CN116933945A - Project construction period prediction method based on multiple linear regression - Google Patents

Project construction period prediction method based on multiple linear regression Download PDF

Info

Publication number
CN116933945A
CN116933945A CN202310968790.1A CN202310968790A CN116933945A CN 116933945 A CN116933945 A CN 116933945A CN 202310968790 A CN202310968790 A CN 202310968790A CN 116933945 A CN116933945 A CN 116933945A
Authority
CN
China
Prior art keywords
construction period
project
working procedure
construction
linear regression
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
Application number
CN202310968790.1A
Other languages
Chinese (zh)
Inventor
郑建勇
郑茜匀
梅飞
郭梦蕾
高昂
解洋
张玺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN202310968790.1A priority Critical patent/CN116933945A/en
Publication of CN116933945A publication Critical patent/CN116933945A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/27—Regression, e.g. linear or logistic regression
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/08—Construction
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • Data Mining & Analysis (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Evolutionary Computation (AREA)
  • Primary Health Care (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Biomedical Technology (AREA)
  • Public Health (AREA)
  • Operations Research (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Automation & Control Theory (AREA)
  • Water Supply & Treatment (AREA)
  • Fuzzy Systems (AREA)
  • Molecular Biology (AREA)
  • Development Economics (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)

Abstract

The invention discloses a project construction period prediction method based on multiple linear regression, belonging to the field of project construction period prediction of electric power engineering; the project construction period prediction method based on multiple linear regression comprises the following steps: constructing a multiple linear regression model; determining each working procedure task of the project, factors influencing the construction period and rated construction periods of each working procedure; based on the construction of a multiple linear regression model, constructing a working procedure construction period linear prediction model by taking factors influencing the working procedure as independent variables and taking the rated construction period of the working procedure as a constant term; drawing a fuzzy network diagram according to the logic relation of each module of the engineering project based on the linear prediction model of the working procedure construction period, determining a key path of project execution, and obtaining the total construction period prediction range of the project by applying a construction period calculation method of a fuzzy membership function; the prediction method is high in reliability, is convenient for a constructor to understand and make reasonable and practical construction progress plans, and is easier for engineering practice and application.

