk point that takes place; 3. latent consequences and seriousness grade thereof;

(3) be directed against the latent consequences of being monitored, the comprehensive order of severity of each risk point in the time period is investigated in assessment, and the contrast through to the comprehensive order of severity of each risk point that obtains can realize the high risk point of the order of severity is carried out early warning.

When being assessed, the order of severity of risk point need consider the problem of risk conduction; Owing to can influence subsequent operation through flow process after a certain risk point of dispatch clearance flow process takes place; If not possessing the debugging ability to this mistake or possess the debugging ability but fail, the subsequent operation flow process do not find; The capital has influence on the correctness of operation, and then influences the normal operation of flight.Therefore assess the order of severity of a risk point, consider the influence of dispatch clearance flow process, promptly will assess based on the conduction of risk from this risk point.

For the arbitrary risk point in the dispatch clearance flow process, it has two kinds of situations, and the one, because the external factor beyond this risk point oneself factor or the operation control system causes this risk point to take place, on the risk conducting path, show as the risk source; The 2nd, risk point in the middle of at this moment this risk point shows as on the risk conducting path takes place owing to this risk point takes place to cause other risk points (risk points of other operations relevant with this operation).These two kinds of situations; To be two kinds of different risk types; First kind of situation is meant because operating personnel's the human factor or the generation of the dangerous incident that external factor causes are a kind of in addition because the order of severity that associative operation goes wrong and causes this risk point to take place on the flow process is as shown in Figure 6.Therefore need assess from this two aspect when assessing the order of severity of risk point.As can beappreciated from fig. 6, risk conduction network is actually the abstract representation to operation flow, therefore in fact just is based on the risk point seriousness assessment of flow process based on the assessment of the risk point order of severity of risk conduction network.

The risk conductive process has following four characteristics:

(1) on a risk conducting path, the risk conduction is " risk amplification " process: the risk of a last node impels the risk point of next node to take place through influencing next node, and the possibility that makes the net result incident take place increases.Therefore between the risk point that points relationship is arranged on the conducting path, the order of severity of follow-up risk point is greater than the order of severity of preorder risk point.

(2) angle that whether takes place from latent consequences; The risk conduction is " risk decay " process: operation has certain debugging ability to preorder in operation follow-up on the flow process; This risk of performance is tackled when conducting to a certain operating process on risk conduction aspect, therefore when the order of severity of assessment risk point, not only need consider the influence to other risk points; Also need consider the interdiction capability of flow to this type of mistake; After a certain risk point took place, the interdiction capability of subsequent operation flow process was high more, and then this risk point is seriously just relatively low.

(3) influence of conduction range: after some risk point takes place; Mistake takes place if for example basic information is obtained with processing links; Can impact a plurality of running node in the dispatch clearance flow process; On range, it is wider to show as the conduction range, sees that from structure more path causes the result who influences the flight operation to take place

(4) influence of multiplicity: if the number of times of a risk point generation is many more, then the order of severity of this risk point is high more.

The assessment of the risk point order of severity comprises middle risk point order of severity assessment, and the risk source order of severity is assessed.

Wherein, the middle risk point order of severity can draw through following Model Calculation:

When the assessment risk point order of severity, confirm that earlier the serious grade of latent consequences " low on fuel " is 5 grades, here with C represent this latent consequences the order of severity (similar, the serious grade of other latent consequences can be used A, B ... Expression successively).

(the risk source must be realized through making at least one risk conducting path form path the influence of final consequence according to the definition of the middle risk point order of severity; Suppose that risk point i is the node (risk point on the risk conducting path) on this paths; The risk that the middle risk point order of severity is defined as risk source k generation is transmitted to risk point i; Percentage contribution to final consequence generation takes place in risk point i; Be designated as

), all risk point order of severity sums are the order of severity of latent consequences on the risk conducting path.Middle risk point will satisfy risk conductive process " risk amplification " character, and therefore on this paths, the order of severity of follow-up middle risk point promptly satisfies partial ordering relation by path greater than the order of severity of risk point before this risk point.Though risk point all is because same mistake causes in fact; Under a lot of situations; Subsequent operation does not have complete debugging ability or does not have the responsibility of debugging; Therefore consider to utilize risk point order of severity difference on the same conducting path is minimised as objective function the order of severity of risk point in the middle of finding the solution.In sum, the order of severity of risk point in the middle of the following model solution capable of using:

$\underset{A_i^k}{\min }\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}I(i,j){(A_i^k-A_j^k)}^2$

Objective function in the above-mentioned model is that the order of severity difference quadratic sum of adjacent middle risk point is minimum; Wherein δ is the adjustment parameter; The minimal difference of representing the two adjacent middle risk point orders of severity; The partial ordering relation that risk point satisfies in the middle of formula (1) expression, all middle risk point contribution degree sums equal the constraint condition of the incident net result order of severity on formula (2) expression one paths.

In the reality, because the risk conduction mode is divided into multiple situation, the order of severity that can be directed against the risk point of different risk conduction modes is carried out evaluates calculation:

(1) the risk conduction mode with individual paths is example, and is as shown in Figure 7, and the risk source is risk point k, and the Optimization Model of the calculating of risk point was suc as formula shown in 4.3 in the middle of the order of severity of latent consequences was represented with A:

$\underset{A_i}{\min }[{(A_2^k-A_1^k)}^2+{(A_3^k-A_2^k)}^2]$

$s.t\left\{\begin{array}{cc}{A}_3^k-A_2^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_2^k-A_1^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ \underset{i=1}{\overset{3}{\mathrm{\&Sigma;}}}{A}_i^k=A& \\ A_i^k\&GreaterEqual;0& \&ForAll;i\end{array}\right.$ (formula 4.3)

(1.1) order of severity of individual paths series model risk conduction mode risk point is calculated, and according to whether middle risk point being arranged, is divided into two kinds of situations (situation 1 and situation 2), and is specific as follows:

(1.1.1) calculating (situation 1) of the risk source order of severity during risk point in the middle of the nothing shown in Figure 8:

On the risk conducting path shown in Figure 8 except that the risk source takes place, do not have other risk points and be triggered, under this conducting path; There is not middle risk point, the order of severity of risk point in the middle of need not calculating, the order of severity in a calculation risk source gets final product; Be after risk point k is triggered, with the operation that directly possibly influence flight, or risk has spread out of operation control system; For operation control system, this risk has exceeded the range of control of operation control system, and the order of severity in risk source equals the order of severity of latent consequences at this moment; The order of severity of supposing latent consequences is A, then the order of severity B in risk source _k=A.

