CN103942739A - Method for construction of construction project risk knowledge base - Google Patents
Method for construction of construction project risk knowledge base Download PDFInfo
- Publication number
- CN103942739A CN103942739A CN201410186535.2A CN201410186535A CN103942739A CN 103942739 A CN103942739 A CN 103942739A CN 201410186535 A CN201410186535 A CN 201410186535A CN 103942739 A CN103942739 A CN 103942739A
- Authority
- CN
- China
- Prior art keywords
- construction
- risk
- knowledge
- project
- engineering
- 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.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A method for construction of a construction project risk knowledge base is characterized by comprising the steps of first performing combing, statistics and analysis according to collected project risk events and accident cases, establishing knowledge expressions of environment factors and engineering factors of the project risk events and the accident cases, classifying risk knowledge, and constructing a risk knowledge map; designing and establishing the risk knowledge base of engineering project construction of factors such as comprehensive scheduling, investment, safety, quality and environmental influence, enabling the risk knowledge base to comprise attribute establishment of the environment factors and the engineering factors; and finally updating data of the knowledge base through data collection and handing of projects in construction. According to the method, by utilizing disasters occurring in the engineering projects under construction or having been constructed, performing knowledge mining on the disasters and combining a computer-assisted construction risk knowledge base to analyze risks of projects to be constructed, project risk prediction and early warning on the projects to be constructed are performed, repeated occurrence of safety accidents to the projects to be constructed is avoided, and safety of the construction project is improved.
Description
Technical field
The invention belongs to construction engineering technical field, it is exactly specifically the construction method that designs a kind of construction project risk knowledge storehouse, the engineering risk knowledge of obtaining is classified and medelling processing, foundation can be called by computer system the data structure of analysis, and carry out knowledge processing warehouse-in according to every data in project, for provide comprehensive dynamic knowledge database data to support in the project of building, meet the needs of project in construction to project risk prediction and early warning.
Background technology
Information management is hereby to be proposed by the president Bill of Microsoft lid at the nineties initial stage the earliest, and the information management that he proposes refers to: enterprise yield-power, competitive power, must lean on " information management ".The arriving of kownledge economy, make knowledge become the key factor of Business survival and development, knowledge base system, also referred to as Knowledge Management System, is the platform that information technology support is provided for enterprise implement knowledge management strategy, is raising enterprise sets up core competitiveness necessary system in the era of knowledge-driven economy.First the useful effect of knowledge base is embodied in and makes information and knowledge ordering; Secondly, knowledge base is accelerated flowing of knowledge and information, is conducive to knowledge sharing and exchanges; The 3rd, knowledge base also helps the cooperation and the communication that realize tissue; The 4th, knowledge base can help enterprise to realize the effective management to customer knowledge
Specific to building field, construction project depends on to a great extent and participates in the knowledge that each individuality of project team provides, and says from this angle, and building industry is a knowledge-based industry.The concept of information management is embodied in the tissue of building industry, no matter be the form with working specification, or the form of experience and lessons, or from infotech should be used in the sector.Since nineteen ninety-nine, the research report of publishing in a large number relates to the application of information management in engineering management field, and the focus of its research mainly concentrates on the following aspects: 1, the motivation of Engineering Project Implementation information management; 2, the information management framed structure under entry environment; 3, the effect of IT in information management; 4, the obstacle that information management is implemented in engineering management.Existing research into knowledge management is mainly paid close attention to genetic analysis and the prediction of construction quality security incident, lacks the structure in risk knowledge storehouse to construction project entirety, cannot meet prediction to construction project risk and the needs of early warning.
Summary of the invention
The object of the invention is the technological deficiency of the potential disaster of project in construction being predicted and being analyzed for not setting up effective risk knowledge storehouse in current building trade, the construction method in a kind of construction project risk knowledge storehouse is provided, utilize collect build or built engineering project occur disaster and disaster is carried out to knowledge excavation, analyze the risk by construction project, thereby to carrying out project risk prediction and early warning by construction project.