Description

Project construction period prediction method based on multiple linear regression
Technical Field
The invention belongs to the field of project duration prediction of electric power engineering, and particularly relates to a project duration prediction method based on multiple linear regression.
Background
The electric power capital construction project generally has the following characteristics: long engineering period, large project planning, high technical requirements, multiple cooperation units and the like. The construction project period is effectively predicted, and a reasonable and practical construction progress plan is formulated, so that the method is one of key factors affecting the quality and cost of the project. If the progress of the engineering project can be controlled, the project progress plan can be effectively predicted, the risk resistance of the project can be enhanced, the quality of the project can be improved to a certain extent, and unnecessary cost is reduced; otherwise, the project management is disordered to a certain extent, so that the project can not be completed on schedule, and the economic benefit and the social benefit brought by the project can be influenced. In the decision-making management process, the project total construction period is often required to be predicted or the personnel and machinery are equipped through construction period prediction and adjustment, so that the progress plan is perfected. Therefore, the method can accurately identify and predict the project construction period of the foundation project, and has important significance for project managers and construction units.
At present, the study of the project period prediction of engineering projects by expert scholars at home and abroad mainly adopts the following methods: artificial neural network methods, decision and laboratory methods, struggle value methods, and the like. These methods have theoretically good results but have limitations in practice. The artificial neural network method has the defects of excessive artificial subjective factors, complex network training and undefined internal operation logic; the decision making and laboratory method calculation processes are complex and complicated, and the calculation efficiency is low; and the unordered project progress deviation of the struggle value method on a specific line directly leads to the fact that the project progress performance monitoring result does not accord with the actual result, and the accuracy of project period prediction is low. The method is not suitable for practical engineering managers and constructors because the calculation process is too complicated, the reliability is low and engineering realization is difficult.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a project construction period prediction method based on multiple linear regression, which solves the problems in the prior art.
The aim of the invention can be achieved by the following technical scheme:
a project construction period prediction method based on multiple linear regression comprises the following steps:
constructing a multiple linear regression model;
determining each working procedure task of the project, factors influencing the construction period and rated construction periods of each working procedure;
based on the construction of a multiple linear regression model, constructing a working procedure construction period linear prediction model by taking factors influencing the working procedure as independent variables and taking the rated construction period of the working procedure as a constant term;
based on the working procedure construction period linear prediction model, according to the logic relation of each module of the engineering project, a fuzzy network diagram is drawn, a key path of project execution is determined, and a construction period calculation method of a fuzzy membership function is applied to obtain the total construction period prediction range of the project.
Further, the multiple linear regression model is:
y＝β 0 +β 1 x 1 +β 2 x 2 +…+β m x m +ε
wherein y is a dependent variable, x 1 ，x 2 ，…，x m Is an independent variable, beta 0 ,β 1 ,β 2 ,…,β m As regression coefficients, ε is the random error of the model.
Further, factors affecting the engineering period include: engineering quantity, voltage class, callable personnel, callable equipment and construction conditions.
Further, the rated period of the process is calculated as:
wherein alpha is i For the nominal working period of step i, P ij For the number of manual and mechanical resources required by the j-th group of professional crews in procedure i, N ij The amount of resources planned to be input for the j-th professional work team in the process i, H ij The method is a shift for inputting the work tools and the human resources of the j-th group of professional work teams in the process i.
Further, the working procedure construction period linear prediction model is as follows:
in which Q i 、V i 、P i 、E i 、C i Respectively representing the actual values of the engineering quantity, the voltage level, the callable personnel, the callable equipment and the construction conditions of the ith procedure data; alpha i The rated construction period of the procedure i; beta i Is a partial regression coefficient, T i Epsilon as the predicted period of the ith step i And (3) obeying normal distribution with the mean value of 0 for the random error of the ith procedure.
Further, the step of drawing the fuzzy network map comprises the following steps:
s421, for the time required by each working procedure in the engineering project, estimating the normal duration, the shortest time and the longest duration of the working procedure by a working procedure construction period linear prediction model, and representing the time by a fuzzy membership function;
s422, programming according to a traditional critical path method, and respectively expressing the logic relations among the working procedures by using a table;
s423, drawing an engineering project fuzzy network diagram according to the logic relation among engineering procedures and the corresponding time parameter table.
Further, the fuzzy membership function is:
wherein μ (x) is a fuzzy membership function, x is the elapsed time of the process, t 1 、t 2 、t 3 The shortest, most probable, and longest completion times of the process are respectively indicated.
Further, in the fuzzy network, under the influence of time difference of each process, the construction period of the total process is a time range, each process may be a key process, and the lines formed by the key processes may be key paths;
the calculation formula of the process time difference is as follows:
T TF (i,j)＝T LF (i,j)-T EF (i,j)
wherein T is LF (i,j)、T EF (i, j) is the process latest completion time and the process earliest completion time, respectively, and can be represented by a fuzzy membership function.