(1.1.2) another kind is the individual paths series model (situation 2) of risk point in the middle of the existence shown in Figure 9:

On conducting path shown in Figure 9, m centre risk point again after the k of risk source, m >=1 wherein,

The middle risk point order of severity in this conducting path can adopt following Optimization Model formula 4.4 to calculate, and the order of severity of middle risk point is the optimum solution of formula 4.4:

$\underset{A_i}{\min }[{(A_2^k-A_1^k)}^2+{(A_3^k-A_2^k)}^2+\&CenterDot;\&CenterDot;\&CenterDot;+{(A_m^k-A_{m-1}^k)}^2]$

$s.t\left\{\begin{array}{cc}{A}_{i+1}^k-A_i^k=\mathrm{\&delta;}\&GreaterEqual;0& i=1,\&CenterDot;\&CenterDot;\&CenterDot;,m\\ \underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{A}_i^k=A& \\ A_i^k\&GreaterEqual;0& \&ForAll;i\end{array}\right.$ (formula 4.4)

In above-mentioned formula 4.4, n _kTotal degree (promptly being the non-safety information frequency in risk source) for risk source k generation with risk point k;

The number of times that

found by risk point 1 place operation; The actual generation of only risky source k this moment, the risk source order of severity is

The number of times that found by risk point 2 place operations, the risk source order of severity

at this moment

The number of times that found by risk point i place behaviour, the risk source order of severity

at this moment

risk has spread out of the operation control system system; By the post beyond the operation control system or department's discovery or the actual generation of final consequence; This moment, the order of severity be

the risk source in this case, and we just think that the order of severity of risk source k this moment has just equaled A.

n _kWith

Satisfy following relation.

$n_k=\underset{i=1}{\overset{m,r}{\mathrm{\&Sigma;}}}{h}_i^k$ (formula 4.5)

In data gathering system, n _kNeed to analyze final consequence for being directed against, risk point k can directly add up as the number of times that the risk source takes place.Interception number of times

can be added up according to concrete non-safety information data.

The risk source order of severity by in 4.6 pairs of risk conducting paths shown in Figure 9 of formula is calculated, and the order of severity of risk source k is formula 4.6 result calculated.

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_i^k}{n_k}{H}_i^k+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_r}{n_k}{H}_r^k$

(formula 4.6)

$=\frac{h_1^k}{n_k}*0+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_i^k}{n_k}*(A_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+A_{i-1}^k)+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_r^k}{n_k}A$

(2) the risk conducting path signal (situation 3) for mulitpath paralleling model (not containing common path) shown in Figure 10

In conducting path shown in Figure 10, there are many risk conducting paths in risk source k after taking place, and every paths all is the series model of individual paths, and does not have common path.

Adopt 4.7 pairs of middle risk point orders of severity shown in Figure 10 of following formula to calculate, the middle risk point order of severity is the optimum solution of Optimization Model formula 4.7.

$\underset{A_i}{\min }[{(A_1-A_2)}^2+{(A_2-A_3)}^2+{(A_5-A_4)}^2]$

$s.t.\left\{\begin{array}{cc}{A}_3^k-A_2^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_2^k-A_1^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_5^k-A_4^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_1^k+A_2^k+A_3^k=A& \\ A_4^k+A_5^k=A& \\ A_i^k\&GreaterEqual;0& \&ForAll;i\end{array}\right.$ (formula 4.7)

The risk source order of severity in the hearsay guiding path shown in Figure 10 is calculated; Wherein the implication of

representative is identical with implication under first kind of situation; When diverse location came to light, all scenario of the risk point that is shown was following after risk source k takes place:

$\begin{array}{cc}{h}_1^k& H_1^k=0\\ h_2^k& H_2^k=A_1^k\\ h_3^k& H_3^k=A_1^k+A_2^k\\ h_4^k& H_4^k=0\\ h_5^k& H_5^k{=A}_4\\ h_r^k& H_r^k=A_4^k+A_5^k=A_1^k+A_2^k+A_3^k=A\\ n_k=\underset{i=1}{\overset{5,r}{\mathrm{\&Sigma;}}}{h}_i& \end{array}$

At this moment, but risk source order of severity through type 4.8 calculate.

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\frac{h_2^k}{n_k}{H}_2^k+\frac{h_3^k}{n_k}{H}_3^k+\frac{h_4^k}{n_k}{H}_4^k+\frac{h_5^k}{n_k}{H}_5^k+\frac{h_r^k}{n_k}{H}_r^k$

(formula 4.8).

$=\frac{h_1^k}{n_k}*0+\frac{h_2^k}{n_k}{A}_1^k+\frac{h_3^k}{n_k}(A_1^k+A_2^k)+\frac{h_4^k}{n_k}*0+\frac{h_5^k}{n_k}{A}_4^K+\frac{h_r^k}{n_k}A$

(3) shown in Figure 11 is a kind of situation (situation 4) that contains the conduction mode of common path

On conducting path shown in Figure 11; The risk that risk source k takes place to produce can influence the generation of final consequence through mulitpath; But the risk source has only a follow-up risk point; Middle risk point can a plurality of later on follow-up risk points, in risk conducting path network, show as the risk source with tightly afterwards the limit between the risk point be the common path of many conducting paths.

The middle risk point order of severity in 4.9 pairs of conducting paths shown in Figure 11 of through type is calculated, and the middle risk point order of severity is the optimum solution of Optimization Model formula 4.9:

$\underset{A_i}{\min }\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}I(i,j){(A_i^k-A_j^k)}^2$

$s.t.\left\{\begin{array}{cc}{A}_3^k-A_2^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_2^k-A_1^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_5^k-A_4^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_6^k-A_2^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_7^k-A_6^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_1^k+A_2^k+A_3^k+A_4^k+A_5^k=A& \\ A_1^k+A_2^k+A_6^k+A_7^k=A& \\ A_i^k\&GreaterEqual;0& \&ForAll;i\end{array}\right.$ (formula 4.9)

In the conducting path shown in Figure 11: when diverse location came to light, all scenario of the risk point that is shown was following after risk source k takes place:

$\begin{array}{cc}{h}_1^k& H_1^k=0\\ h_2^k& H_2^k=A_1^k\\ h_3^k& H_3^k=A_1^k+A_2^k\\ h_4^k& H_4^k=A_1^k+A_2^k+A_3^k\\ h_5^k& H_5^k=A_1^k+A_2^k+A_3^k+A_4^k\\ h_6^k& H_6^k=A_1^k+A_2^k\\ h_7^k& H_7^k=A_1^k+A_2^k+A_6^k\\ h_r^k& H_r^k=A_1^k+A_2^k+A_6^k+A_7^k=A\\ n_k=\underset{i=0}{\overset{7,r}{\mathrm{\&Sigma;}}}{h}_i& \end{array}$

Wherein

expression interception number of times can be added up according to concrete non-safety information data.To risk source order of severity B _kUtilize formula 4.10 to calculate.

