CN102270190B - Computer modeling and solving processing method of complex decision-making problem - Google Patents
Computer modeling and solving processing method of complex decision-making problem Download PDFInfo
- Publication number
- CN102270190B CN102270190B CN 201010193108 CN201010193108A CN102270190B CN 102270190 B CN102270190 B CN 102270190B CN 201010193108 CN201010193108 CN 201010193108 CN 201010193108 A CN201010193108 A CN 201010193108A CN 102270190 B CN102270190 B CN 102270190B
- Authority
- CN
- China
- Prior art keywords
- model
- knowledge
- scheme
- decision
- dictionary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
The invention relates to a computer modeling and solving processing method of a complex decision-making problem, wherein the method comprises the following steps of: 1) preprocessing an intelligent man-machine interaction agent based on problem comprehension; 2) intelligently modeling; and 3) solving the complex problem. Compared with the prior art, the invention has the advantages of being strong in dynamic adaptive capacity to problem change and capable of optimizing in real time, etc.
Description
Technical field
The present invention relates to a kind of modeling and method for solving, especially relate to a kind of microcomputer modelling of complicated decision-making problems and find the solution disposal route.
Background technology
Decision-making is recurrent a kind of activity in the modern enterprise operation control process, how to use effectively and management enterprise knowledge, supports the complex decision activity of enterprise, deals with problems timely and accurately, has become the life-and-death thing of enterprise.Research with computer science and infotech introducing decision science, and decision support system (DSS) (Decision Support System proposed first, be called for short DSS) this concept be in one piece of classical paper " A Framework for Information SystemDesign " of showing in the seventies in 20th century of A.Gorry and M.S.S.Morton, from then on, the hot issue studied of the scholars that how to assist the research of human decision behavior just to become global a plurality of fields for computing machine.
At present in the modeling of challenge and most correlative studys of finding the solution, its method thinking is normally understood mathematical model → expert that the challenge → expert of enterprise takes out problem along: expert and is worked out or call according to mathematical model and find the solution software → operating software and find the solution that a kind of like this thinking of problem carries out.This thinking is too dependent on the human expert for understanding and the modeling ability of management decision problem, it makes the process range of problem only limit to finding the solution among the software of mathematical model that the expert sets up and establishment on the one hand, famine carries out the dynamically adapting ability of respective handling at the problem variation, on the other hand, make the performance advantage of computing machine can not get effective performance again, computing machine is understood with the degree of support of modeling extremely limited for problem, therefore, in a single day the model and the competent process range of software that have exceeded the expert when problem, just must be re-recognized the characteristics of decision problem by the expert, again modeling, again find the solution, the real-time processing of decision problem can't realize at all.Therefore, the achievement under this thinking only, user more stable for conditions of problems can accurately expect, and has that comparatively sufficient to find the solution the management decision problem of time be more effective.
In the applied environment of present management decision problem, along with advancing by leaps and bounds of electronic information and network communications technology, people's business transaction activity no longer is subjected to the restriction of region, time, scattered and small-scale commercial activity exists in a large number, they are difficult to form enterprise commerce management activity regular, that concentrate, cause the enterprise management decision making problem conditional instability, can not survey even fast changing, this variation tend to exceed single mathematical model or computer software the scope that can handle; And as a rule; not only the client of enterprise requires enterprise that " following service ability or following service time " provided and reply immediately through regular meeting; and corporate decision maker itself also needs the management decision problem is done with real-time monitoring, to ensure the smooth implementation of enterprise plan.In a word, many-sided reason will force the decision maker to handle immediately each state that problem changes, at this moment, no matter be how skilled modeling expert, or working method how efficiently, do not change the processing thinking that undue dependence expert deals with problems, computing machine is not transferred in the modeling work of problem and finished, the real-time processing of decision problem just is difficult to realize.