Technical scheme
In order to realize above-mentioned technical purpose, the present invention designs the construction method in a kind of construction project risk knowledge storehouse, it is characterized in that, it comprises the following steps: first according to Project Risk Event and the accident case collected, carry out combing, statistics, analysis, set up the knowledge expression of its envirment factor and the engineering factor, carry out the attribute description of envirment factor and the engineering factor, and set up risk knowledge catalogue and obtained project risk knowledge is classified and medelling processing, the data structure that foundation can be accepted by various computing machine; Then risk knowledge is classified, build risk knowledge map, comprise structure and the instance analysis of Knowledge Map concept, risk knowledge map; Then design and set up the risk knowledge storehouse of the Engineering Project Construction of the aspect factors such as comprehensive progress, investment, safety, quality, environmental impact, comprise that the attribute of envirment factor, the engineering factor is set up; Finally arrange the data in the storehouse of then refreshing one's knowledge by the Data Collection to project in construction.
In the described first step, set up in the data structure process that can be accepted by various computing machine and need Project Risk Event and event case to carry out the data cleansing based on project in construction building structure, in discovery correction of data file, discernible mistake comprises inspection data consistency, processes invalid value and missing values.
The method of in described second step, risk knowledge being carried out to take in Classification and Identification process is divided into decision analysis in macroscopical field and the concrete analysis of microscopic fields, specifically comprises building on-stream analysis, risk expert investigation enumeration method and Decomposition analysis.
Beneficial effect
The present invention by utilize collect build or built engineering project occur disaster and disaster is carried out to knowledge excavation, analyze the risk by construction project by building risk knowledge storehouse in conjunction with area of computer aided, thereby to carrying out project risk prediction and early warning by construction project, avoid the project of building to repeat to occur security incident, improved the security of construction project.
Brief description of the drawings
Accompanying drawing 1 is process flow diagram of the present invention.
Accompanying drawing 2 is single accident-accident factor Bayesian networks in the embodiment of the present invention.
Accompanying drawing 3 is bayesian algorithm structural drawing of knowledge base unitary construction.
Embodiment
Below in conjunction with drawings and Examples, the present invention will be further described.
As shown in Figure 1, the construction method in a kind of construction project risk knowledge storehouse, it is characterized in that, it comprises the following steps: first according to Project Risk Event and the accident case collected, carry out combing, statistics, analysis, set up the knowledge expression of its envirment factor and the engineering factor, carry out the attribute description of envirment factor and the engineering factor, and set up risk knowledge catalogue and obtained project risk knowledge is classified and medelling processing, the data structure that foundation can be accepted by various computing machine;
Detailed process is as shown in Figure 2, on Computer Database, the result of segmenting is carried out to database above and builds table handling, builds disaster reason table.The representation of knowledge of its envirment factor of Database and the engineering factor, this representation of knowledge is mainly made up of a linear equation, for example causing a reason of this disaster of fire is wire short-circuiting, this data table items of wire short-circuiting becomes the external key of disaster occurrence cause factor table so, and this external key is stored in a concrete data table items as the major key of disaster occurrence cause factor table.Each list item is the reason of envirment factor and the engineering factor except major key, and what this factor was preserved is the raw name of initiation that its factor occurs for this disaster reason, as: wire short-circuiting | electric wire is aging | there is curtain | curtain is inflammable ...Then set up respectively envirment factor table and engineering factor table to carrying out the attribute description of envirment factor and the engineering factor, as: electric wire is aging is to exceed and cause its tenure of use due to the electric wire operating period.Finally set up risk knowledge catalogue and to obtained project risk knowledge by input database table,
The data structure that foundation can be accepted by various computing machine: first Project Risk Event and event case are carried out to the data cleansing based on project in construction building structure, repeat and wrong data in the hope of removing.Data cleansing process is undertaken by computing machine, and in discovery correction of data file, last one program of discernible mistake, comprises inspection data consistency, processes invalid value and missing values etc.And in the process of data cleansing, carry out following twice operation:
A) consistency check
Consistency check, is reasonable value scope and the mutual relationship according to each variable, checks whether data meet the requirements, and finds to exceed normal range, unreasonable or conflicting data in logic.For example, the Main Basis as project in construction risk management with a target date, bullets and entry name.Having the answer of inconsistency in logic may occur in a variety of forms: for example, there is identical entry name, identical Project dates still has different bullets, this should treat as two projects, because in a project, this divides with respect to different building tasks may different bullets.Find when inconsistent, list a target date, bullets and project name etc., be convenient to further check and correct.