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\frac{h_2^k}{n_k}{H}_2^k+\frac{h_3^k}{n_k}{H}_3^k+\frac{h_4^k}{n_k}{H}_4^k+\frac{h_5^k}{n_k}{H}_5^k+\frac{h_6^k}{n_k}{H}_6^k+\frac{h_7^k}{n_k}{H}_7^k$

$=\frac{h_1^k}{n_k}*0+\frac{h_2^k}{n_k}{A}_1^k+\frac{h_3^k}{n_k}(A_1^k+A_2^k)+\frac{h_4^k}{n_k}(A_1^k+A_2^k+A_3^k)+\frac{h_5^k}{n_k}(A_1^k+A_2^k+A_3^k+A_4^k)$ (formula 4.10).

$+\frac{h_6^k}{n_k}(A_1^k+A_2^k)+\frac{h_7^k}{n_k}(A_1^k+A_2^k+A_6^k)+\frac{h_r^k}{n_k}A$

(4) shown in Figure 12 is the another kind of situation (situation 5) that contains the conduction mode of common path

On conducting path shown in Figure 12, can influence the generation of net result after the risk source takes place through mulitpath, but the risk source there is a plurality of tight backs risk point, in the conduction network, shows as common path for not comprising risk source risk point.

The middle risk point order of severity in the conducting path of 4.11 pairs of conduction modes shown in Figure 12 of through type is calculated, and the middle risk point order of severity is the optimum solution of Optimization Model formula 4.11:

$\underset{A_i}{\min }\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}I(i,j){(A_i^k-A_j^k)}^2$

$s.t.\left\{\begin{array}{cc}{A}_3^k-A_2^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_2^k-A_1^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_5^k-A_4^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_6^k-A_3^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_6^k-A_5^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_7^k-A_6^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_4^k+A_5^k+A_6^k+A_7^k=A& \\ A_1^k+A_2^k+A_3^k+A_6^k+A_7^k=A& \\ A_i^k\&GreaterEqual;0& \&ForAll;i\end{array}\right.$ (formula 4.11)

In conducting path shown in Figure 12, when diverse location came to light, all scenario of the risk point that is shown was following after risk source k takes place:

$\begin{array}{cc}{h}_1^k& H_1^k=0\\ h_2^k& H_2^k=A_1^k\\ h_3^k& H_3^k=A_1^k+A_2^k\\ h_4^k& H_4^k=0\\ h_5^k& H_5^k=A_4^k\\ h_6^k& H_6^k=A_1^k+A_2^k+A_3^k\\ h_7^k& H_7^k=A_1^k+A_2^k+A_3^k+A_6^k\\ h_r^k& H_r^k=A_1^k+A_2^k+A_3^k+A_6^k+A_7^k=A\\ n_k=\underset{i=0}{\overset{7,r}{\mathrm{\&Sigma;}}}{h}_i& \end{array}$

At this moment, the risk source order of severity through type in the conducting path shown in Figure 12 4.12 calculates.

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\frac{h_2^k}{n_k}{H}_2^k+\frac{h_3^k}{n_k}{H}_3^k+\frac{h_4^k}{n_k}{H}_4^k+\frac{h_5^k}{n_k}{H}_5^k+\frac{h_6^k}{n_k}{H}_6^k+\frac{h_7^k}{n_k}{H}_7^k+\frac{h_r^k}{n_k}{H}_r^k$

$=\frac{h_1^k}{n_k}*0+\frac{h_2^k}{n_k}{A}_1^k+\frac{h_3^k}{n_k}(A_1^k+A_2^k)+\frac{h_4^k}{n_k}*0+\frac{h_5^k}{n_k}{A}_4^k+\frac{h_6^k}{n_k}(A_1^k+A_2^k+A_3^k)$ (formula 4.12).

$+\frac{h_7^k}{n_k}(A_1^k+A_2^k+A_3^k+A_6^k)+\frac{h_r^k}{n_k}A$

(5) shown in Figure 13 is the complicated case (situation 6) that contains the conduction mode of common path:

Risk conducting path pattern shown in Figure 13 has comprised several kinds of situations in front, and can there be a plurality of tight backs risk point in the risk source, and arbitrary middle risk point has a plurality of tight backs risk point.

The middle risk point order of severity in the conducting path of 4.13 pairs of conduction modes shown in Figure 13 of through type is calculated, and the middle risk point order of severity is the optimum solution of Optimization Model formula 4.13:

$\underset{A_i}{\min }\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}I(i,j){(A_i^k-A_j^k)}^2$

$s.t.\left\{\begin{array}{cc}{A}_3^k-A_2^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_2^k-A_1^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_5^k-A_4^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_3^k-A_4^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_6^k-A_5^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_7^k-A_6^k\&GreaterEqual;\mathrm{\&delta;}>0& \\ A_4^k+A_5^k+A_6^k+A_7^k=A& \\ A_1^k+A_2^k+A_3^k+A_6^k+A_7^k=A& \\ A_4^k+A_3^k+A_6^k+A_7^k=A& \\ A_i^k\&GreaterEqual;0& \&ForAll;i\end{array}\right.$ (formula 4.13)

Under conduction mode shown in Figure 13, the risk that risk source k triggers is following by the order of severity and frequency that interception shows at diverse location:

$\begin{array}{cc}{h}_1^k& H_1^k=0A_1^k\\ h_2^k& H_2^k=A_1^k\\ h_3^k& H_3^k=A_1^k+A_2^k\\ h_4^k& H_4^k=0\\ h_5^k& H_5^k=A_4^k\\ h_6^k& H_6^k=A_1^k+A_2^k+A_3^k\\ h_7^k& H_7^k=A_1^k+A_2^k+A_3^k+A_6^k\\ h_7^k& H_r^k=A_1^k+A_2^k+A_3^k+A_6^k+A_7^k=A\\ n_k=\underset{i=0}{\overset{7,r}{\mathrm{\&Sigma;}}}{h}_i& \end{array}$

At this moment, the risk source order of severity is calculated through following formula 4.14.

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\frac{h_2^k}{n_k}{H}_2^k+\frac{h_3^k}{n_k}{H}_3^k+\frac{h_4^k}{n_k}{H}_4^k+\frac{h_5^k}{n_k}{H}_5^k+\frac{h_6^k}{n_k}{H}_6^k+\frac{h_7^k}{n_k}{H}_7^k+\frac{h_r^k}{n_k}{H}_r^k$

$=\frac{h_1^k}{n_k}*0+\frac{h_2^k}{n_k}{A}_1^k+\frac{h_3^k}{n_k}(A_1^k+A_2^k)+\frac{h_4^k}{n_k}*0+\frac{h_5^k}{n_k}{A}_4^k+\frac{h_6^k}{n_k}(A_1^k+A_2^k+A_3^k)$ (formula 4.14).