An engineering project duration prediction system based on multiple linear regression, comprising:
the regression model building module: constructing a multiple linear regression model;
the construction period factor confirming module: determining each working procedure task of the project, factors influencing the construction period and rated construction periods of each working procedure;
the prediction model building module: based on the construction of a multiple linear regression model, constructing a working procedure construction period linear prediction model by taking factors influencing the working procedure as independent variables and taking the rated construction period of the working procedure as a constant term;
and a total construction period prediction module: based on the working procedure construction period linear prediction model, according to the logic relation of each module of the engineering project, a fuzzy network diagram is drawn, a key path of project execution is determined, and a construction period calculation method of a fuzzy membership function is applied to obtain the total construction period prediction range of the project.
A computer storage medium storing a readable program capable of executing the above-described prediction method when the program is running.
The invention has the beneficial effects that:
the method has the advantages that a clear mathematical relationship between each influence factor of the project construction period and each working procedure construction period can be obtained through researching by introducing a multiple linear regression model, the working procedure construction period can be reasonably corrected and predicted according to the actual value of each working procedure influence factor on the basis of the general working procedure construction period (constant term), the project construction period prediction method is suitable for the prediction situation of the general project construction period, the construction project construction periods of different regions and different scales can be effectively predicted, and a predicted construction time standard can be provided for the project construction period; compared with other methods in the prior art, the mathematical relationship of the model is more visual, the reliability is high, a constructor can understand and make reasonable construction progress plans, the actual construction progress plans are met, and engineering practice and application are easier.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort.
FIG. 1 is a schematic diagram of F test;
FIG. 2 is a diagram of two types of construction;
FIG. 3 is a graph of a triangle blur distribution function;
FIG. 4 is a diagram of engineering project network nodes.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A project construction period prediction method based on multiple linear regression comprises the following steps:
s1, constructing a multiple linear regression model;
the method comprises the following specific steps:
s11, constructing a mathematical model;
the expression form of the multiple linear regression model can be regarded as popularization of the single linear regression model, and the expression is shown as follows:
y＝β 0 +β 1 x 1 +β 2 x 2 +…+β m x m +ε (1)
wherein y is a dependent variable, x 1 ，x 2 ，…，x m Is an independent variable, beta 0 ,β 1 ,β 2 ,…,β m As regression coefficients, epsilon is the random error of the model and the mean value is 0.
Assuming that n observations are made, n sets of data samples can be obtained, the multiple linear regression model can be expressed as:
wherein ε 1 ,ε 2 ,…,ε m Are independent of each other and obey ε to N (0, σ) 2 )。
The above formula can be written in a matrix form as shown in the following formula:
y＝Xβ+ε (3)
in the method, in the process of the invention,
s12, estimating unknown parameters by adopting a least square method;
the parameter estimation of the common multiple linear regression model adopts least square estimation; setting a parameter beta 0 ,β 1 ,β 2 ,…,β m Is estimated as (1)The sum of squares of the deviations of the model from the observed values is:
according to the least square method, the sum of squares Q of the deviations should be minimized, at which time:
for X in formula (3), if present (X T X) -1 Then the least squares estimate of the regression coefficients can be found as:
β＝(X T X) -1 X T Y (6)。
s13, carrying out regression significance test on the multiple linear regression model;
on model inspection, fitting goodness inspection, multiple regression equation overall saliency inspection and multiple regression coefficient saliency inspection are three hypothesis inspection methods of regression models. Among the most commonly used are the saliency tests, also called F-tests, which are able to test the degree of saliency of the linear relationship between all the explained and interpreted variables in the whole regression equation. The F test steps are as follows:
(1) the original hypothesis and alternative hypothesis of the proposed problem;
(2) constructing statistics under the original assumption condition;
(3) calculating the value of the statistic according to the sample information;
(4) comparing the value of the statistic with the value of the theoretical F distribution, and if the value of the calculated statistic exceeds the theoretical value, rejecting the original assumption, otherwise, accepting the original assumption.
Definition F test assume the following formula:
wherein H is 0 For the original assumption, H 1 To select the hypothesis, the F test rejects the original hypothesis under the following conditions: the calculated F-test value is greater than the theoretical F-value found.
The F test calculation process is shown in figure 1; here, assuming a plane in which one of the points is located, a sum of squares of errors (sum of squares of true and estimated value differences) and a sum of squares of regression deviations (sum of squares of estimated value and average value differences) are calculated from the distance relationship between the point and the line, in the figureIs the sample mean value->Is an estimated value.
The total sum of squares of dispersion (SST), sum of Squares of Error (SSE) and Sum of Squares of Regression (SSR) are calculated as follows:
the F statistic is constructed from the above formula:
at a given beta 0 If under the condition of (1)Reject H 0 Assuming that the explanation "linear build" assumption holds; the opposite is not true.
S2, determining each working procedure task of the project, factors influencing the construction period and rated construction periods of each working procedure; the specific process is as follows:
s21, determining a single process task of the project;