B) processing of invalid value and missing values
Due to item development personnel's difference, coding and typing error, in data, may there is some invalid values and missing values, need to give suitable processing.Disposal route of the present invention is had: estimation, whole example is deleted, variable deletion and paired deletion.
The simplest way is exactly to replace invalid value and missing values with the sample average of certain variable, median or mode.This way is simple, but does not take into full account existing information in data, and error may be larger.Another kind of way is exactly the answer to other problems according to respondent, estimates by the correlation analysis between variable or logical deduction.
It is to reject the sample that contains missing values that example is deleted.Because a lot of questionnaires all may exist missing values, the possibility of result of this way causes effective sample volume to greatly reduce, and cannot make full use of the data of having collected.Therefore, be only suitable for key variables disappearance, or the very little situation of sample proportion that contains invalid value or missing values.
If the invalid value of a certain variable and missing values are a lot, and this variable is not particular importance for studied problem, can consider this variable deletion.This way has reduced for the variables number of analyzing, but does not change sample size.Delete in pairs (pairwise deletion) and represent invalid value and missing values by a specific code (normally 9,99,999 etc.), retain whole variablees and the sample of data centralization simultaneously.But, in the time of concrete calculating, only adopt the sample of complete answer, thereby the variable difference that relates to of different point factorials, its effective sample volume also can be different.This is a kind of conservative disposal route, has retained to greatest extent the available information of data centralization.
Adopt different disposal routes to exert an influence to analysis result, especially when obviously relevant between the appearance of missing values nonrandom and variable.Therefore, in investigation, should avoid occurring invalid value and missing values as far as possible, ensure the integrality of data.
Then risk knowledge is classified, build risk knowledge map, comprise structure and the instance analysis of Knowledge Map concept, risk knowledge map; Then design and set up the risk knowledge storehouse of the Engineering Project Construction of the aspect factors such as comprehensive progress, investment, safety, quality, environmental impact, the attribute that comprises envirment factor, the engineering factor is built and first project risk knowledge is carried out to medelling and process and mainly mixing of following several method of identification.Risk Identification Method of the present invention, can be divided into decision analysis in macroscopical field and the concrete analysis of microscopic fields.Reach the possible of the largest ground analysis and utilization data by this several method being mixed use.
1) building on-stream analysis: claim again method of Flow Chart.Building flow process is named again and is built flow process, refers in construction of buildings process, puts into the completion of building from building material, the process using continuously in order by certain equipment.This kind of method emphasized, according to different flow processs, to every one-phase and link, to investigate and analyse one by one, finds out the reason that risk exists.
2) risk expert investigation enumeration method: listed one by one to this building or building the risk that building site may face by risk manager, and classify according to different standards.The related face of expert should be extensive as far as possible, has certain representativeness.General criteria for classification is: directly or indirectly, and nature or artificial, economic factors or non-economic factors etc.
3) Decomposition analysis: refer to the things of a complexity to be decomposed into multiple fairly simple things, large architectural process is decomposed into concrete element, the risk that therefrom analysis may exist and the threat of potential loss.Fault Tree Analysis is to carry out the situation of all error events before survey item loss occurrence with the method for graphic representation, or the various reasons that cause accident are carried out to decomposition analysis, specifically judges which error most probable causes losing risk and occurs.
The identification of risk also has additive method, such as environmental analysis, insurance investigation, crash analysis etc.In the time of this patent identification risk, make alternately in all sorts of ways.
In the time of concrete identification risk, need to fully utilize some technical skill and instrument, to ensure that identifying expeditiously risk does not omit, these methods comprise Delphi method, braistorming, look-up table method, SWOT technology, look-up table and graphic technique etc.
4) the dynamic risk appraisal procedure based on Bayesian network technology is proposed.
1. first according to the kind of event in naive Bayesian method statistic raw data base:
Obtain disaster accident sets of factors and disaster accident set by statistical means
2. obtain certain class disaster accident mapping ruler by algorithm below:
When wherein initial, be illustrated in t culprit of j sample concrete in i class culprit, the complete or collected works of the accident occurrence cause of expression disaster i.Greatly collect by the accident factor that the calculating of this formula is obtained to whole accidents.And greatly collect and build knowledge table according to accident factor in building knowledge table, and using this set content as list item.
3. the accident factor collection of pair each accident is divided, and is divided into natural cause and human factor.