$+\frac{h_7^k}{n_k}(A_1^k+A_2^k+A_3^k+A_6^k)+\frac{h_r^k}{n_k}A$

Obtain under the various conduction modes in aforementioned calculation; After the order of severity (being equivalent to the independent order of severity) of each risk point (risk point and risk source in the middle of comprising); Can calculate the whole order of severity of each risk point; Can (the whole order of severity of risk point is defined as: the whole order of severity of risk point is meant to take all factors into consideration investigates the order of severity after as the order of severity of risk source and middle risk point of risk point in the time period according to the definition of the whole order of severity of risk point; Be that the whole order of severity comes from two aspects, the one, the order of severity that shows as the risk source, the 2nd, as other the order of severity that middle risk point showed in risky source.Therefore the whole order of severity of risk point is the risk source order of severity of this risk point and the weighted sum of the middle risk point order of severity), utilize the whole order of severity of 4.15 pairs of risk points of following formula to calculate:

$r_i={\mathrm{\&beta;}}_i{B}_i+\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_k{A}_i^k$

β _i＝n _i/N _i

(formula 4.15)

${\mathrm{\&alpha;}}_k=n_i^k/N_i$

$N_i=n_i+\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{n}_i^k$

In the formula 4.15, N _iExpression risk point i investigates the total degree that non-safety information relevant with the target latent consequences in the time period takes place,

n _i: risk point i is investigating the number of times that takes place as the risk source in the time period, can obtain through the non-safety information of collecting is added up.

risk point i is the frequency of the middle risk point of risk source k the most.Therefore risk point i can need add up respectively each risk source simultaneously as the middle risk point in a plurality of risks source.

M: investigating has its conducting path of generation back, m risk source to comprise risk point i, therefore

in the time period

Obtain the whole order of severity of each risk point in aforementioned calculation after; Can calculate the comprehensive order of severity of each risk point; Can (the comprehensive order of severity definition of risk point---risk point of consideration of multiplicity possibly have than the higher whole order of severity in the time period though investigate according to the definition of the comprehensive order of severity of risk point; If but the possibility that this risk point takes place is very little, the number of times that just takes place seldom, its comprehensive order of severity maybe can't be lower than those whole orders of severity so; But frequency; The relative possibility that promptly takes place is higher than the comprehensive order of severity of higher risk point, so the comprehensive severity of risk point is the order of severity of considering after the multiplicity that risk point takes place, representes the comprehensive order of severity of risk point i with R.), investigate the consequence type that will study to institute in the time period, the comprehensive order of severity of a certain risk point can be utilized following formula 4.16 calculating:

$R_i={\mathrm{\&eta;}}_i{r}_i=(N_i/\underset{i}{\mathrm{\&Sigma;}}{N}_i)r_i$ (formula 4.16)

In the formula 4.16, N _iBe the total degree relevant of risk point i generation in the investigation time period with this latent consequences,

For investigating the total degree that all relevant with this latent consequences in time period risk points take place.

Can calculate the comprehensive order of severity of each risk point of the latent consequences that is directed against " low on fuel " through above-mentioned steps 3, thereby can carry out step 4, carry out Risk-warning according to the order of severity of risk point.Specifically can sort by the comprehensive order of severity under the corresponding serious grade of the latent consequences of " low on fuel " to all relevant risk points; Thereby draw the highest risk point of the comprehensive order of severity; As the early warning risk point; Also can carry out Risk-warning to the risk point that reaches early-warning conditions according to the early-warning conditions of setting.

On the basis of above-mentioned steps 3; Can also continue the comprehensive order of severity of calculation risk point under each serious grade as required; The order of severity of risk point under consequence seriousness grade equals it under this grade; Relatively the weighted mean of the order of severity of each latent consequences with, wherein weight be the generation accounting of all latent consequences under relative this seriousness grade of this latent consequences.The example that is calculated as with the comprehensive order of severity of each risk point under 7 grades of serious grades.Suppose that the latent consequences under this rank has m, respectively to be numbered 1 to m (each latent consequences all has corresponding numbers), with E _{I, 7}The comprehensive order of severity of expression risk point i under this severity level, R _{I, j}Be illustrated in the j class incident under this severity level, the comprehensive order of severity of risk point i, R _{I, j}Value in above-mentioned each step, calculate, with Num _jRepresent the frequency of 7 times non-safety informations relevant of serious grade, then have with j class incident:

$E_{i,7}=\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{\&pi;}}_j{R}_{i,j},$ Wherein ${\mathrm{\&pi;}}_j={\mathrm{Num}}_j/\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{Num}}_k.$

Can calculate the comprehensive order of severity of the overall situation of same risk point under the latent consequences of the serious grade of difference (being the comprehensive order of severity of the overall situation of risk point) through following formula.Thereby can press the ordering rule of dictionary preface according to the comprehensive order of severity of this overall situation of risk point, when comparing the sequencing of two risk points; The comprehensive order of severity that can at first under the 7th grade of seriousness rank, compare two risk points; If inequality, can confirm that then the ordering of the risk point that the comprehensive order of severity is high is forward, then need not to continue again comparison between these two risk points; If the order of severity of two risk points is identical under this rank, then compare other order of severity of next stage.And then can obtain the comprehensive ordering relatively (the low more expression ordering of value is forward more) under each risk point global state, and then also can when carrying out step 4, carry out Risk-warning through the comprehensive order of severity of the overall situation of risk point.

Following combination is described further the present invention the processing procedure of the actual early warning of latent consequences of " low on fuel ".

Latent consequences through finishing collecting is the non-safety information of " low on fuel "; Assess each risk point order of severity of " low on fuel " this latent consequences relatively; Wherein employed non-safety information is not all non-safety informations in the time span, just the non-safety information relevant with " low on fuel ".

The data is collected from certain airline's operation control system, and with the relevant non-safety information of dispatch clearance flow process, time span is 200901～200908.The latent consequences of these non-safety informations is " low on fuel ".Here " low on fuel " is meant the situation that does not meet CA's standard of fuel according to the flight fuel oil; Be not to be to be not enough to support flight to fly to reach airport of destination at the flight course intermediate fuel oil; As long as be less than more than standard of fuel a certain amount of, outside the scope that allows, all belong to low on fuel.The non-safety information that provides in operation control center of CA mainly is the description to event procedure; What the latent consequences of incident is; The risk source, these information such as discovery risk point do not provide, and project team is after analyzing data; Tentatively confirmed latent consequences each data item, it has been satisfied risk point is carried out the data demand that seriousness is analyzed for the relevant non-safety information of " low on fuel ".