The residing environment of complicated decision-making problems or condition are at every moment all to change, and when this variation surpasses certain scope, have just formed another problem.For the new problem of up-to-date appearance, how to obtain the process that the optimizing decision scheme is gone control problem fast in time, this is the key link of handling the dynamic decision problem.For this reason, the present invention has proposed the natural language understanding that the auxiliary mankind of machine finish problem first, and the problem that machine is understood is carried out intelligent Modeling, finally obtains the exact solution that necessary for human is wanted.The present invention not only will play a significant role in the real-time optimization field of complicated decision-making problems, as real-time optimization scheduling of the optimal control of production run and scheduling, e-commerce distribution logistics etc., but also can be applied to resources such as investment decision and manpower and materials optimization allocation, plan and planning problem, produce storage problem etc., and more wide application prospect is arranged also in CIMS, FMS.
Summary of the invention
Purpose of the present invention is exactly to provide a kind of dynamically adapting ability that problem is changed strong in order to overcome the defective that above-mentioned prior art exists, and the microcomputer modelling of complicated decision-making problems that can real-time optimization with find the solution disposal route.
Purpose of the present invention can be achieved through the following technical solutions:
A kind of microcomputer modelling of complicated decision-making problems with find the solution disposal route, it is characterized in that, may further comprise the steps:
1) the intelligent human-machine interaction intelligence body pre-service of understanding based on problem;
2) intelligent modeling;
3) complex problem solving is handled.
Described intelligent human-machine interaction intelligence body preprocessing process based on problem understanding is as follows:
1) receptor perception problems if judge and to be the problem imported of user then to deposit the interaction problems tabulation in, carry out step 2); If judge it is domain knowledge extraneous or that other intelligent bodies transmit, then be sent to learning machine, carry out step 8);
2) the interaction problems Understanding Module obtains one with the problem statement of natural language expressing from problem list, send morphology and syntactic analysis resume module;
3) morphology and syntactic analysis module are finished the processing that the centre word of participle, part-of-speech tagging and subject and predicate to the problem statement, guest's composition is judged by the dictionary for word segmentation storehouse;
4) the conceptual Center word that obtains by morphology and syntactic analysis of interaction problems Understanding Module, search coupling in the knowledge dictionary, thereby understanding problem;
5) the alternate statement maker is searched for the knowledge concepts word in the knowledge segment tree that obtains in the matching process in the knowledge dictionary with the interaction problems Understanding Module, and the natural statement that the user begins to provide when mutual is the basis, generates the natural language statement of computing machine and user interactions;
6) send the alternate statement of natural language expressing to user by receptor, form man-machine interaction;
7) after man-machine interaction is confirmed, will finally understand the result by receptor and deliver to communicator, and understand result's problem domain of living in through the communicator analysis and judgement, and the corresponding intelligent body of handover is handled;
8) dictionary for word segmentation storehouse, knowledge dictionary are carried out consistance and redundancy inspection, according to check result, upgrade dictionary for word segmentation storehouse, knowledge dictionary.
Described intelligent modeling process is as follows:
1) the model case management system is treated to deal with problems in case library and is carried out historical case search and coupling;
2) judge whether to find the coupling case, if judged result is otherwise execution in step 3), if judged result for being execution in step 5);
3) model management system makes up model by model bank;
4) model content output finishes;
5) the model case management system is filled the case model framework that mates in the case library;
6) judge whether this case model framework is fit to new problem, turns back to step 4) if the judgment is Yes, if judged result is for denying execution in step 7);
7) the case model framework is carried out adaptability revision, return step 4) afterwards.
Described complex problem solving processing procedure is as follows:
1) initialization module of scheme carries out initialization to scheme;
2) detector of scheme is surveyed scheme, judges whether to reach optimization aim according to the end condition of scheme iteration, then finishes if the judgment is Yes, then forwards step 3) if the judgment is No to;
3) maker of scheme generates new route scheme, forwards step 2 afterwards to).
Compared with prior art, the present invention has the following advantages:
1, the dynamically adapting ability that problem is changed is strong;
2, can carry out real-time optimization to complicated decision-making problems.
Description of drawings
Fig. 1 is intelligent human-machine interaction intelligence body structure;
Fig. 2 is that complicated decision-making problems is based on principle and the structural drawing of the intelligent modeling method of textural difference;
Fig. 3 is the execution flow process of complicated decision-making problems derivation algorithm;
Fig. 4 is the solving model tree construction synoptic diagram of complicated decision-making problems.
