CN103281207A - Method for calculating decision-making system web service capability under SOA - Google Patents

Abstract

The invention discloses a method for calculating decision-making system web service capability under an SOA. Firstly, after receiving a decision task submitted by a user, a service capability evaluation module calculates goal-achieving capability value of the corresponding web service according to the number of goals and the goal importance degree both of which can be achieved by different decision web services. Secondly, time-achieving capability value of the web service is calculated according to the time of finishing decision making by the decision-making web service. Lastly, the cost-price capability value of the web service is calculated according to the cost needed for finishing the decision making task by the decision-making web service. By means of the method, the goal-achieving capability value, the time-achieving capability value and the cost-price capability value are achieved and are calculated in a weighting mode to obtain the web service capability. According to the method for calculating the decision making system web service capability under the SOA, in the network, a caller can automatically recognize the capability of the decision-making web service, the purpose that the capability of the decision-making web service under the network environment is calculated is achieved.

Description

A kind of computational methods of the SOA making policy decision web of system service ability

Technical field

The present invention relates to computer network decision-making field, more particularly, relate to the ability computational methods of decision-making web service in a kind of internet environment.

Background technology

The DSS development is moved towards the network cooperation decision-making from the unit DSS for many years at present.This transformation shows as highlightedly by network environment, and the solving model of DSS is carried out shared collaboration, finds the solution for complex task and serves.

In the evolution of network DSS, and Enterprise SOA (Service Oriented Architecture---SOA) be an important solution.SOA impels each the function decision-making solving model in the implementation decision back-up system to change the Web service form into, formation has the independent software module of calculating the ability of finding the solution, is deployed in the different network DSS terminals by distributed, loosely-coupled mode.When implementing the network decision-making, the user can have decision making function web service with these by certain mode and organize, and finishes the decision-making of complex task by cooperation mode.

Single decision system is to the variation of SOA decision support pattern as shown in Figure 1, and the basic procedure of computer aided decision support system is: decision task-determine decision objective-trade-off decision solving model-implementation decision occurs and find the solution.In single DSS, after decision task occurred, DSS was within it in the decision model storehouse of portion, and trade-off decision solving model, these models are software modules independently, find the solution ability work according to its software.Under the SOA environment, DSS has been broken away from the restriction of single DSS, has realized the decision-making solving model shared collaboration between a plurality of decision systems in the network.Among the figure, the center service terminal that the service ability evaluation module is installed realizes two-way communication by the Internet and a plurality of decision-making terminal that has the heterogeneous networks decision system, and the network decision system in each decision-making terminal all has decision-making web service library separately.After the user is committed to task service ability evaluation module in the center service terminal, will above-mentionedly have the task that user's submission was organized and finished in decision making function web service by the service ability evaluation module according to corresponding principle of decision-making.In order to satisfy the basic demand of SOA under the network environment, the decision-making solving model of these decision systems is converted into the web service software form that meets the SOA standard, has standardized interface, can obtain effectively calling of other decision systems.The decision service of each network decision-making end all can be used as the candidate, by the assessment of service ability, selects and meets task needs person, participative decision making most.

In the present invention, the decision-making of these network decision systems is found the solution software model and is converted into web service under the SOA, is referred to as decision-making web service.

A key technology that realizes decision-making web service combination is accurately to assess the ability size of these web services, judges whether it is fit to certain concrete decision-making.The merit rating of tradition web service mainly depends on means such as interface satisfies, target satisfies, and can't carry out comprehensive assessment to the performance of whole web service, especially the ability of time dimension, cost dimension is calculated more to lack.Therefore, a kind of decision-making web service ability computational methods of composite type are badly in need of in network decision-making, can various dimensions ground estimate the ability of a service, promote the effect of making a strategic decision.

Summary of the invention

At the defective that exists in the prior art, the purpose of this invention is to provide a kind of computational methods of the SOA making policy decision web of system service ability.