in order to realize reasonable construction period calculation of power transmission and transformation engineering, it is first necessary to determine which links and procedures exist in the current project execution process. According to the activity sub-items in the power construction project summary budget quota use guideline (comprising building engineering and electric equipment installation engineering), determining an activity WBS decomposition module of the power transmission and transformation project, wherein the activity WBS decomposition module comprises pile foundation engineering, construction engineering construction, electric equipment installation engineering, secondary protection debugging, measurement and control networking debugging, civil engineering tail sweeping, acceptance checking, quality monitoring and the like.
Representative procedures of the present research project were determined one by means of investigation of a construction site and the like, as shown in table 1. The key path formed by the process flows can cover the construction period of the whole construction project, and plays a key role in drawing a network diagram of the project and calculating the construction period.
TABLE 1 engineering Process form
S22, determining rated completion time of each procedure;
calculating rated working procedures of each link of the project according to a rated working period calculating method, wherein the rated working period comprises the following steps:
s221, determining the total engineering quantity Q of a certain single procedure i of engineering projects i ；
S222, according to parameter quota M of construction project construction operation ij Calculating the number P of manual and mechanical resources required by the j-th professional work team in the process ij The method comprises the following steps:
s223, determining the j-th professional work in the working procedure according to the construction organization design or contractThe resource amount of team plan input is N ij ；
S224, determining that the shift of the j-th group of professional work team work tools and human resources input in the process is H according to the construction organization design ij ；
S225, calculating the rated construction period alpha of the process i The method comprises the following steps:
wherein alpha is i For the nominal working period of step i, P ij For the number of manual and mechanical resources required by the j-th group of professional crews in procedure i, N ij The amount of resources planned to be input for the j-th professional work team in the process i, H ij The method is a shift for inputting the work tools and the human resources of the j-th group of professional work teams in the process i.
S23, analyzing factors influencing the engineering construction period;
and taking each link in project engineering as a research object, and determining main factors influencing the construction period of the project engineering based on historical project data and empirical analysis. Reasonable construction period influencing factors are selected and divided and weighted, so that the accuracy of multi-linear regression model prediction can be effectively improved, and a powerful basis is provided for follow-up project optimization and intelligent early warning. In general, the selection of the impact indicators is based on the principles of comprehensiveness, hierarchy, operability, and differentiation.
Referring to the index system established by numerous scholars in the past for project construction period prediction problems, the main influencing factors for the general project construction period can be known: engineering quantity, site construction conditions, equipment conditions, labor work efficiency, climate conditions, material supply, management level and the like. Through investigation analysis and related quantitative consideration results, the following 5 factors are determined as main indexes for influencing the construction period of the historical engineering project: engineering quantity Q, voltage class V, callable person P, callable equipment E and construction condition C. The values are shown in Table 2:
table 2 engineering project Q, V, P, E, C parameter values
According to engineering project practice and the principles of availability, applicability and the like of parameters, the acquisition mode of the parameters can be obtained as follows:
s231, the engineering quantity Q can be directly inquired from the engineering quantity list approximation, and the unit of each process is different. Through actual field investigation, the representative engineering quantity data with forward corresponding relation with the construction period in each procedure is determined, so that the construction unit can be filled conveniently, and the quota and fitting calculation can be carried out at the later stage conveniently.
S232, the voltage level V takes on the value:
s233, the callable person P is the sum of the actual callable persons in the procedure.
S234, calling device E:
when there are only 1 apparatus for the j procedure:
when there are multiple devices in the j procedure:
in the formula e k,j 、p k,j Respectively representing the available quantity and the available unit quantity of k equipment in the j procedure;
s235, taking 1,2 and 3 grades in the construction condition C; the south is hot in summer and not cold in winter, so that the south area is set to be 1 in spring and autumn, 2 in winter and 3 in summer; the northern summer is not very hot, and the winter is severe cold, so that the northern area is set to be 1 in spring and autumn, 2 in summer and 3 in winter (wherein three, four and five months are spring, six, seven and eight months are summer, nine, ten and ten months are autumn, and twelve, one and two months are winter).
S3, constructing a working procedure construction period linear prediction model based on the multiple linear regression model constructed in the S1, taking factors influencing the engineering construction period as independent variables and taking the rated construction period of the working procedure as a constant term so as to predict the construction period of each working procedure of the engineering;
the method comprises the following specific steps:
s31, an assumption of the construction type of the engineering project is proposed;
construction types of engineering projects can be divided into two categories: linear construction and nonlinear construction, as shown in fig. 2 (a) and (b). The linear construction means that the construction strength is linearly related to the influence factors of the construction progress and is expressed as a uniform increase, a uniform decrease or an equilibrium state; nonlinear building means that the building strength and the progress influencing factors are in a nonlinear state and are in a disordered increasing or decreasing state.