Embodiment
The present invention is described in detail below in conjunction with the drawings and specific embodiments.
Embodiment
(1) the intelligent human-machine interaction intelligence body preprocessing process implementation method of understanding based on problem
The intelligent human-machine interaction intelligence body of understanding based on problem of the present invention's design comprises: lexical analysis dictionary library management, the management of lexical analysis word sex determination rule base, lexical analysis system, syntactic analysis systems and management problem understanding system.
To the understanding with the problem of management statement of natural language description, the present invention uses morphology and syntactic analysis.Lexical analysis mainly is that sentence is carried out participle, on the basis of participle, sentence structure is carried out syntactic analysis, i.e. the grammatical item born at sentence of analysing word.Then, participle and sentence element analysis all are to carry out on the dictionary basis, and accurately whether whether abundant in content dictionary is, whether rational in infrastructure, determining sentence comprehension.
The present invention is categorized as word by its grammatical function word: noun, verb, adjective, adverbial word, interrogative, pronoun, numeral-classifier compound, preposition, conjunction, auxiliary word, negative word 11 big classes.In natural language, many words all have the characteristics of the many property of a word.Yet in a complete natural language statement, each word all can only have an attribute.Therefore, in order accurately to understand the natural language statement, the present invention carries out part of speech to the word with the many property of a word and judges.
Lexical analysis is the analysis to the word of forming natural language.The sentence of Chinese is composed of words, and between word and the word being does not have to separate.This point is different with English.Therefore, the present invention separates word exactly to the first step that Chinese sentence is handled, and participle just becomes the main contents of sentence lexical analysis.Participle has two kinds of artificial participle and machine automatic word segmentations.There are shortcomings such as participle is inconsistent, processing speed is slow, length consuming time, cost height in artificial participle.At these shortcomings, computer generation replaces artificial participle to become and presses for, and automatic word segmentation technology arises at the historic moment.The present invention in conjunction with the needs that problem of management is understood, has designed three scanning Words partition systems summing up on the existing Words partition system basis.
The implication of sentence is by the sentence pattern structural meaning of sentence and forms the meaning decision of the centre word of sentence element.Therefore, to understand statement be very important to the meaning of correctly holding the sentence pattern structural meaning of sentence and centre word for accurate.The syntactic analysis method of the problem of management statement that the present invention proposes, be on the augmented transition network method basis of grammatical analysis, the sentence pattern structure analysis of sentence is combined with the centre word judgement, and according to centre word and the clear and definite problem domain of management domain knowledge and find the solution target, finally reach the purpose of understanding the problem of management statement.
The present invention is on the basis of above dictionary, lexical analysis system and the design of syntactic analysis system, and simulating human problem understanding process has designed a problem of management and understood intelligence system, and its treatment step is as follows:
Step 1: accept the user with the problem of management statement of natural language input;
Step 2: the participle and the part-of-speech tagging that the problem of management statement are carried out lexical analysis are handled;
Step 3: the question sentence that participle and part-of-speech tagging were handled carries out the subject and predicate of syntactic analysis, the centre word of guest's composition is judged;
Step 4: by the knowledge dictionary, with interactive form problem is carried out alternately;
Step 5: according to the mutual information that obtains, in the knowledge dictionary, search for, namely problem of management is understood;
Step 6: the result is understood in output, allows the user confirm.If the user confirms, then understand and finish to change problem model class coupling subsystem; Otherwise, at the new problem that the user proposes, changeed for 1 step.
(2) challenge intelligent modeling method
The challenge intelligent modeling method based on textural difference that the present invention proposes comprises: the mathematics model analysis method of complicated decision-making problems, the knowledge representation method of complicated decision-making problems mathematical model and the modeling of complicated decision-making problems.