4. construct by formula below the probability that each accident factor occurs, and assert that the generation of each accident factor is relatively independent:
,,…,,,,…,,…,,,…,
5. calculate in the situation that a certain accident occurs the probability that this accident is produced by a certain particular incident factor:
, ...,,, ..., ...,, ...,, t accident factor of wherein expression accident n.
6. enough make its corresponding Bayesian network for each accident i, as shown in Figure 2.
7. merge accident factor according to Fig. 3 and build the Bayesian network of whole accidents.
8. calculate a certain contingency occurrence probability according to formula below:
Wherein Parents represents the associating of the direct precursor node of accident, and probable value can be found in corresponding list item conditional probability from knowledge base.
9. the accident and the factor thereof that constantly occur in the process of building by obtaining actual construction project, constantly update accident factor probability, adjusts Bayesian network, thereby make knowledge base more accurate to the estimation of a certain accident.
Finally arrange the data in the storehouse of then refreshing one's knowledge by the Data Collection to project in construction.
The present invention by utilize collect build or built engineering project occur disaster and disaster is carried out to knowledge excavation, analyze the risk by construction project by building risk knowledge storehouse in conjunction with area of computer aided, thereby to carrying out project risk prediction and early warning by construction project, avoid the project of building to repeat to occur security incident, improved the security of construction project.
Claims (3)
1. the construction method in construction project risk knowledge storehouse, is characterized in that, it comprises the following steps:
(1) according to Project Risk Event and the accident case collected, carry out combing, statistics, analysis, set up the knowledge expression of its envirment factor and the engineering factor, carry out the attribute description of envirment factor and the engineering factor, and set up risk knowledge catalogue and obtained project risk knowledge is classified and medelling processing, the data structure that foundation can be accepted by various computing machine;
(2) risk knowledge is classified, build risk knowledge map, comprise structure and the instance analysis of Knowledge Map concept, risk knowledge map;
(3) design and set up the risk knowledge storehouse of the Engineering Project Construction of the aspect factors such as comprehensive progress, investment, safety, quality, environmental impact, comprise that the attribute of envirment factor, the engineering factor is set up;
(4) Data Collection by project in construction arranges the data in the storehouse of then refreshing one's knowledge.
2. the construction method in a kind of construction project risk knowledge as claimed in claim 1 storehouse, it is characterized in that: in the data structure process that in described step (1), foundation can be accepted by various computing machine, need Project Risk Event and event case to carry out the data cleansing based on project in construction building structure, in discovery correction of data file, discernible mistake comprises inspection data consistency, processes invalid value and missing values.
3. the construction method in a kind of construction project risk knowledge as claimed in claim 1 storehouse, it is characterized in that: the method for in described step (2), risk knowledge being carried out to take in Classification and Identification process is divided into decision analysis in macroscopical field and the concrete analysis of microscopic fields, specifically comprise building on-stream analysis, risk expert investigation enumeration method and Decomposition analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410186535.2A CN103942739A (en) | 2014-05-05 | 2014-05-05 | Method for construction of construction project risk knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410186535.2A CN103942739A (en) | 2014-05-05 | 2014-05-05 | Method for construction of construction project risk knowledge base |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103942739A true CN103942739A (en) | 2014-07-23 |
Family
ID=51190391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410186535.2A Pending CN103942739A (en) | 2014-05-05 | 2014-05-05 | Method for construction of construction project risk knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103942739A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318408A (en) * | 2014-11-25 | 2015-01-28 | 上海建科工程咨询有限公司 | Multi-target and multi-dimension risk management system for large building engineering construction |
| CN107832942A (en) * | 2017-11-02 | 2018-03-23 | 苏州热工研究院有限公司 | A kind of Human Factor in Nuclear Power Plant is because of affair analytical method and instrument |