For achieving the above object, the present invention adopts following technical scheme:

A kind of computational methods of the SOA making policy decision web of system service ability, the concrete steps of these computational methods are:

A. the service ability evaluation module calculates the realization target capability value that corresponding web serves according to different decision-making web the service destination number that can realize and the target significance level that can realize after receiving the decision task that the user submits to;

B. serve the realization time capacity value that the Time Calculation that can finish decision task obtains this web service according to decision-making web;

C. finish the cost price ability value that the needed cost price value of decision task calculates this web service according to decision-making web service;

D. by realization target capability value, the realization time capacity value of step B and the cost price ability value of step C of steps A, weighted calculation obtains decision-making web service ability;

The decision task that user in the steps A submits to is: and Task=(Task_goal, Task_time, Task_cost), wherein:

Task_goal=(tg ₁, tg ₂...) and be the set of decision task sub-goal;

Task_time=(max_time, best_time (tg ₁), best_time (tg ₂) ...) be that the corresponding time requirement of each decision-making sub-goal is gathered;

Task_cost=(max_cost, best_cost (tg ₁), best_cost (tg ₂) ...) cost that can bear for each decision-making sub-goal institute gathers;

Decision task sub-goal set Task_goal=(tg ₁, tg ₂...) and in each sub-goal be designated as tg _i(being designated as i sub-goal down), this sub-goal correspondence has an optimal period that allows to finish this sub-goal and treats time best_time (tg _i), the optimum cost of the corresponding expectation of this sub-goal is best_cost (tg _i); Simultaneously, max_time refers to the maximum permission time total amount of making a strategic decision of finishing of decision task regulation; Max_cost refers to that the decision-making maximum of finishing of decision task regulation allows the cost total amount;

Decision-making web in steps A service is designated as and is represented as D, its ability that has be designated as D=(D_goal, D_time, D_cost), D_goal=(dg wherein ₁, dg ₂...) and for serving the sub-goal that can realize, gathers decision-making web; D_time=(D_time (dg ₁), D_time (dg ₂) ...) realize needed time requirement set for each decision-making sub-goal; D_cost=(D_cost (dg ₁), D_cost (dg ₂) ...) realize the set of needed cost for each decision-making sub-goal;

In described decision task, decision-making web service realizes sub-goal set D_goal=(dg ₁, dg ₂...) and in each sub-goal be designated as dg _j(be designated as j sub-goal down, value is positive integer), this sub-goal correspondence has time D _ time (dg that needs are finished this sub-goal _j), this sub-goal finish in requisition for cost be D_cost (dg _j).

The concrete steps of described steps A are:

Suppose to exist target x and target y, then adaptation function N of equal value (x) → y refers to that x and y are identical; Described adaptation function of equal value is used for defining the equivalence relation between the sub-goal that calculates sub-goal that decision-making web service can realize and decision task; If dg _jWith a tg _iEquivalence so just is designated as N (dg _j) → tg _i, mean tg _iCan be by the dg of decision-making web service _jFinish;

The realization target capability value of decision-making web service D is designated as score_goal, and value is the real number between the 0-1; Assess according to two standard extended targets: the destination number that can realize and the target significance level that can realize;

For decision task subclass Task_goal=(tg ₁, tg ₂...) and in each tg _i(being designated as i target down), the weight that it has is designated as w (tg _i) (w (tg _i) span be the real number of 0-1), and all target tg _iWeight satisfy condition

The realization target capability of this decision-making web service may be calculated so:

$\mathrm{score}\_\mathrm{goal}\left(D\right)=\left\{\begin{array}{cc}\underset{N\left({\mathrm{dg}}_j\right)\→{\mathrm{tg}}_i}{\mathrm{\Σ}}w\left({\mathrm{tg}}_i\right)\×{\left(\frac{m}{n}\right)}^{(1-\frac{1}{m})}& m\≥2\\ \underset{N\left({\mathrm{dg}}_j\right)\→{\mathrm{tg}}_i}{\mathrm{\Σ}}w\left({\mathrm{tg}}_i\right)\×\frac{1}{n}& m=1\end{array}\right.$

In the following formula, n is the quantity that comprises all targets among the decision task subclass Task_goal, and m satisfies N (dg among the decision-making web service sub-goal set D_goal that can realize _j) → tg _iDg _jQuantity.