The mathematical model of the present invention's design is comprising variable, objective function and is retraining this three principal ingredients.The variable of complicated decision-making problems mathematical model comprises decision variable and deviation variables two classes.Wherein, decision variable is the unknown quantity that will determine in the problem, and it comes the solution of problem of representation in the mode of quantity.Objective function is the function of model variable, the purpose requirement that its problem of representation will reach.According to the expectation of decision maker's difference, the target of complicated decision-making problems is different, and for example the target of vehicle route scheduling problem can be that distance is the shortest, cost is minimum, use that vehicle number is minimum, the time is minimum etc.Correspondingly, the objective function of this problem mathematical model should give expression to one or more combination of these targets.According to the difference of implication, the constraint of complicated decision-making problems mathematical model mainly contains following a few class: 1) limit relevant constraint with Enterprise Resource.Such constraint often represents that the human and material resources of enterprise, financial resources use can not surpass enterprise's existing resources upper limit etc.2) with the functional relevant constraint of prediction.Such constraint need give expression to a series of implications of business forcast function, as the starting material that estimate to drop into, estimate quantum of output etc.3) constraint relevant with planning function.Such constraint need give expression to a series of implications of enterprise plan function, as enterprise's manufacturing program of a year, recruitment plan etc.4) constraint relevant with controlling function.Such constraint need give expression to a series of implications of enterprise's controlling function, as production implementation status analysis etc.
In the mathematical model of complicated decision-making problems, variable is basicvocabulary, and objective function and equation of constraint then are to use the mathematics statement of this lexical representation conditions of problems implication.The certain structure feature all followed in these statements: each variable has a coefficient in each equation of constraint or objective function, all coefficients can constitute a matrix of coefficients; Each equation of constraint has an equation or inequality symbol, and a right-hand member constant, and they form a model parameter matrix after can being added in matrix of coefficients, and this matrix can be expressed the data message of complicated decision-making problems model fully.Since matrix based on be a bivariate table structure, this structure can be by the mode of entity information in the additional corresponding problem knowledge of raw column data, strengthen this structure for the expression ability of model knowledge, thereby further strengthen computing machine for the understandability of model data, for the identification of model in the modeling process and modification provide the basis.For this reason, the present invention uses tableau format to represent the mathematical model of complicated decision-making problems: gauge outfit is indicated the implication of each row, comprises title, inequality symbol, right-hand member constant and the corresponding entity title of every capable information of all variablees; The first three rows of form records the corresponding relation of entity information in each type of variables, implication and variable and the tree-shaped structure of knowledge of problem respectively; The parameter information of fourth line record cast objective function; Parameter information from the model constrained equation of all line items of back that fifth line begins.
" research and sum up human thinking's universal law, and utilize its realization of computer simulation " is the clear and definite guiding theory that the scientists of artificial intelligence field is followed.Intelligent modeling method proposed by the invention is based upon on the basis of the human modeling law of thought just.The present invention is based on the human modeling law of thought, proposed a kind of intelligent modeling method.This method receives with problem knowledge represents the problem conceptual model described, comes the modeling ability of simulating human by the reasoning of all kinds of knowledge, finally exports the concrete form that the problem mathematical model is represented.
(3) the complex problem solving method has been proposed
The solving model of the complicated decision-making problems among the present invention be one based on the executable program of the problem of finding the solution of knowledge.This program must include the maker of scheme, the detector of scheme, the controller of solution procedure, the initialization of scheme and these five modules of end condition of scheme iteration according to the heuristic feature of finding the solution of complicated decision-making problems; In addition, as the data source of solving model, the information model of complicated decision-making problems and mathematical model be requisite module also.For this reason, the present invention proposes a kind of hexa-atomic group of knowledge representation method of this problem solving model in conjunction with above-mentioned six modules.
The solving model M that defines 1 one complicated decision-making problems can be expressed as one hexa-atomic group:
M＝(B，I，O，P，E，D)
Wherein: B---the set of complicated decision-making problems master data information (Basic information), recording information model and the mathematical model of problem;
I---initial scheme (Initial solution) is for the initial scheme of describing heuristic solution procedure, perhaps the generation knowledge of this scheme;
O---dbjective state (Object state) is for the end condition of describing heuristic solution procedure, perhaps the generation knowledge of this condition;
The reasoning of P---solution procedure and control strategy (Control policy) are the controllers of program operation;
The maker of E---scheme (Solution enumerator) is for the generation knowledge of describing next heuristic scheme;
The detector of D---scheme (Solution detector) be used for to be described the optimization knowledge of solution procedure, uses this tuple to reduce and to enumerate number of times by improving threshold value or reducing mode such as non-feasible branch, optimizes enumeration process.