From totally 51 data of 200901～200908 of collecting, the specifying information of data is seen below table 5.Different month " low on fuel " non-safety information generations distribute shown in figure 14.

Table 5 low on fuel " the non-safety information table

For each bar non-safety information in the last table, in risk point order of severity when assessment, need following data item: 1. latent consequences, and the 2. risk of latent consequences conduction network, with the adjacency matrix stored in form, 3. risk point is found in the risk source 4..The above-mentioned data item of every non-safety information is all definite.Revised " low on fuel " risk conducting path network is as shown in Figure 4.

The treatment step of early warning is specially:

(1) confirms " low on fuel " latent consequences as final early warning

(2) table 6 is seen in the risk source in the non-safety information of confirming to assess, and record risky source of institute and corresponding non-safety information thereof.

Table 6 risk source statistics

The adjacency matrix of " low on fuel " correspondence is with L _{N * n+1}Expression (n=42 here).The corresponding serious grade of latent consequences of " low on fuel " latent consequences is 5 grades, representes its order of severity with letter C.

(3) calculate the order of severity in each risk source:

The order of severity in risk source is the order of severity performance through middle risk point, so the order of severity in calculation risk source, the order of severity of risk point in the middle of needing at first to calculate.

The calculating of the middle risk point order of severity is after definite latent consequences and risk source, from adjacency matrix, reads computation model, and then solving model, obtains the middle risk point order of severity of middle risk point under a certain risk source.With risk source 3 is example, the computation process of risk point in the middle of explaining.

(3.1) read risk point 3 and conduct sub-road network network to the risk of latent consequences:

For " low on fuel " this latent consequences, risk source 3 corresponding risk conduction subpath networks are illustrated in fig. 15 shown below.The risk source is a risk point 3, and middle risk point is followed successively by risk point 5, risk point 7 risk points 14, and latent consequences is " low on fuel ".When the risk source was risk point 3, the network that its corresponding risk conduction subpath constitutes was illustrated in fig. 15 shown below.

The computation model of risk point is as shown in the formula, adjustment parameter δ=1/42 in the middle of this moment.

$\underset{A_i}{\min }Z={(A_{14}^3-{\mathrm{<figure-callout\; id="7"\; label="A"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">A</figure-callout>}}_7^3)}^2+{(A_7^3-{\mathrm{<figure-callout\; id="5"\; label="A"\; filenames="HDA0000135449990000071.png,HDA0000135449990000072.png"\; state="\{\{state\}\}">A</figure-callout>}}_5^3)}^2$

$s.\mathrm{<figure-callout\; id="7"\; label="t\; A"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">t</figure-callout>}\left\{\begin{array}{c}{A}_{}^{}\end{array}\right.\left\{\begin{array}{c}{}_7^3-{\mathrm{<figure-callout\; id="5"\; label="A"\; filenames="HDA0000135449990000071.png,HDA0000135449990000072.png"\; state="\{\{state\}\}">A</figure-callout>}}_5^3\&GreaterEqual;\mathrm{\&delta;}\\ A_{14}^3-{\mathrm{<figure-callout\; id="7"\; label="A"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">A</figure-callout>}}_7^3\&GreaterEqual;\mathrm{\&delta;}\\ A_5^3+A_7^3+A_{14}^3=\mathrm{<figure-callout\; id="5"\; label="C\; A"\; filenames="HDA0000135449990000071.png,HDA0000135449990000072.png"\; state="\{\{state\}\}">C</figure-callout>}& \\ A_{}^{}{}_5^3\&GreaterEqual;0\\ {\mathrm{<figure-callout\; id="7"\; label="A"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">A</figure-callout>}}_7^3\&GreaterEqual;0\\ A_{14}^3\&GreaterEqual;0\end{array}\right.$

Separate above-mentioned model, in the time of can getting risk point 3 as the risk source, the order of severity such as the following table 4.8 of risk point in the middle of each.

Table 4.8. risk source 3 each middle risk point order of severity

The order of severity of the risk that risk source 3 causes when different risks come to light does

${\mathrm{<figure-callout\; id="5"\; label="H"\; filenames="HDA0000135449990000071.png,HDA0000135449990000072.png"\; state="\{\{state\}\}">H</figure-callout>}}_5^3=0,$ ${\mathrm{<figure-callout\; id="7"\; label="H"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">H</figure-callout>}}_7^3{=A}_5^3=0.3095C,$ ${\mathrm{<figure-callout\; id="14"\; label="H"\; filenames="HDA0000135449990000082.png"\; state="\{\{state\}\}">H</figure-callout>}}_{14}^3=A_5^3+A_7^3=0.6428C,$ ${\mathrm{<figure-callout\; id="3"\; label="H\; r"\; filenames="HDA0000135449990000061.png,HDA0000135449990000063.png"\; state="\{\{state\}\}">H</figure-callout>}}_r^{}=C$

Under other risk sources in the middle of each the computation process of risk point identical with risk source 3, the result of calculation in risky source as shown in table 7 below.

The middle risk point order of severity in each risk source of table 7

(4) add up the position of the discovery risk point of 3 times every non-safety informations of this risk source risk point, and the interception number of times, according to the risk source with find that risk point confirms to also have on the risk conducting path risk point in the middle of which, and add up its frequency.The risky source of institute to deserved risk intercepting position and risk point frequency statistics see Table 8 with table 9.

Table 8 is that the risk interception in each risk source distributes

Table 9 is the frequency statistics of risk source and each risk point

(4.1), calculate the order of severity in each risk source according to the computing formula of the risk source order of severity.

The computing formula of risk source 3 orders of severity is:

${\mathrm{<figure-callout\; id="3"\; label="B"\; filenames="HDA0000135449990000061.png,HDA0000135449990000063.png"\; state="\{\{state\}\}">B</figure-callout>}}_3=\frac{{\mathrm{<figure-callout\; id="5"\; label="h"\; filenames="HDA0000135449990000071.png,HDA0000135449990000072.png"\; state="\{\{state\}\}">h</figure-callout>}}_5^3}{n_3}{\mathrm{<figure-callout\; id="5"\; label="H"\; filenames="HDA0000135449990000071.png,HDA0000135449990000072.png"\; state="\{\{state\}\}">H</figure-callout>}}_5^3+\frac{{\mathrm{<figure-callout\; id="7"\; label="h"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">h</figure-callout>}}_7^3}{n_3}{\mathrm{<figure-callout\; id="7"\; label="H"\; filenames="HDA0000135449990000072.png,HDA0000135449990000073.png"\; state="\{\{state\}\}">H</figure-callout>}}_7^3++\frac{{\mathrm{<figure-callout\; id="14"\; label="h"\; filenames="HDA0000135449990000082.png"\; state="\{\{state\}\}">h</figure-callout>}}_{14}^3}{n_3}{\mathrm{<figure-callout\; id="14"\; label="H"\; filenames="HDA0000135449990000082.png"\; state="\{\{state\}\}">H</figure-callout>}}_{14}^3+\frac{h_r^3}{n_3}{\mathrm{<figure-callout\; id="3"\; label="H\; r"\; filenames="HDA0000135449990000061.png,HDA0000135449990000063.png"\; state="\{\{state\}\}">H</figure-callout>}}_r^{}$