The concrete steps of described step B are:

Realize whether the time capacity web service D that represents to make a strategic decision can finish decision-making in the decision task official hour.More short decision-making deadline response, the time capacity value of acquisition is more high.To be designated as the max_time(value be the arithmetic number type to the maximum permission time total amount of decision-making of finishing of decision task regulation, expression clock cycle quantity), for each decision-making subtask tg _iThe optimal finish time amount of expectation is best_time (tg _i) (value is the arithmetic number type, expression clock cycle quantity).The fulfillment capability value of decision-making web service D is designated as score_time (D).

D is to each sub-goal dg in decision-making web service _jThe decision-making time that can reach consume and to be D_time (dg _j).Suppose sub-goal dg _jWith a tg _iNamely there is N (dg in equivalence _j) → tg _i, D_time (dg so _j) be designated as D_time (tg _i).The time fulfillment capability value of D may be calculated so:

$\mathrm{score}\_\mathrm{time}\left(D\right)=\frac{m}{n}\×(1-\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}D\_\mathrm{time}{\left({\mathrm{tg}}_i\right)}^{\mathrm{\η}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\mathrm{match}(D\_\mathrm{time}\left({\mathrm{tg}}_i\right))}{\max \_\mathrm{time}})$

In following formula, η (η ∈ [0,1]) is factor of influence; N is the quantity that comprises all targets among the set Task_goal, and m refers to that D can satisfy the sub-goal quantity of task; Match (D_time (tg _i)) be an adaptation function, be used for calculating D_time (tg _i) and the expectation best_time (tg _i) between matching degree, this function calculation is as follows:

The concrete steps of described step C are:

The cost price ability value of decision-making web service D judges content is whether the needed price of decision service exceeds the scope that decision task allows, and the ability value that more low price obtains is more high; Suppose that it is arithmetic number that maximum cost that decision task can allow is designated as the max_cost(span), for each decision-making subtask tg _iThe optimum cost of expectation is best_cost (tg _i) (value is arithmetic number).D is to each sub-goal tg in decision-making web service _iNeeded cost is D_cost (tg _i), the fulfillment capability value of decision-making web service D is designated as score_cost (D), may be calculated:

$\mathrm{score}\_\cos t\left(D\right)=\frac{m}{n}\×(1-\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}D\_\cos t{\left({\mathrm{tg}}_i\right)}^{\mathrm{\η}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\mathrm{over}(D\_\cos t\left({\mathrm{tg}}_i\right))}{\max \_\cos t})$

In following formula, η (η ∈ [0,1]) is factor of influence, and n is the quantity that comprises all targets among the set Task_goal, and m refers to that D can satisfy the sub-goal quantity of task; Over (D_cost (tg _i)) be the premium function, be calculated as follows:

The concrete steps of described step D are:

The whole capability assessed value of decision-making web service D is designated as capacity (D), is calculated as follows:

capacity(D)=σ ₁×score_goal(D)+σ ₂×score_time(D)+σ ₃×score_cost(D)，

σ wherein _k(K=1,2,3) is weighted value (value is the real number of 0-1), and satisfies constraints $\underset{k=1}{\overset{3}{\mathrm{\Σ}}}{\mathrm{\σ}}_k=1.$

Compared with prior art, employing the present invention proposes a kind of computational methods of the SOA making policy decision web of system service ability, can make the web service of making a strategic decision in the Internet obtain caller to the automatic identification of its ability, realize the ability calculating of decision-making web service in a kind of internet environment, from this decision-making web service the realization target capability, realize time capacity and three dimensions of cost price ability, realize that the whole capability value calculates.

Description of drawings

Fig. 1 is that unit DSS of the prior art is to the differentiation schematic diagram of network DSS;

Fig. 2 is the flow chart of the computational methods of a kind of SOA making policy decision web of system service ability of the present invention.