Hexa-atomic group of M=(B, I, O, P, E D) is referred to as the BIOPED representation of complicated decision-making problems solving model.In six tuples, set B is used for the data knowledge of expression complicated decision-making problems, and its content is the conceptual model of problem and the mathematical model of problem, i.e. two files representing of the representation of knowledge of problem and mathematical model; The IOED tuple then is used for storage and conceals the knowledge of enumerating of finding the solution, and its content will be determined according to the corresponding derivation algorithm of different problem knowledges; The P tuple then is to utilize the executable program of finding the solution knowledge IOED, extracting data, control problem solution procedure from given data source B, also be to make whole solution procedure produce " source " of intelligence, the algorithm of carrying out this tuple realization complicated decision-making problems solution procedure is:
Algorithm 1Step 1: carry out the I tuple, initial scheme is deposited among the variable bestScheme;
Step 2: if bestScheme miss the mark state O then changes next step; Otherwise changeed for the 6th step;
Step 3: carry out the E tuple, generate the next vehicle route scheme of bestScheme, and charge to variable nextSolution;
Step 4: carry out the D tuple, check the feasibility of nextScheme scheme, and solution procedure is optimized, if the nextSolution scheme is feasible and be better than the bestScheme scheme, then cover the bestScheme variable with it;
Step 5: return Step 2, proceed the heuristic iteration work of next round;
Step 6: solution procedure finishes, and bestScheme is best driving scheme.
The BIOPED method for expressing of complicated decision-making problems solving model has been realized separating of knowledge base (IOED) and inference machine (P) on the one hand, when finding the solution dissimilar complicated decision-making problems, can be under the constant situation of reasoning flow process according to each module of problem characteristics displacement knowledge base; Realize data knowledge (B) on the other hand and found the solution separating of knowledge (IOED), all complicated decision-making problems are all followed the knowledge representation method of tree structure, system can not change under the situation of finding the solution the knowledge data interface, processing has the practical problems of different tree structures, the consistance that this has just guaranteed each solver data structure makes each solver can bring into play ability separately under a uniform platform.
According to hexa-atomic group of requirement, each derivation algorithm of complicated decision-making problems all should be decomposed into four knowledge tuples of IOED, these tuples and the mutually integrated executable program that obtains for problem solving of BP tuple.Yet, because the complicacy of complicated decision-making problems and derivation algorithm thereof, each of above-mentioned model representation found the solution knowledge tuple (IOED tuple) and tended to contain a plurality of modules, each module can be subdivided into a plurality of submodules again, so segmentation is gone down, and can be summed up as the solving model of complicated decision-making problems the tree structure of the handstand of a stratification.So, the structure work of solving model has also just developed into the generative process of this tree structure.To this, we adopt the construction method of " brick pattern " to build this tree structure of this problem solving model.Its basic thought is the method according to the building blocks recreation, build a bigger module by each submodule according to certain institutional framework, build a bigger module by some bigger modules according to certain institutional framework, according to said method continue, finally can constitute the solving model tree of a complicated decision-making problems, as shown in Figure 4.
Because different complicated decision-making problems has the different knowledge (IOED tuple) of finding the solution, with different problem and model knowledge (B tuple), thereby different solving model trees is just arranged also; Therefore, the build process of a concrete complicated decision-making problems solving model tree is just to carry out after the coupling of the identification of problem and types of models and problem solving algorithm.
Though the content of the solving model of each complicated decision-making problems tree is not quite similar, their basic structure all should be followed following principle:
1) root node of the tree structure of building according to " brick pattern " method has and has only 6 child nodes, corresponds respectively to 6 tuples of the complicated decision-making problems solving model representation of knowledge;
2) in 6 child nodes of tree structure root node, " B tuple " child node is only built by two submodules, and they are exactly the data file that records complicated decision-making problems information model and mathematical model respectively;
3) the solving model elementary cell of building is " module ", " module " refers to the program file that exists with text formatting without compiling, the derivation algorithm that the meeting basis matched when system moved is organized into a program engineering with the module of correspondence, and compiles automatically and carry out;
4) according to the Miller rule of human cognitive process, the number of the submodule that each module comprises is not above 7.