$=\frac{1}{3}*0+0*0.3095C+0*0.6428C+\frac{2}{3}*C=0.6667C$

The order of severity in risky source as shown in table 10 below:

Table 10 is the risk source order of severity in each risk source

(4.2) calculate the whole order of severity of each risk point

According to whole order of severity computing formula; Calculate the whole order of severity of each risk point; The whole order of severity of each risk point equals the weighted sum of its risk source order of severity and the middle risk point order of severity; With risk point 19 is example (though risk point 19 is not in the risk conducting path network in risk source 3, even if the relative complex of its overall risk point is representative).28 is the total degree that risk point 19 takes place in the formula, and 21 is its frequency as the risk source, and other integers are as the frequency of the middle risk point in other risk sources.

$r_{19}=\frac{21}{28}{B}_{19}+\frac{2}{28}{A}_{19}^{15}+\frac{1}{28}{A}_{19}^{16}+\frac{4}{28}{A}_{19}^{38}$

$=\frac{21}{28}*0.095C+\frac{2}{28}*0.476C+\frac{1}{28}*0.476C+\frac{4}{28}*0.476C=0.190C$

The whole order of severity of other risk points is as shown in table 11.

Table 11 is the whole order of severity of each risk point and the comprehensive order of severity

(4.3) calculate the comprehensive order of severity of each risk point:

The comprehensive order of severity is on the basis of overall risk point, with the accounting of each risk point frequency as weight, the adjusted order of severity, as shown in table 12.

Several treatment steps more than the process; To " low on fuel " this latent consequences; The comprehensive order of severity of each risk point is confirmed; Because the risk point order of severity under the serious grade need can be confirmed after other latent consequences is also assessed, and does not therefore further calculate in the above-mentioned processing.

What obtain is as shown in table 12 for the comprehensive order of severity of each risk point of " low on fuel " to latent consequences.

Table 12 is each risk point order of severity assessment result of " low on fuel " for being directed against latent consequences

Contrast according to above-mentioned table 12; Can confirm to " low on fuel " this latent consequences; Influence in the risk point of operation control system safe condition; The order of severity of risk point 19 (information that is used to calculate flight oil mass, carrying capacity data is inaccurate) is the highest relatively, and risk point 15 (Computer flight plan system data maintenance, system management is untimely, inaccurate or method fault influences Computer flight plan and makes) takes second place.In the reality, can carry out early warning to risk point 19, risk point 15 and risk point 14 to " low on fuel " this latent consequences, thus these several risk points of emphasis monitoring.

Those skilled in the art can know; It is above-mentioned that only to carry out Risk-warning with the latent consequences of " low on fuel " be that example describes; Those skilled in the art can realize each risk point of other latent consequences is carried out Risk-warning according to thought of the present invention, repeat no more at this.

The above; Be merely the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, any technician who is familiar with the present technique field is in the technical scope that the present invention discloses; The variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.

Claims (13)

1. the method for prewarning risk based on flow process is characterized in that, may further comprise the steps:

Steps A is obtained the information of each included service processing node of business processing flow that operation control system carries out, finds out each service processing node that has security risk in the said information, sets up each flow process risk point according to each service processing node correspondence of finding out;

Step B; From the corresponding potential dangerous incident of said business processing flow; Selected potential dangerous incident of being monitored; From the flow process risk point, find out the relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model;

Step C assesses each relevant risk point according to the serious grade and the said combined data information of said risk conducting path model of setting up and the said potential dangerous incident obtained, confirms the order of severity of each relevant risk point;

Step D if the order of severity of relevant risk point meets or exceeds preset serious grade, then to said operation control system output early warning information, carries out early warning by said operation control system according to said early warning information.

2. the method for prewarning risk based on flow process according to claim 1 is characterized in that, said steps A specifically comprises:

Steps A 1 is obtained the information of each included service processing node of business processing flow that operation control system carries out;

Steps A 2 is found out each service processing node that comprises non-safety information in the said information, sets up risk point as the flow process risk point corresponding to each service processing node of finding out.

3. the method for prewarning risk based on flow process according to claim 1 is characterized in that, among the said step B, specifically comprises according to each relevant risk point and said potential dangerous event establishment risk conducting path model:

Step B1 confirms pairing each service processing node of each relevant risk point;

Step B2 finds out each corresponding service processing node of each relevant risk point conducts to said potential dangerous incident in business processing flow incidence relation;

Step B3 is connected to each relevant risk point by said incidence relation and forms risk conducting path model after the said potential dangerous incident.

4. the method for prewarning risk based on flow process according to claim 1 is characterized in that, among the said step B, the combined data information of obtaining the non-safety information of each relevant risk point of said risk conducting path model in the certain hour section comprises:

Each relevant risk point in the certain hour section during as the risk source of triggering said potential dangerous incident the frequency of its non-safety information and each relevant risk point it conducts discovery number of times of next non-safety information to other risk point during as middle the risk point that triggers said potential dangerous incident in the certain hour section.

5. the method for prewarning risk based on flow process according to claim 1 is characterized in that, among the said step B, the combined data information of obtaining the non-safety information of each relevant risk point of said risk conducting path model in the certain hour section also comprises:

The frequency in the frequency of each relevant risk point non-safety information relevant with said potential dangerous incident in the certain hour section, each relevant risk point risk source of the said potential dangerous incident of conduct triggering in the certain hour section and each relevant risk point are as the frequency of the middle risk point in each risk source of triggering said potential dangerous incident.