| CN109242243A (en) * | 2018-07-27 | 2019-01-18 | 南京航空航天大学 | A kind of flight operation risk dynamic analysing method based on improvement random set Bayesian network |
| CN111460033A (en) * | 2020-03-23 | 2020-07-28 | 盛安保险技术股份有限公司 | Quality risk guarantee supply analysis method and system for use stage of building engineering based on engineering quality insurance |
| CN113743690A (en) * | 2020-05-27 | 2021-12-03 | 中国石油化工股份有限公司 | International engineering quotation risk management method |
| CN117689213A (en) * | 2024-01-31 | 2024-03-12 | 四川省华地建设工程有限责任公司 | Mud-rock flow risk assessment method and system based on artificial intelligence |
-
2014
- 2014-05-05 CN CN201410186535.2A patent/CN103942739A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 张静: "基于知识库的建筑工程施工质量控制平台研究", 《中国优秀博硕士学位论文全文数据库(硕士)经济与管理科学辑（月刊）》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318408A (en) * | 2014-11-25 | 2015-01-28 | 上海建科工程咨询有限公司 | Multi-target and multi-dimension risk management system for large building engineering construction |
| CN107832942A (en) * | 2017-11-02 | 2018-03-23 | 苏州热工研究院有限公司 | A kind of Human Factor in Nuclear Power Plant is because of affair analytical method and instrument |
| CN109242243A (en) * | 2018-07-27 | 2019-01-18 | 南京航空航天大学 | A kind of flight operation risk dynamic analysing method based on improvement random set Bayesian network |
| CN109242243B (en) * | 2018-07-27 | 2022-08-05 | 南京航空航天大学 | Flight operation risk dynamic analysis method based on improved random set Bayesian network |
| CN111460033A (en) * | 2020-03-23 | 2020-07-28 | 盛安保险技术股份有限公司 | Quality risk guarantee supply analysis method and system for use stage of building engineering based on engineering quality insurance |
| CN113743690A (en) * | 2020-05-27 | 2021-12-03 | 中国石油化工股份有限公司 | International engineering quotation risk management method |
| CN117689213A (en) * | 2024-01-31 | 2024-03-12 | 四川省华地建设工程有限责任公司 | Mud-rock flow risk assessment method and system based on artificial intelligence |
| CN117689213B (en) * | 2024-01-31 | 2024-04-05 | 四川省华地建设工程有限责任公司 | Mud-rock flow risk assessment method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110119413B (en) | Data fusion method and device | |
| Poh et al. | Safety leading indicators for construction sites: A machine learning approach | |
| Peng et al. | A hybrid data mining approach on BIM-based building operation and maintenance | |
| CN103942739A (en) | Method for construction of construction project risk knowledge base | |
| CN103226743B (en) | Aircraft equipment technology maturity based on TRL assesses information processing method | |
| CN105574593B (en) | Track state static detection and control system and method based on cloud computing and big data | |
| CN102117443A (en) | Analyzing anticipated value and effort in using cloud computing to process a specified workload | |
| CN104281525B (en) | A kind of defect data analysis method and the method utilizing its reduction Software Testing Project | |
| Assaad et al. | Determining critical combinations of safety fatality causes using spectral clustering and computational data mining algorithms | |
| CN109447658A (en) | The generation of anti-fraud model and application method, device, equipment and storage medium | |
| Mantha et al. | Cybersecurity in construction: Where do we stand and how do we get better prepared | |
| CN105631612A (en) | System and method of evaluating individual performance and capability of public servant based on big data | |
| Jiang et al. | Improvement of random forest by multiple imputation applied to tower crane accident prediction with missing data | |
| Senthilkumar et al. | Case study–based testing of design interface management system | |
| Sriliasta et al. | Overview of life cycle assessment of current emerging technologies | |
| Franch et al. | Community data for OSS adoption risk management | |
| Vilochani et al. | Consolidation of management practices for Sustainable Product Development: A systematic literature review | |
| Al-Roumi et al. | Exploring the rate of adoption and implementation depth of building information modeling (BIM): A case of Kuwait | |
| CN106056300A (en) | Crisscrossed mixed resource allocation rule mining method | |
| CN116703148B (en) | Cloud computing-based mine enterprise risk portrait method | |
| Amies et al. | Success factors for dam engineering industry: systematic literature review and conceptual classification | |
| Lee et al. | Software fault-proneness analysis based on composite developer-module networks | |
| KR101985961B1 (en) | Similarity Quantification System of National Research and Development Program and Searching Cooperative Program using same | |
| CN100517225C (en) | Method for automatically digging high-performance task in software course task warehouse and system thereof | |
| CN113642669A (en) | Fraud prevention detection method, device and equipment based on feature analysis and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140723 |
|
| WD01 | Invention patent application deemed withdrawn after publication |