Embodiment

Further specify technical scheme of the present invention below in conjunction with accompanying drawing and embodiment.

The process of decision-making is exactly at decision task, finds decision-making web service to find the solution, and reaches desired target in the scope that decision task allows.

The present invention at first defines decision task and decision-making web service:

Define 1 decision task.Decision task refers to the problem that decision system need solve, and comprises target, the time range of permission and the cost that can bear that needs are realized.

Decision task be represented as Task=(Task_goal, Task_time, Task_cost), Task_goal=(tg wherein ₁, tg ₂...) and be the set of decision task sub-goal; Task_time=(max_time, best_time (tg ₁), best_time (tg ₂) ...) be that the corresponding time requirement of each decision-making sub-goal is gathered; Task_cost=(max_cost, best_cost (tg ₁), best_cost (tg ₂) ...) cost that can bear for each decision-making sub-goal institute gathers.

In above-mentioned decision task is described, decision task sub-goal set Task_goal=(tg ₁, tg ₂...) and in each sub-goal be designated as tg _i(being designated as i sub-goal down), this sub-goal correspondence has an optimal period that allows to finish this sub-goal and treats time best_time (tg _i), the optimum cost of the corresponding expectation of this sub-goal is best_cost (tg _i); Simultaneously, max_time refers to the maximum permission time total amount of making a strategic decision of finishing of decision task regulation; Max_cost refers to that the decision-making maximum of finishing of decision task regulation allows the cost total amount.

Definition 2 decision-making web services.Decision-making web service refers to the software model that can realize that decision task is found the solution, comprises the target that can reach, reaches the needed time of target and reach the needed price of target.

Decision-making web service is designated as and is represented as D, its ability that has be designated as D=(D_goal, D_time, D_cost), D_goal=(dg wherein ₁, dg ₂...) and for serving the sub-goal that can realize, gathers decision-making web; D_time=(D_time (dg ₁), D_time (dg ₂) ...) realize needed time requirement set for each decision-making sub-goal; D_cost=(D_cost (dg ₁), D_cost (dg ₂) ...) realize the set of needed cost for each decision-making sub-goal.

In above-mentioned decision task was described, decision-making web service realized sub-goal set D_goal=(dg ₁, dg ₂...) and in each sub-goal be designated as dg _j(be designated as j sub-goal down, value is positive integer), this sub-goal correspondence has time D _ time (dg that needs are finished this sub-goal _j), this sub-goal finish in requisition for cost be D_cost (dg _j).

Below, provide the present invention's computational methods that the web service ability estimates of making a strategic decision:

A kind of computational methods of the SOA making policy decision web of system service ability, the concrete steps of these computational methods are divided into:

A. the service ability evaluation module calculates the realization target capability value that corresponding web serves according to different decision-making web the service destination number that can realize and the target significance level that can realize after receiving the decision task that the user submits to;

B. serve the realization time capacity value that the Time Calculation that can finish decision task obtains this web service according to decision-making web;

C. finish the cost price ability value that the needed cost price value of decision task calculates this web service according to decision-making web service;

D. by realization target capability value, the realization time capacity value of step B and the cost price ability value of step C of steps A, weighted calculation obtains decision-making web service ability;

Traditional services performance evaluation is foundation with the quantity that realizes task object mostly, it is considered herein that, except realizing destination number, realizes time and cost that required by task will spend, all is very important.Therefore, than traditional services performance evaluation, the present invention is more comprehensive.

(1) the realization target capability value of decision-making web service is calculated

Realize that the calculating of decision task target capability is intended to estimate decision-making web service D and can finishes decision task on much degree.The present invention assesses according to two standard extended targets: the destination number that can realize and the target significance level that can realize.For this reason, we at first provide a kind of definition of adaptation function of equal value:

Definition 3 is if exist target x and target y, and then adaptation function N of equal value (x) → y refers to that x and y are identical.

This equivalence adaptation function is used for defining the equivalence relation between the sub-goal that calculates sub-goal that decision-making web service can realize and decision task.If dg _jWith a tg _iEquivalence so just is designated as N (dg _j) → tg _i, mean tg _iCan be by the dg of decision-making web service _jFinish.

Decision-making web service is designated as D, and the decision objective set that it can be realized is designated as D_goal=(dg ₁, dg ₂...), each dg _jThe sub-goal that (subscript j is natural number, represents j sub-goal) can realize for D.The realization target capability value of decision-making web service D is designated as score_goal, and value is the real number between the 0-1.Assess according to two standard extended targets: the destination number that can realize and the target significance level that can realize.

Suppose to exist decision task set Task_goal=(tg ₁, tg ₂...).For each tg _i(being designated as i target down), the weight that it has is designated as w (tg _i) (w (tg _i) span be the real number of 0-1), and all target tg _iWeight satisfy condition

The realization target capability of this decision-making web service may be calculated so:

$\mathrm{score}\_\mathrm{goal}\left(D\right)=\left\{\begin{array}{cc}\underset{N\left({\mathrm{dg}}_j\right)\→{\mathrm{tg}}_i}{\mathrm{\Σ}}w\left({\mathrm{tg}}_i\right)\×{\left(\frac{m}{n}\right)}^{(1-\frac{1}{m})}& m\≥2\\ \underset{N\left({\mathrm{dg}}_j\right)\→{\mathrm{tg}}_i}{\mathrm{\Σ}}w\left({\mathrm{tg}}_i\right)\×\frac{1}{n}& m=1\end{array}\right.---\left(1\right)$

In the above-mentioned formula 1, n is the quantity that comprises all targets among the set Task_goal, and m satisfies N (dg among the set D_goal _j) → tg _iDg _jQuantity.

(2) the realization time capacity value of decision-making web service is calculated

Realize whether the time capacity web service D that represents to make a strategic decision can finish decision-making in the decision task official hour.More short decision-making deadline response, the time capacity value of acquisition is more high.To be designated as the max_time(value be the arithmetic number type to the maximum permission time total amount of decision-making of finishing of decision task regulation, expression clock cycle quantity), for each decision-making subtask tg _iThe optimal finish time amount of expectation is best_time (tg _i) (value is the arithmetic number type, expression clock cycle quantity).The fulfillment capability value of decision-making web service D is designated as score_time (D).

D is to each sub-goal dg in decision-making web service _jThe decision-making time that can reach consume and to be D_time (dg _j).Suppose sub-goal dg _jWith a tg _iNamely there is N (dg in equivalence _j) → tg _i, D_time (dg so _j) be designated as D_time (tg _i).The time fulfillment capability value of D may be calculated so:

$\mathrm{score}\_\mathrm{time}\left(D\right)=\frac{m}{n}\×(1-\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}D\_\mathrm{time}{\left({\mathrm{tg}}_i\right)}^{\mathrm{\η}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\mathrm{match}(D\_\mathrm{time}\left({\mathrm{tg}}_i\right))}{\max \_\mathrm{time}})---\left(2\right)$

In above-mentioned formula 2, η (η ∈ [0,1]) is factor of influence; N is the quantity that comprises all targets among the set Task_goal, and m refers to that D can satisfy the sub-goal quantity of task; Match (D_time (tg _i)) be an adaptation function, be used for calculating D_time (tg _i) and the expectation best_time (tg _i) between matching degree, this function calculation is as follows:

(3) the cost price ability value of decision-making web service calculates

The D cost price ability value of decision-making web service judges content is whether the needed price of decision service exceeds the scope that decision task allows.The ability value that more low price obtains is more high.Suppose that it is arithmetic number that maximum cost that decision task can allow is designated as the max_cost(span), for each decision-making subtask tg _iThe optimum cost of expectation is best_cost (tg _i) (value is arithmetic number).D is to each sub-goal tg in decision-making web service _iNeeded cost is D_cost (tg _i), the fulfillment capability value of decision-making web service D is designated as score_cost (D), may be calculated:

$\mathrm{score}\_\cos t\left(D\right)=\frac{m}{n}\×(1-\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}D\_\cos t{\left({\mathrm{tg}}_i\right)}^{\mathrm{\η}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\mathrm{over}(D\_\cos t\left({\mathrm{tg}}_i\right))}{\max \_\cos t})---\left(4\right)$

Here η is identical with parameter in the formula 2, and n is the quantity that comprises all targets among the set Task_goal, and m refers to that D can satisfy the sub-goal quantity of task; We are same over (D_cost (tg _i)) be the premium function, be calculated as follows:

(4) the whole capability value of decision-making web service is calculated

According to realization target capability, realization time capacity and the cost price ability value of decision-making web service D, the whole capability assessed value of decision-making web service D is designated as capacity (D), is calculated as follows:

capacity(D)=σ ₁×score_goal(D)+σ ₂×score_time(D)+σ ₃×score_cost(D) (6)

σ in the following formula 6 _k(K=1,2,3) is weighted value (value is the real number of 0-1), and satisfies constraints

Embodiment:

We have provided following example description time prediction and price evaluation calculates.Suppose to exist a decision task Task=(Task_goal, Task_time, Task_cost), wherein:

$\left\{\begin{array}{c}\mathrm{Task}\_\mathrm{goal}=({\mathrm{tg}}_1,{\mathrm{tg}}_2,{\mathrm{tg}}_3)\\ \mathrm{best}\_\mathrm{time}\left({\mathrm{tg}}_1\right)=4,\mathrm{best}\_\mathrm{time}\left({\mathrm{tg}}_2\right)=7,\mathrm{best}\_\mathrm{time}\left({\mathrm{tg}}_3\right)=6,\max \_\mathrm{time}=20\\ \mathrm{best}\_\cos t\left({\mathrm{tg}}_1\right)=10,\mathrm{best}\_\cos t\left({\mathrm{tg}}_2\right)=10,\mathrm{best}\_\cos t\left({\mathrm{tg}}_3\right)=15,\max \_\cos t=50\end{array}\right.,$

Two decision-making web service D1 and D2 are arranged simultaneously as candidate service, they are from two different decision systems, under the correlated performance:

$D1:\left\{\begin{array}{c}D\_\mathrm{goal}=({\mathrm{tg}}_1,{\mathrm{tg}}_2,{\mathrm{tg}}_3)\\ D\_\mathrm{time}\left({\mathrm{tg}}_1\right)=4,D\_\mathrm{time}\left({\mathrm{tg}}_2\right)=5,D\_\mathrm{time}\left({\mathrm{tg}}_3\right)=8\\ D\_\cos t\left({\mathrm{tg}}_1\right)=13,D\_\cos t\left({\mathrm{tg}}_2\right)=8,D\_\cos t\left({\mathrm{tg}}_3\right)=20\end{array}\right.$

$D2:\left\{\begin{array}{c}D\_\mathrm{goal}=({\mathrm{tg}}_1,{\mathrm{tg}}_3)\\ D\_\mathrm{time}\left({\mathrm{tg}}_1\right)=3,D\_\mathrm{time}\left({\mathrm{tg}}_3\right)=7\\ D\_\cos t\left({\mathrm{tg}}_1\right)=12,D\_\cos t\left({\mathrm{tg}}_3\right)=13\end{array}\right.$

Concrete parameter situation is as shown in table 1:

Table 1

It is as follows to obtain corresponding calculated value according to the above-mentioned parameter value:

$D1:\left\{\begin{array}{c}\mathrm{score}\_\mathrm{goal}\left(D1\right)=(0.5+0.2+0.3)\×{\left(\frac{3}{3}\right)}^{1-\frac{1}{3}}=1\\ \mathrm{score}\_\mathrm{time}\left(D1\right)=1\×(1-\frac{{(4+5+5)}^{0.7}+(0+0+2)}{20})=0.58\\ \mathrm{score}\_\cos t\left(D1\right)=1\×(1-\frac{{(13+8+20)}^{0.7}+(3+0+5)}{50})=0.57\\ \mathrm{capacity}\left(D1\right)=\frac{1}{3}\×(1+0.58+0.57)=0.72\end{array}\right.$

$D2:\left\{\begin{array}{c}\mathrm{score}\_\mathrm{goal}\left(D2\right)=(0.5+0.3)\×{\left(\frac{2}{3}\right)}^{1-\frac{1}{2}}=0.65\\ \mathrm{score}\_\mathrm{time}\left(D2\right)=\frac{2}{3}\×(1-\frac{{(3+7)}^{0.7}+(0+1)}{20})=0.46\\ \mathrm{score}\_\cos t\left(D2\right)=\frac{2}{3}\×(1-\frac{{(12+13)}^{0.7}(2+0)}{50})=0.51\\ \mathrm{capacity}\left(D2\right)=\frac{1}{3}\×(0.65+0.46+0.51)=0.54\end{array}\right.$

In example, service ability evaluation module in the center service terminal is according to the D1 that calculates and the capability evaluation value of D2, the option program of decision-making Web service can (for example ability value be high preferential by certain rule, but should rule can independently be determined by the user that the task that concrete enforcement is selected is submitted to), the decision-making web service that meets mission requirements is selected in decision, finds the solution thereby carry out decision task.Above-mentioned example can specifically be applied in the functional assessment to the different decision-making of network terminals, thereby is a complex task when selecting different decision-making terminals when solving, and needed performance analysis is offered help.

Those of ordinary skill in the art will be appreciated that, above embodiment illustrates purpose of the present invention, and be not as limitation of the invention, as long as in essential scope of the present invention, all will drop in the scope of claim of the present invention variation, the modification of the above embodiment.

Claims (6)

1. the computational methods of the SOA making policy decision web of a system service ability is characterized in that:

The concrete steps of these computational methods are:

A. the service ability evaluation module calculates the realization target capability value that corresponding web serves according to different decision-making web the service destination number that can realize and the target significance level that can realize after receiving the decision task that the user submits to;

B. serve the realization time capacity value that the Time Calculation that can finish decision task obtains this web service according to decision-making web;

C. finish the cost price ability value that the needed cost price value of decision task calculates this web service according to decision-making web service;

D. by realization target capability value, the realization time capacity value of step B and the cost price ability value of step C of steps A, weighted calculation obtains decision-making web service ability.

2. computational methods according to claim 1 is characterized in that:

The decision task that user in the steps A submits to is: and Task=(Task_goal, Task_time, Task_cost), wherein:

Task_goal=(tg ₁, tg ₂...) and be the set of decision task sub-goal;

Task_time=(max_time, best_time (tg ₁), best_time (tg ₂) ...) be that the corresponding time requirement of each decision-making sub-goal is gathered;

Task_cost=(max_cost, best_cost (tg ₁), best_cost (tg ₂) ...) cost that can bear for each decision-making sub-goal institute gathers;

Decision task sub-goal set Task_goal=(tg ₁, tg ₂...) and in each sub-goal be designated as tg _i(being designated as i sub-goal down), this sub-goal correspondence has an optimal period that allows to finish this sub-goal and treats time best_time (tg _i), the optimum cost of the corresponding expectation of this sub-goal is best_cost (tg _i); Simultaneously, max_time refers to the maximum permission time total amount of making a strategic decision of finishing of decision task regulation; Max_cost refers to that the decision-making maximum of finishing of decision task regulation allows the cost total amount;

Decision-making web in steps A service is designated as and is represented as D, its ability that has be designated as D=(D_goal, D_time, D_cost), D_goal=(dg wherein ₁, dg ₂...) and for serving the sub-goal that can realize, gathers decision-making web; D_time=(D_time (dg ₁), D_time (dg ₂) ...) realize needed time requirement set for each decision-making sub-goal; D_cost=(D_cost (dg ₁), D_cost (dg ₂) ...) realize the set of needed cost for each decision-making sub-goal;

In described decision task, decision-making web service realizes sub-goal set D_goal=(dg ₁, dg ₂...) and in each sub-goal be designated as dg _j(be designated as j sub-goal down, value is positive integer), this sub-goal correspondence has time D _ time (dg that needs are finished this sub-goal _j), this sub-goal finish in requisition for cost be D_cost (dg _j).

3. computational methods according to claim 2 is characterized in that:

The concrete steps of described steps A are:

Suppose to exist target x and target y, then adaptation function N of equal value (x) → y refers to that x and y are identical; Described adaptation function of equal value is used for defining the equivalence relation between the sub-goal that calculates sub-goal that decision-making web service can realize and decision task; If dg _jWith a tg _iEquivalence so just is designated as N (dg _j) → tg _i, mean tg _iCan be by the dg of decision-making web service _jFinish;

The realization target capability value of decision-making web service D is designated as score_goal, and value is the real number between the 0-1; Assess according to two standard extended targets: the destination number that can realize and the target significance level that can realize;

For decision task subclass Task_goal=(tg ₁, tg ₂...) and in each tg _i(being designated as i target down), the weight that it has is designated as w (tg _i) (w (tg _i) span be the real number of 0-1), and all target tg _iWeight satisfy condition

The realization target capability of this decision-making web service may be calculated so:

In the following formula, n is the quantity that comprises all targets among the decision task subclass Task_goal, and m satisfies N (dg among the decision-making web service sub-goal set D_goal that can realize _j) → tg _iDg _jQuantity.

4. computational methods according to claim 2 is characterized in that:

The concrete steps of described step B are:

Realize whether the time capacity web service D that represents to make a strategic decision can finish decision-making in the decision task official hour.More short decision-making deadline response, the time capacity value of acquisition is more high; To be designated as the max_time(value be the arithmetic number type to the maximum permission time total amount of decision-making of finishing of decision task regulation, expression clock cycle quantity), for each decision-making subtask tg _iThe optimal finish time amount of expectation is best_time (tg _i) (value is the arithmetic number type, expression clock cycle quantity); The fulfillment capability value of decision-making web service D is designated as score_time (D);

D is to each sub-goal dg in decision-making web service _jThe decision-making time that can reach consume and to be D_time (dg _j).Suppose sub-goal dg _jWith a tg _iNamely there is N (dg in equivalence _j) → tg _i, D_time (dg so _j) be designated as D_time (tg _i).The time fulfillment capability value of D may be calculated so:

In following formula, η (η ∈ [0,1]) is factor of influence; N is the quantity that comprises all targets among the set Task_goal, and m refers to that D can satisfy the sub-goal quantity of task; Match (D_time (tg _i)) be an adaptation function, be used for calculating D_time (tg _i) and the expectation best_time (tg _i) between matching degree, this function calculation is as follows:

5. computational methods according to claim 2 is characterized in that:

The concrete steps of described step C are:

The cost price ability value of decision-making web service D judges content is whether the needed price of decision service exceeds the scope that decision task allows, and the ability value that more low price obtains is more high; Suppose that it is arithmetic number that maximum cost that decision task can allow is designated as the max_cost(span), for each decision-making subtask tg _iThe optimum cost of expectation is best_cost (tg _i) (value is arithmetic number).D is to each sub-goal tg in decision-making web service _iNeeded cost is D_cost (tg _i), the fulfillment capability value of decision-making web service D is designated as score_cost (D), may be calculated:

In following formula, η (η ∈ [0,1]) is factor of influence, and n is the quantity that comprises all targets among the set Task_goal, and m refers to that D can satisfy the sub-goal quantity of task; Over (D_cost (tg _i)) be the premium function, be calculated as follows:

6. computational methods according to claim 2 is characterized in that:

The concrete steps of described step D are:

The whole capability assessed value of decision-making web service D is designated as capacity (D), is calculated as follows:

capacity(D)=σ ₁×score_goal(D)+σ ₂×score_time(D)+σ ₃×score_cost(D)，

σ wherein _k(K=1,2,3) is weighted value (value is the real number of 0-1), and satisfies constraints

Images