The multiple linear regression model is a statistical method for processing linear dependency relationship between two or more variables, and can correspondingly describe the linear construction process relationship, so that under the condition of linear construction, the multiple linear regression model can be adopted to learn and train each factor, and when regression is remarkable, the model is used for predicting construction period. Otherwise, the engineering is proved to be of a nonlinear construction type, and can be predicted by using models such as a neural network.
S32, establishing a working procedure construction period linear prediction model;
for each single task link executed in the engineering project, taking a factor influencing the engineering construction period as an independent variable, taking the rated construction period of a working procedure as a constant term, and taking the working procedure construction period as the dependent variable, a linear prediction model shown in the following formula can be obtained:
in which Q i 、V i 、P i 、E i 、C i Respectively representing five parameters (engineering quantity Q, voltage class V and capability ofActual values of the caller P, the callable device E and the construction condition C) correspond to table 2; alpha i The value of the constant term is the rated completion time of each procedure of the project obtained by the project construction period calculation model; beta i Is a partial regression coefficient, T i Epsilon as the predicted period of the ith step i And (3) obeying normal distribution with the mean value of 0 for the random error of the ith procedure.
Learning and training are carried out through multiple linear regression model engineering data of the formula (17), and unknown parameter beta is solved i . The parameter estimation here requires the least square method to be used on the premise that the sum of squares of errors is minimized.
S4, drawing a fuzzy network diagram according to the logical relation of each module of the engineering project based on the working procedure construction period linear prediction model constructed in the S3, determining a key path of project execution, and obtaining the total construction period prediction range of the project by applying a construction period calculation method of a fuzzy membership function;
the method comprises the following specific steps:
s41, analyzing the logic relation among the working procedures of the engineering project;
after each process of power transmission and transformation engineering is defined, the logic relation among the processes is determined according to engineering practice management experience and engineering arrangement. In power transmission and transformation engineering, the following aspects are generally considered in determining a logic relationship:
(1) arranged in program execution order; this is determined by the logic of the item itself;
(2) overlap relationship between professional activities; for example, various equipment (e.g., water, electricity, etc.) installations must intersect and overlap with civil construction activities;
(3) rules of the professional engineering system; if the frame foundation is finished, the frame can be installed. This logical relationship is inherent to the engineering system with mandatory dependencies;
(4) requirements of technical specifications; if the requirements of technological intermittence exist among some procedures, the next step can be carried out when the requirements of maintenance period are met for the activities related to electricity, otherwise, the quality cannot be ensured;
(5) constructing relations on organizations; for example, usually, each process is not completed after one activity is completed, and then another activity is completed, and the construction is performed sequentially, or parallel construction is performed, or sectional flow construction is adopted, which is arranged by a construction organization plan;
(6) the automatic dispatch is carried out after the secondary protection debugging, and the latter completes how much of the former is carried out to complete corresponding work;
providing a calculation basis for calculating the total construction period of the engineering project, analyzing the engineering immediately preceding activities of each working procedure, and determining the logic relationship; the logic relation among the working procedures is based on engineering practice.
S42, drawing a fuzzy network diagram, and determining a key path by applying a fuzzy membership function;
firstly, blurring process time parameters of an engineering project, and estimating the operation time and the total construction period of each process according to the characteristics of a power engineering project network progress plan, wherein three-point estimation is generally adopted, namely the shortest operation time, the most probable operation time and the longest operation time; the fuzzy network time distribution is mainly a triangular distribution, and the distribution type is shown in fig. 3.
The membership function is expressed as:
wherein μ (x) is a fuzzy membership function, x is time, and t1, t2 and t3 respectively represent the shortest, the most probable and the longest time of the procedure.
By fuzzy membership function t 1 (i,j)，t 2 (i,j)，t 3 (i,j)]To describe the duration of process i-j; wherein t is 1 (i, j) represents the shortest time of the step i-j; t is t 2 (i, j) represents the most likely time of the process i-j; t is t 3 (i, j) represents the longest time of the process i-j.
The node relationships of the fuzzy network technique are deterministic with invariance. And drawing a fuzzy network node diagram of the project task through analysis of the project Gantt chart and the like based on the logic relation of each link of the project. The fuzzy network diagram drawing method comprises the following steps:
s421, for the time required by each working procedure in the engineering project, estimating the normal duration, the shortest time and the longest duration of the working procedure by a working procedure construction period linear prediction model, and representing the time by a fuzzy membership function;
s422, programming according to a traditional critical path method, and respectively expressing the logic relations among the working procedures by using a table;
s423, drawing an engineering project fuzzy network diagram according to the logic relation among engineering procedures and the corresponding time parameter table.
Calculating the process time difference T TF (i, j) which represents that if the actual end time of process i-j can be at an intermediate value between its earliest completion time and its latest completion time without affecting the total construction period, the calculation of the process time difference can be represented by the following equation:
T TF (i,j)＝T LF (i,j)-T EF (i,j)(19)
wherein T is LF (i,j)、T EF (i, j) is the process latest completion time and the process earliest completion time, respectively, and can be represented by a fuzzy membership function.