6. the method for prewarning risk based on flow process according to claim 1 is characterized in that, said step C specifically comprises:

Step C1 from each relevant risk point of the said risk conducting path model set up, finds out the conducting path that conducts to said potential dangerous incident as the risk point in risk source through middle risk point; And the frequency that obtains the non-safety information in risk source in the said combined data information with each in the middle of the discovery number of times of risk point non-safety information that other risk point conduction is come;

Step C2; The frequency of the conducting path that finds according to step C1, the non-safety information in risk source and in the middle of each risk point to the discovery number of times of the non-safety information of other risk point conduction; And the serious grade of said potential dangerous incident; The independent order of severity of risk point in the middle of calculating through following assessment models one, assessment models one is:

$\underset{A_i^k}{\min }\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}I(i,j){(A_i^k-A_j^k)}^2$

In the above-mentioned model; δ is the adjustment parameter; The minimal difference of representing two adjacent middle risk point degrees of risk; The partial ordering relation that risk point satisfies in the middle of formula (1) expression, all middle risk point contribution degree sums equal the constraint condition of the serious grade of potential dangerous incident on conducting path of formula (2) expression, and A is the serious grade of said potential dangerous incident;

Step C3, the independent order of severity of the middle risk point that obtains according to step C2 calculates the independent order of severity B in risk source through following assessment models two _k, assessment models two is:

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_i^k}{n_k}{H}_i^k+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_r}{n_k}{H}_r^k$

$=\frac{h_1^k}{n_k}*0+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_i^k}{n_k}*(A_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+A_{i-1}^k)+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_r^k}{n_k}A$

In the above-mentioned model two, n _kFor the frequency of the non-safety information of risk source k, promptly with the non-safety information frequency of risk point k as the risk source;

is the number of times that non-safety information is found by risk point 1 place operation; The actual generation of risk source k this moment, the degree of risk of risk source k is

Be the number of times that non-safety information is found by the operation of risk point i place, this moment, the degree of risk of risk source k did $H_i^k=A_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+A_{i-1}^k;$

spread out of operation control system for non-safety information; Found beyond the operation control system or the actual generation of final potential dangerous incident; This moment risk source k degree of risk is

in this case; This moment, the order of severity of risk source k equaled A, and A is the serious grade of said potential dangerous incident;

n _kWith

Between satisfy relation:

Step C4, the independent order of severity that above-mentioned assessment is obtained is as the order of severity of each self-corresponding each relevant risk point.

7. the method for prewarning risk based on flow process according to claim 6 is characterized in that, among the said step C, after step C2, also comprises:

Step C20 based on the independent order of severity of each relevant risk point being assessed each the relevant risk point that obtains, carries out whole order of severity assessment to each relevant risk point, obtains the whole order of severity of each relevant risk point.

8. the method for prewarning risk based on flow process according to claim 7 is characterized in that, saidly each relevant risk point is carried out the assessment of the whole order of severity specifically comprises:

Step C201; Each relevant risk point is carried out whole order of severity assessment one by one; At first select a certain relevant risk point, from said combined data information, obtain the frequency of the middle risk point in the frequency of this relevant risk point non-safety information relevant in the certain hour section, the frequency in this relevant risk point risk source of the said potential dangerous incident of conduct triggering in the certain hour section and each risk source that the conduct of this relevant risk point triggers said potential dangerous incident with said potential dangerous incident;

Step C202; The frequency of this relevant risk point non-safety information relevant with said potential dangerous incident in the certain hour section that obtains according to step C201, this relevant risk point as the frequency in the risk source of triggering said potential dangerous incident and this relevant risk point frequency as the middle risk point in each risk source of triggering said potential dangerous incident, draw the whole order of severity r of this relevant risk point through following assessment models three calculating in the certain hour section _i, assessment models three is:

$r_i={\mathrm{\&beta;}}_i{B}_i+\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_k{A}_i^k$

β _i＝n _i/N _i

${\mathrm{\&alpha;}}_k=n_i^k/N_i$

$N_i=n_i+\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{n}_i^k$

In the above-mentioned assessment models three, N _iFrequency for risk point i non-safety information relevant in the certain hour section with said potential dangerous incident;

n _iBe risk point i number of times as the generation of risk source in the certain hour section;

is the frequency of risk point i as the middle risk point of risk source k;

M is illustrated in has m risk source to take place to comprise risk point i in its conducting path of back in the certain hour section;

Step 203 repeats above-mentioned steps 201,202 until the whole order of severity of obtaining each relevant risk point.

9. the method for prewarning risk based on flow process according to claim 7 is characterized in that, among the said step C, after step C20, also comprises:

Step C21 according to the independent order of severity and the whole order of severity of each relevant risk point being assessed each the relevant risk point that obtains, carries out comprehensive order of severity assessment to each relevant risk point, obtains the comprehensive order of severity of each relevant risk point.

10. the method for prewarning risk based on flow process according to claim 9 is characterized in that, saidly each relevant risk point is carried out the assessment of the comprehensive order of severity specifically comprises:

Step C211; Each relevant risk point is carried out comprehensive order of severity assessment one by one; Selected a certain relevant risk point obtains the frequency of all risk points that this relevant risk point is correlated with said potential dangerous incident in the frequency relevant with said potential dangerous incident, the certain hour section in the certain hour section from said combined data information;

Step C212; This relevant risk point that obtains according to step C211 in the certain hour section in the number of times relevant, the certain hour section with said potential dangerous incident with said potential dangerous incident the frequency of all relevant risk points and the whole order of severity of this risk point, calculate the comprehensive order of severity R of this relevant risk point through following assessment models four _i, assessment models four is:

$R_i={\mathrm{\&eta;}}_i{r}_i=(N_i/\underset{i}{\mathrm{\&Sigma;}}{N}_i)r_i$

In the above-mentioned assessment models four, N _iBe the number of times relevant of risk point i generation in the certain hour section with said latent consequences,

Frequency for all relevant in certain hour section risk points with said latent consequences;

Step C213 repeats above-mentioned steps C211, C212 until the comprehensive order of severity of obtaining each relevant risk point.

11. the method for prewarning risk based on flow process is characterized in that, may further comprise the steps:

Steps A is obtained the information of each included service processing node of business processing flow that operation control system carries out, finds out each service processing node that has security risk in the said information, corresponding each flow process risk point of setting up with each service processing node of finding out;

Step B; From the corresponding potential dangerous incident of said business processing flow; Selected a kind of potential dangerous incident; From said flow process risk point, find out each relevant risk point that can trigger this potential dangerous incident,, and obtain the combined data information of the non-safety information of each relevant risk point of said risk conducting path model in serious grade and the certain hour section of said potential dangerous incident according to each relevant risk point and said potential dangerous event establishment risk conducting path model;

Step C assesses each relevant risk point according to the serious grade and the said combined data information of said risk conducting path model of setting up and the said potential dangerous incident obtained, confirms the comprehensive order of severity of each relevant risk point;

Step D; Repeat above-mentioned steps B, the comprehensive order of severity of C all risk points in drawing said flow process risk point under the different potential dangerous incident of same serious grade, calculate the comprehensive order of severity of the overall situation of each risk point based on the order of severity of each risk point;

Step e if the comprehensive order of severity of the overall situation of risk point meets or exceeds the preset order of severity, then to said operation control system output early warning information, is carried out early warning by said operation control system according to said early warning information.