5) number of each tuple of solving model (except " B tuple ") submodule that can comprise and the degree of depth without limits, the purpose that submodule is set just is: 1. make things convenient for the exploitation of solver; 2. make things convenient for MAINTENANCE OF KNOWLEDGE BASE; 3. make things convenient for building of solving model.This be because: 1. the performance history of a new problem solver is actually the performance history of each module from top to bottom of this problem solving model tree, this from the top to down progressively the thought of refinement be the theoretical foundation of program design in the soft project; 2. when the solver of a new problem need incorporate system, only need weave with the regulated procedure language module file of text formatting, and each module file registered in the registration table of knowledge base get final product; 3. when existing solver in the utilization system is built the solving model tree of a particular problem, only need the record according to the EDISAN registration table, all that find that this problem matches are found the solution module, and are built into a program engineering according to the hierarchical relationship in the registration table with " brick pattern " method and get final product.
Claims (3)
- The microcomputer modelling of a complicated decision-making problems with find the solution disposal route, it is characterized in that, may further comprise the steps:1) the intelligent human-machine interaction intelligence body pre-service of understanding based on problem;2) intelligent modeling;3) complex problem solving is handled;Described complex problem solving processing procedure is as follows:1) initialization module of scheme carries out initialization to scheme;2) detector of scheme is surveyed scheme, judges whether to reach optimization aim according to the end condition of scheme iteration, then finishes if the judgment is Yes, then forwards step 3) if the judgment is No to;3) maker of scheme generates new route scheme, forwards step 2 afterwards to).
- The microcomputer modelling of a kind of complicated decision-making problems according to claim 1 with find the solution disposal route, it is characterized in that the described intelligent human-machine interaction intelligence body preprocessing process of understanding based on problem is as follows:1) receptor perception problems if judge and to be the problem imported of user then to deposit the interaction problems tabulation in, carry out step 2); If judge it is domain knowledge extraneous or that other intelligent bodies transmit, then be sent to learning machine, carry out step 8);2) the interaction problems Understanding Module obtains one with the problem statement of natural language expressing from problem list, send morphology and syntactic analysis resume module;3) morphology and syntactic analysis module are finished the processing that the centre word of participle, part-of-speech tagging and subject and predicate to the problem statement, guest's composition is judged by the dictionary for word segmentation storehouse;4) centre word that obtains by morphology and syntactic analysis of interaction problems Understanding Module, search coupling in the knowledge dictionary, thereby understanding problem;5) the alternate statement maker is searched for the knowledge concepts word in the knowledge segment tree that obtains in the matching process in the knowledge dictionary with the interaction problems Understanding Module, and the natural statement that the user begins to provide when mutual is the basis, generates the natural language statement of computing machine and user interactions;6) send the alternate statement of natural language expressing to user by receptor, form man-machine interaction;7) after man-machine interaction is confirmed, will finally understand the result by receptor and deliver to communicator, and understand result's problem domain of living in through the communicator analysis and judgement, and the corresponding intelligent body of handover is handled;8) dictionary for word segmentation storehouse, knowledge dictionary are carried out consistance and redundancy inspection, according to check result, upgrade dictionary for word segmentation storehouse, knowledge dictionary.