Determining a fuzzy critical path according to the project fuzzy network diagram and task duration estimation; in the fuzzy network, the finishing time of the working procedures has uncertainty, the construction period of the total working procedure is a time range under the influence of the time difference of each working procedure, each working procedure can be a key working procedure, and the lines formed by the working procedures can be key paths.
Setting the criticality C of a certain path p in a fuzzy network diagram rp Can be expressed as:
wherein S is the shortest blurring period, S T Is thatIs (are) fuzzy set, < >>For fuzzy triangle count, ++>Is the fuzzy triangle number of the path p.
S43, predicting the total construction period range of the engineering project;
after the fuzzy critical path is defined, each critical task link can be defined, and the total construction period is defined to be composed of the critical tasks of the fuzzy critical path. According to the fuzzy network analysis technology, based on the fuzzy membership function construction period calculation method, the prediction result of the total construction period can be calculated and obtained according to the construction period prediction time of the key task, and the prediction result is a time range.
And obtaining the relation of the engineering procedure in the front and the back according to the network planning diagram, and drawing a network node diagram as shown in fig. 4. Based on which the total project time limit can be calculated.
Examples
Calculating the rated construction period (constant term) of the engineering procedure;
in the embodiment, construction period data of ten groups of outdoor newly-built transformer substation engineering procedures in a certain area are collected, and rated completion periods of each procedure are obtained through calculation in the formula (13) according to the national unified building installation engineering period quota which is queried for research contents of the invention. Since the procedures and the engineering quantities covered by the engineering are different, in order to uniformly study, the rated construction period of each procedure of the project is finally determined through expert experience correction and induction as shown in table 3:
TABLE 3 rated construction period of engineering Process (constant term)
Coefficient calculation of multiple linear regression model
Only the data of the foundation project of the newly-built 500kV transformer substation in the region is collected in the investigation process, so that the regression coefficient of the influence factor of the voltage level V cannot be solved. Here, we consider only the extent to which the four factors of the engineering quantity Q, the callable person P, the callable device E, and the construction condition C affect the construction period. Namely:
T i ＝a i +β 1i Q i +β 2i P i +β 3i E i +β 4i C i (22)
matlab software is selected for programming, and the values of the four influence factor coefficients are obtained through fitting calculation, as shown in Table 4. The checking and accepting and eliminating part is fixed in the time of completion in different projects, the influence factor calculation is not considered, and the rated construction period of the working procedure is added when the total construction period is finally calculated.
Table 4 multiple linear regression model coefficient determination
Through program verification, the p value of all the process F tests is found to be obviously smaller than the significance level (0.05), which indicates that the linearity of the regression model of each process is obviously established.
Total construction period calculation
The critical path of the project is determined according to the construction flow, as shown in fig. 4. And analyzing and calculating the obtained predicted construction period data based on the fuzzy membership function algorithm to obtain the predicted range of the total construction period of the project.
(1) The construction time of each step of the verification project was calculated from the above multiple linear regression model equation, as shown in table 5.
Table 5 validates engineering fill-in data and construction period prediction calculations
(2) The most probable total construction period for the validation project was calculated to be 449 days based on the network node map figure 4.
(3) By adopting the construction period calculation method based on the fuzzy membership function, which is established by the invention, the possibility analysis is carried out on the proposed schedule construction period.
And calculating according to the network node diagram to obtain the total construction period range (360,449,510) of the 500kV newly-built substation construction. Wherein 360 days corresponds to t in the fuzzy membership function 1 Day 449 corresponds to t in the fuzzy membership function 2 510 is t in the function corresponding to fuzzy membership 3 . Thus, the likelihood analysis was performed for the estimated time periods of 390 days (13 months), 420 days (14 months) and 480 days (16 months), respectively.
If the estimated time period is 390 days, using equation (18), the possibility of implementation is:
P(T＝390)＝33.71％ (23)
if the estimated time period is 420 days, the implementation possibilities are:
P(T＝420)＝67.42％ (24)
if the estimated time period is 480 days, the implementation possibilities are:
P(T＝480)＝49.18％ (25)
and the time uncertainty is subjected to fuzzy processing through a fuzzy membership function, so that the feasibility of the project predicted construction period is improved.
In summary, the method is based on historical basic construction project data, takes the main influencing factors of project working procedures as independent variables, takes the working procedures as dependent variables, calculates the rated working procedures as constants, establishes a multiple linear regression model, and predicts the expected completion time of each working procedure of the current engineering project. By using a network planning technology, a network node diagram of the project is drawn by analyzing the project Gantt chart and the like, and a key path of project execution is determined. And finally, obtaining the total construction period prediction range of the project based on the fuzzy membership function construction period calculation method. Compared with the existing construction period prediction method, the project construction period correction model mathematical relationship adopted by the invention is more visual, has high reliability, is convenient for construction parties to understand and make reasonable construction progress plans which accord with actual construction progress, and is easy for practice and application in engineering.
In the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims.