12. the method for prewarning risk based on flow process according to claim 11 is characterized in that, said step C specifically comprises:

Step C1 from each relevant risk point of the said risk conducting path model set up, finds out the conducting path that conducts to said potential dangerous incident as the risk point in risk source through middle risk point; And the frequency that obtains the non-safety information in risk source in the said combined data information with each in the middle of the discovery number of times of risk point non-safety information that other risk point conduction is come;

Step C2; The frequency of the conducting path that finds according to step C1, the non-safety information in risk source and in the middle of each risk point to the discovery number of times of the non-safety information of other risk point conduction; And the serious grade of said potential dangerous incident; The independent order of severity of risk point in the middle of calculating through following first assessment models, first assessment models is:

$\underset{A_i^k}{\min }\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}I(i,j){(A_i^k-A_j^k)}^2$

In above-mentioned first assessment models; δ is the adjustment parameter; Be the inverse of all risk point sums, represent the minimal difference of two adjacent middle risk point degrees of risk, the partial ordering relation that risk point satisfies in the middle of formula (1) expression; All middle risk point contribution degree sums equal the constraint condition of the serious grade of potential dangerous incident on conducting path of formula (2) expression, and A is the serious grade of said potential dangerous incident;

Step C3, the independent order of severity of the middle risk point that obtains according to step C2 calculates the independent order of severity B in risk source through following second assessment models _k, second assessment models is:

$B_k=\frac{h_1^k}{n_k}{H}_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_i^k}{n_k}{H}_i^k+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_r}{n_k}{H}_r^k$

$=\frac{h_1^k}{n_k}*0+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_i^k}{n_k}*(A_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+A_{i-1}^k)+\&CenterDot;\&CenterDot;\&CenterDot;+\frac{h_r^k}{n_k}A$

In above-mentioned second assessment models, n _kFor the frequency of the non-safety information of risk source k, promptly with the non-safety information frequency of risk point k as the risk source;

Be the number of times that non-safety information is found by risk point 1 place operation, the actual generation of risk source k this moment, the degree of risk of risk source k does $H_1^k=0;$

Be the number of times that non-safety information is found by the operation of risk point i place, this moment, the degree of risk of risk source k did $H_i^k=A_1^k+\&CenterDot;\&CenterDot;\&CenterDot;+A_{i-1}^k;$

spread out of operation control system for non-safety information; Found beyond the operation control system or the actual generation of final potential dangerous incident; This moment risk source k degree of risk is

in this case; This moment, the order of severity of risk source k equaled A, and A is the serious grade of said potential dangerous incident;

n _kWith

Between satisfy relation:

Step C4; Each relevant risk point is carried out whole order of severity assessment one by one; Select a certain relevant risk point, from said combined data information, obtain the frequency of the middle risk point in the frequency of this relevant risk point non-safety information relevant in the certain hour section, the frequency in this relevant risk point risk source of the said potential dangerous incident of conduct triggering in the certain hour section and each risk source that the conduct of this relevant risk point triggers said potential dangerous incident with said potential dangerous incident;

Step C5; The frequency of this relevant risk point non-safety information relevant with said potential dangerous incident in the certain hour section that obtains according to step C5, this relevant risk point as the frequency in the risk source of triggering said potential dangerous incident and this relevant risk point frequency as the middle risk point in each risk source of the said potential dangerous incident of triggering, calculate the whole order of severity r of this relevant risk point through following the 3rd assessment models in the certain hour section _i, the 3rd assessment models is:

$r_i={\mathrm{\&beta;}}_i{B}_i+\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_k{A}_i^k$

β _i＝n _i/N _i

${\mathrm{\&alpha;}}_k=n_i^k/N_i$

$N_i=n_i+\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{n}_i^k$

In above-mentioned the 3rd assessment models, N _iFrequency for risk point i non-safety information relevant in the certain hour section with said potential dangerous incident;

n _iBe risk point i number of times as the generation of risk source in the certain hour section;

is the frequency of risk point i as the middle risk point of risk source k;

Its conducting path comprised risk point i after m was illustrated in and has m risk source to take place in the certain hour section;

Step C6 repeats above-mentioned steps C4, C5 until the whole order of severity of obtaining each relevant risk point;

Step C7; Each relevant risk point is carried out comprehensive order of severity assessment one by one; Selected a certain relevant risk point obtains the frequency of all risk points that this relevant risk point is correlated with said potential dangerous incident in the number of times relevant with said potential dangerous incident, the certain hour section in the certain hour section from said combined data information;

Step C8; This relevant risk point that obtains according to step C7 in the certain hour section in the number of times relevant, the certain hour section with said potential dangerous incident with said potential dangerous incident the frequency of all relevant risk points and the whole order of severity of this risk point, calculate the comprehensive order of severity R of this relevant risk point through following the 4th assessment models _i, the 4th assessment models is:

$R_i={\mathrm{\&eta;}}_i{r}_i=(N_i/\underset{i}{\mathrm{\&Sigma;}}{N}_i)r_i$

In above-mentioned the 4th assessment models, N _iBe the number of times relevant of risk point i generation in the certain hour section with said latent consequences,

is the frequency of all risk points relevant with said latent consequences in the certain hour section;

Step C9 repeats above-mentioned steps C7, C8 until the comprehensive order of severity of obtaining each relevant risk point.

13. the method for prewarning risk based on flow process according to claim 12 is characterized in that, among the said step D, the comprehensive order of severity of the overall situation that calculates each risk point according to the order of severity of each risk point specifically comprises:

Step D1; Each relevant risk point is carried out comprehensive order of severity assessment one by one; Selected a certain risk point; According to the frequency of the relevant non-safety information of the different potential dangerous incident of this risk point that obtains under the comprehensive order of severity under the different potential dangerous incident of same serious grade and this same serious grade of obtaining, calculate the comprehensive order of severity E of the overall situation of this risk point through following the 5th assessment models _{I, x}, the 5th assessment models is:

$E_{i,x}=\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{\&pi;}}_j{R}_{i,j},$ Wherein ${\mathrm{\&pi;}}_j={\mathrm{Num}}_j/\underset{k=1}{\overset{m}{\mathrm{\&Sigma;}}}{\mathrm{Num}}_k$

In above-mentioned the 5th assessment models, R _{I, j}Be the potential dangerous incident of the j class under this severity level, the comprehensive order of severity of risk point i; X is the grade of said dangerous incident; Num _jBe the frequency of non-safety information relevant with the potential dangerous incident of j class under this serious grade, m is the quantity of the potential dangerous incident that comprises under this serious grade.

Images