- The microcomputer modelling of a kind of complicated decision-making problems according to claim 1 with find the solution disposal route, it is characterized in that described intelligent modeling process is as follows:1) the model case management system is treated to deal with problems in case library and is carried out historical case search and coupling;2) judge whether to find the coupling case, if judged result is otherwise execution in step 3), if judged result for being execution in step 5);3) model management system makes up model by model bank;4) model content output finishes;5) the model case management system is filled the case model framework that mates in the case library;6) judge whether this case model framework is fit to new problem, turns back to step 4) if the judgment is Yes, if judged result is for denying execution in step 7);7) the case model framework is carried out adaptability revision, return step 4) afterwards.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010193108 CN102270190B (en) | 2010-06-04 | 2010-06-04 | Computer modeling and solving processing method of complex decision-making problem |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010193108 CN102270190B (en) | 2010-06-04 | 2010-06-04 | Computer modeling and solving processing method of complex decision-making problem |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102270190A CN102270190A (en) | 2011-12-07 |
| CN102270190B true CN102270190B (en) | 2013-07-17 |
Family
ID=45052497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010193108 Expired - Fee Related CN102270190B (en) | 2010-06-04 | 2010-06-04 | Computer modeling and solving processing method of complex decision-making problem |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102270190B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111626060B (en) * | 2020-05-21 | 2021-05-11 | 山东师范大学 | Product innovation design method and system based on object field analysis and rule reasoning |
| WO2024007604A1 (en) * | 2022-07-08 | 2024-01-11 | 华为云计算技术有限公司 | Mathematical model solving method and apparatus, and computing device and computing device cluster |
| CN117114387B (en) * | 2023-10-25 | 2024-02-27 | 联通在线信息科技有限公司 | Interactive customer service model building method and device, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1435781A (en) * | 2003-02-24 | 2003-08-13 | 杨炳儒 | Intelligent decision supporting configuration method based on information excavation |
| CN1480884A (en) * | 2003-07-16 | 2004-03-10 | 中南大学 | Swarm intelligence man-machine decision method based on Internet structure |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002067151A1 (en) * | 2001-02-16 | 2002-08-29 | Va Tech Elin Ebg Elektronik Gmbh & Co | Configurator |
-
2010
- 2010-06-04 CN CN 201010193108 patent/CN102270190B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1435781A (en) * | 2003-02-24 | 2003-08-13 | 杨炳儒 | Intelligent decision supporting configuration method based on information excavation |
| CN1480884A (en) * | 2003-07-16 | 2004-03-10 | 中南大学 | Swarm intelligence man-machine decision method based on Internet structure |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102270190A (en) | 2011-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sharp et al. | A survey of the advancing use and development of machine learning in smart manufacturing | |
| Wang et al. | Research on optimization of big data construction engineering quality management based on RNN-LSTM | |
| Zabin et al. | Applications of machine learning to BIM: A systematic literature review | |
| Sahinidis | Optimization under uncertainty: state-of-the-art and opportunities | |
| Hsieh et al. | Enabling the development of base domain ontology through extraction of knowledge from engineering domain handbooks | |
| Muñoz et al. | Towards an ontological infrastructure for chemical batch process management | |
| CN102682122A (en) | Method for constructing semantic data model for material science field based on ontology | |
| CN106127395A (en) | A kind of product unified model constructing system based on semantic meta-model and construction method | |
| CN102243649A (en) | Semi-automatic information extraction processing device of ontology | |
| CN107103363A (en) | A kind of construction method of the software fault expert system based on LDA | |
| CN115293507A (en) | Situation knowledge graph construction method and system for power grid dispatching operation management | |
| Zhong et al. | Automatically generating assembly sequences with an ontology-based approach | |
| CN102270190B (en) | Computer modeling and solving processing method of complex decision-making problem | |
| CN101266628A (en) | Automatic real-time emulation and its paralleling method based on emulated knowledge library | |
| CN106447174A (en) | Ontology modeling method for high-value-added steel product production technology knowledge based on three-dimensional framework | |
| Qiu | Analysis of human interactive accounting management information systems based on artificial intelligence | |
| CN115757810A (en) | Method for constructing standard ontology of knowledge graph | |
| CN101930566B (en) | Hydrological experimental simulation system and method based on parallel system | |
| CN101706840A (en) | Product node tree based presentation method of product performance simulation information | |
| Min et al. | Data mining and economic forecasting in DW-based economical decision support system | |
| Khayut et al. | Intelligent user interface in fuzzy environment | |
| Wang et al. | Customized product configuration rule intelligent extraction and dynamic updating method based on the least recently used dynamic decision tree | |
| Agrawal | Ontologies for Structural Engineering Models Ontologien für Tragwerksmodelle | |
| Fang | The characteristics of artificial intelligence model description language and its research direction | |
| CN107247810B (en) | Method for constructing knowledge base system based on device operation of styrene chemical process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130717 Termination date: 20160604 |