Claims (10)

1. The project construction period prediction method based on multiple linear regression is characterized by comprising the following steps of:
constructing a multiple linear regression model;
determining each working procedure task of the project, factors influencing the construction period and rated construction periods of each working procedure;
based on the construction of a multiple linear regression model, constructing a working procedure construction period linear prediction model by taking factors influencing the working procedure as independent variables and taking the rated construction period of the working procedure as a constant term;
based on the working procedure construction period linear prediction model, according to the logic relation of each module of the engineering project, a fuzzy network diagram is drawn, a key path of project execution is determined, and a construction period calculation method of a fuzzy membership function is applied to obtain the total construction period prediction range of the project.
2. The project duration prediction method based on multiple linear regression according to claim 1, wherein the multiple linear regression model is:
y＝β 0 +β 1 x 1 +β 2 x 2 +…+β m x m +ε
wherein y is a dependent variable, x 1 ，x 2 ，…，x m Is an independent variable, beta 0 ,β 1 ,β 2 ,…,β m As regression coefficients, ε is the random error of the model.
3. The method for predicting construction period of an engineering project based on multiple linear regression according to claim 1, wherein the factors influencing the construction period include: engineering quantity, voltage class, callable personnel, callable equipment and construction conditions.
4. The method for predicting construction period of engineering project based on multiple linear regression according to claim 1, wherein the rated construction period of the process is calculated by the following formula:
wherein alpha is i For the nominal working period of step i, P ij For the number of manual and mechanical resources required by the j-th group of professional crews in procedure i, N ij The amount of resources planned to be input for the j-th professional work team in the process i, H ij The method is a shift for inputting the work tools and the human resources of the j-th group of professional work teams in the process i.
5. The method for predicting the construction period of an engineering project based on multiple linear regression according to claim 2, wherein the linear prediction model of the construction period of the working procedure is:
in which Q i 、V i 、P i 、E i 、C i Respectively representing the actual values of the engineering quantity, the voltage level, the callable personnel, the callable equipment and the construction conditions of the ith procedure data; alpha i The amount of step iSetting a construction period; beta i Is a partial regression coefficient, T i Epsilon as the predicted period of the ith step i And (3) obeying normal distribution with the mean value of 0 for the random error of the ith procedure.
6. The method for predicting construction period of engineering project based on multiple linear regression according to claim 5, wherein the step of drawing the fuzzy network map comprises:
s421, for the time required by each working procedure in the engineering project, estimating the normal duration, the shortest time and the longest duration of the working procedure by a working procedure construction period linear prediction model, and representing the time by a fuzzy membership function;
s422, programming according to a traditional critical path method, and respectively expressing the logic relations among the working procedures by using a table;
s423, drawing an engineering project fuzzy network diagram according to the logic relation among engineering procedures and the corresponding time parameter table.
7. The method for predicting construction period of engineering project based on multiple linear regression according to claim 6, wherein the fuzzy membership function is:
wherein μ (x) is a fuzzy membership function, x is the elapsed time of the process, t 1 、t 2 、t 3 The shortest, most probable, and longest completion times of the process are respectively indicated.
8. The method for predicting construction period of engineering project based on multiple linear regression according to claim 6, wherein in the fuzzy network, under the influence of time difference of each process, the construction period of the total process is a time range, each process may be a critical process, and the lines formed by the critical processes may be critical paths;
the calculation formula of the process time difference is as follows:
T TF (i,j)＝T LF (i,j)-T EF (i,j)
wherein T is LF (i,j)、T EF (i, j) is the process latest completion time and the process earliest completion time, respectively.
9. A multiple linear regression-based project duration prediction system, comprising:
the regression model building module: constructing a multiple linear regression model;
the construction period factor confirming module: determining each working procedure task of the project, factors influencing the construction period and rated construction periods of each working procedure;
the prediction model building module: based on the construction of a multiple linear regression model, constructing a working procedure construction period linear prediction model by taking factors influencing the working procedure as independent variables and taking the rated construction period of the working procedure as a constant term;
and a total construction period prediction module: based on the working procedure construction period linear prediction model, according to the logic relation of each module of the engineering project, a fuzzy network diagram is drawn, a key path of project execution is determined, and a construction period calculation method of a fuzzy membership function is applied to obtain the total construction period prediction range of the project.
10. A computer storage medium storing a readable program, characterized in that the prediction method according to any one of claims 1-8 is executable when the program is run.
CN202310968790.1A 2023-08-03 2023-08-03 Project construction period prediction method based on multiple linear regression Pending CN116933945A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310968790.1A CN116933945A (en) 2023-08-03 2023-08-03 Project construction period prediction method based on multiple linear regression

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310968790.1A CN116933945A (en) 2023-08-03 2023-08-03 Project construction period prediction method based on multiple linear regression

Publications (1)

Publication Number Publication Date
CN116933945A true CN116933945A (en) 2023-10-24

Family

ID=88394043

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310968790.1A Pending CN116933945A (en) 2023-08-03 2023-08-03 Project construction period prediction method based on multiple linear regression

Country Status (1)

Country Link
CN (1) CN116933945A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117391264A (en) * 2023-12-12 2024-01-12 中建三局集团有限公司 Method, system and medium for calculating construction period quota based on BIM model

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117391264A (en) * 2023-12-12 2024-01-12 中建三局集团有限公司 Method, system and medium for calculating construction period quota based on BIM model
CN117391264B (en) * 2023-12-12 2024-03-19 中建三局集团有限公司 Method, system and medium for calculating construction period quota based on BIM model

Similar Documents

Publication Publication Date Title
US10223167B2 (en) Discrete resource management
CN106940833A (en) A kind of power grid enterprises&#39; sale of electricity side methods of risk assessment based on fuzzy number and improved AHP method
CN103310388B (en) Based on the composite index of grid operation method of calculation of information source entropy
Cavone et al. Intermodal terminal planning by petri nets and data envelopment analysis
CN116933945A (en) Project construction period prediction method based on multiple linear regression
CN111199354A (en) Hopfield neural network-based power communication network operation and maintenance work order scheduling method
CN111666706A (en) Maintenance planning optimization method based on particle swarm optimization
Albayrak et al. A state of art review on metaheuristic methods in time-cost trade-off problems
Zohrehvandi et al. A fuzzy project buffer management algorithm: a case study in the construction of a renewable project
Gonzalez-Gonzalez et al. Fuzzy reliability analysis with only censored data
Yuill et al. Application of Adaptive Neuro Fuzzy Inference System (ANFIS) based short term load forecasting in South African power networks
Liu et al. Efficiency of risk management for tunnel security of megaprojects construction in China based on system dynamics
CN116644929A (en) Intelligent management method for operation and maintenance work orders of power distribution network equipment based on Internet of things perception
CN116307481A (en) Distribution area progress arrangement method, system and equipment based on BIM and fuzzy decision
Boussabaine et al. An expert‐simulation system for construction productivity forecasting: Conformation method used to illustrate the uncertainty associated with expert's knowledge
Tai et al. Learn curve for precast component productivity in construction
Song et al. Quality Risk Management Algorithm for Cold Storage Construction Based on Bayesian Networks
Zhang et al. Development of risk assessment model in construction project using fuzzy expert system
Bosenko et al. Development of an automated system to improve the efficiency of the oil pipeline construction management
Jia et al. Preparation and scheduling system of emergency supplies in disasters
CN116011722A (en) Large power grid-oriented man-machine cooperation regulation and control method, module and device
Al‐Tabtabai et al. Expert judgment in forecasting construction project completion
Pan et al. Research on Project Duration Prediction Based on Multiple Linear Regression
Dharssini et al. Benefits of smart meters in institutional building–a case study
Wu et al. Research on Project Duration Prediction Based on Artificial Neural Network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination