CN113807596A - Management method and system for informatization project cost - Google Patents

Management method and system for informatization project cost Download PDF

Info

Publication number
CN113807596A
CN113807596A CN202111112821.0A CN202111112821A CN113807596A CN 113807596 A CN113807596 A CN 113807596A CN 202111112821 A CN202111112821 A CN 202111112821A CN 113807596 A CN113807596 A CN 113807596A
Authority
CN
China
Prior art keywords
data
period
project
cost
future
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.)
Granted
Application number
CN202111112821.0A
Other languages
Chinese (zh)
Other versions
CN113807596B (en
Inventor
胡健坤
陈志坚
吉小恒
王海吉
解文艳
孙浩
卢雪莹
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.)
China Energy Engineering Group Guangdong Electric Power Design Institute Co Ltd
Original Assignee
China Energy Engineering Group Guangdong Electric Power Design Institute Co Ltd
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 China Energy Engineering Group Guangdong Electric Power Design Institute Co Ltd filed Critical China Energy Engineering Group Guangdong Electric Power Design Institute Co Ltd
Priority to CN202111112821.0A priority Critical patent/CN113807596B/en
Publication of CN113807596A publication Critical patent/CN113807596A/en
Application granted granted Critical
Publication of CN113807596B publication Critical patent/CN113807596B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06313—Resource planning in a project environment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0283—Price estimation or determination
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Tourism & Hospitality (AREA)
  • Biophysics (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Primary Health Care (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a management method and a system for informatization project cost, which comprises the following steps: step S1, constructing a construction cost period prediction model based on real data of the construction cost of the historical project in the historical project period, and predicting the prediction data of the construction cost of the existing project in the future project period by using the construction cost period prediction model; step S2, judging whether the planning data of the future engineering period is reasonable or not; and step S3, adjusting the unreasonable planning data of the future engineering period. The invention can judge the rationality of the initial project cost planning, finish the unreasonable re-planning of the project cost, finally realize the stage adjustment of the project cost according to the project progress, improve the accuracy and feasibility of the project cost, and provide the best project cost management for enterprises.

Description

Management method and system for informatization project cost
Technical Field
The invention relates to the technical field of engineering management, in particular to a management method of informatization engineering cost.
Background
The construction cost refers to the construction price of the project, and refers to the total sum of all the expenses expected or actually required for completing the construction of one project. With the continuous advance of the economic construction pace of China, the vigorous development of urban construction is greatly promoted, and in the engineering construction process, because the project construction is a complex implementation process, the general investment is large, the actual construction period is long, meanwhile, the project implementation process needs multi-aspect coordination, and the cost management of multiple engineering projects is increased, so that the accurate engineering cost of the actual project cannot be obtained specifically before the whole project is completed. The uncertain factors of the engineering project need to be analyzed before the bidding of the engineering project, so that the construction cost of the engineering project can be grasped on the whole.
The existing engineering cost management has more uncertain factors, so that the engineering cost is difficult to adjust in stages according to the actual condition of the engineering progress, the difficulty of the engineering cost is greatly increased, the problems of poor accuracy and insufficient feasibility of the engineering cost exist, the optimal engineering cost cannot be provided for enterprises, and the economic benefit of related enterprises is influenced.
Disclosure of Invention
The invention aims to provide an information-based engineering cost management method and system, and aims to solve the technical problems that in the prior art, the engineering cost management has more uncertain factors, so that the engineering cost is difficult to adjust in stages according to the actual condition of the engineering progress, the engineering cost difficulty is greatly increased, the problems of poor accuracy and poor feasibility of the engineering cost exist, the optimal engineering cost cannot be provided for an enterprise, and the economic benefit of the related enterprise is influenced.
In order to solve the technical problems, the invention specifically provides the following technical scheme:
a method for managing information engineering cost includes:
constructing a construction cost period prediction model based on real data of the construction cost of the historical project in the historical project period, and predicting the prediction data of the construction cost of the existing project in the future project period by using the construction cost period prediction model based on the real data of the construction cost of the existing project in the existing project period;
the forecast data of the project cost in the future project period is compared with the planning data of the project cost in the future project period in a difference mode, and whether the planning data of the future project period is reasonable or not is judged according to the difference comparison result;
and adjusting unreasonable planning data of the future engineering period until the difference between the prediction data of the engineering cost in the future engineering period and the planning data of the future engineering period is within a threshold range.
Optionally, the comparing the difference between the predicted data of the construction cost in the future engineering period and the planned data of the construction cost in the future engineering period includes:
the construction cost of the existing engineering is set in the existing engineering period [1, N]True data of { xk|k∈[1,N]Inputting the predicted data out of the project cost in the future project period N +1 into the project cost period prediction modelN+1；
The predicted data out of the project cost in the future project period N +1N+1Planning data y of the project cost in the future project period N +1N+1A difference comparison is made, wherein,
if the difference comparison result exceeds the difference threshold range, the workerPlanning data y of project cost in future project period N +1N+1Unreasonable;
if the difference comparison result does not exceed the difference threshold range, the planning data y of the project cost in the future project period N +1N+1And (4) the method is reasonable.
Optionally, the calculating method of the difference comparison result includes:
calculating the prediction data outN+1With the planning data yN+1As a difference comparison result, the difference comparison result is calculated by the formula:
Figure BDA0003271173720000021
where Δ e is the difference comparison result, T is the transpose operator, WT＝{wr|r∈[1,p]}，WTFor planning data yN+1Of the respective data class, wrFor planning data yN+1P is planning data yN+1Total number of data categories of (1).
Optionally, constructing a cost cycle prediction model based on real data of the project cost of the historical project in the historical project cycle, including:
quantizing the real data of each engineering period in the historical engineering period into a vector form to serve as a training sample, reserving the period number of the engineering period as the time sequence attribute of the corresponding training sample, and sequentially arranging all the training samples according to the time sequence attribute to form a training time sequence sample { x }t|t∈[1,n]}; wherein x istThe real data of the t-th engineering period in the historical engineering period is obtained, the training sample of the t-th time sequence is obtained, and n is the total period number of the engineering period in the historical engineering period;
and applying the training time sequence sample to the CNN-LSTM hybrid neural network for model training to obtain a cost period prediction model.
Optionally, applying the training time sequence sample to the CNN-LSTM hybrid neural network for model training to obtain a cost cycle prediction model, including:
inputting the training time sequence sample into a CNN convolutional neural network for feature extraction, and outputting a feature sequence; wherein the setting of the training parameters of the CNN convolutional neural network comprises: setting 64 network filters, setting an activation function as a Relu function, setting pooling processing as a max-polling mode, and setting dropout probability as 0.25;
performing time sequence prediction training on the characteristic sequence input value LSTM long and short term memory network, and outputting prediction data of a future engineering period;
the training parameter setting of the LSTM long-short term memory network comprises the following steps:
setting the network layer time step as a characteristic type m of the characteristic sequence;
the network Attention coefficient is set as alpha ═ softmax (A)T*sigmoid(O))；
The Attention Value of the network is set to S ═ O alphaT；
The predicted value of the network output is set as out which is sigmoid (B)T*S)；
Wherein, A is { a }, O is { O ═ O }i|i∈[1,m]}，α＝{αi|i∈[1,m]}，S＝{si|i∈[1,m]}，B＝{b}，out＝{outputi|i∈[1,m]}；
softmax is a softmax operation function body, sigmoid is a sigmoid operation function body, O is an output predicted value when an Attention mechanism is not introduced into the LSTM long and short term memory network, and O isiWhen the Attention mechanism is not introduced into the LSTM long and short term memory network, the output value of the ith network layer time step is a preset weight of O, alpha is an Attention coefficient vector formed by the Attention coefficients of all the network layer time steps, and alpha isiIs the Attention coefficient of the i-th network, S is the Attention Value vector formed by the Attention values of all network layer time step, SiThe Value of the Attention Value of the ith network layer time step, b is the preset weight of S, out is the output predicted Value when the Attention mechanism is introduced into the LSTM long-short term memory network, outputiWhen introducing the Attention mechanism for the LSTM long and short term memory network, the output value of the ith network layer time step is transposedAnd (4) sign.
Optionally, the training mode of the LSTM long-short term memory network is set as a reverse transfer mode of seq2 seq;
error settings of the LSTM long and short term memory network
Figure BDA0003271173720000041
Wherein n is the total period number of the engineering period in the historical engineering period, xtTrue data, out, for the project cost in the tth project cycle in the historical project cycletAnd (3) predicted data of the t-th project period of the project cost output when an Attention mechanism is introduced into the LSTM long-short term memory network in the historical project period.
Optionally, adjusting unreasonable planning data of the future engineering period until a difference between the predicted data of the engineering cost in the future engineering period and the planning data of the future engineering period is within a threshold range, including:
planning data yN+1' where the data category to be adjusted is re-planned for the project cost to obtain new planning data yN+1", until the predicted data outN+1With the planning data yN+1"is within a threshold range.
Optionally, adjusting unreasonable planning data of the future engineering period until a difference between the predicted data of the engineering cost in the future engineering period and the planning data of the future engineering period is within a threshold range, including:
the planning data yN+1' and the prediction data outN+1Sequentially performing data weight difference operation on each data type, and performing data weight difference operation on the result { delta e } of each data typer|r∈[1,p]Comparing with the difference threshold range; wherein the content of the first and second substances,
if Δ erIf the difference is within the range of the threshold value, the planning data yN+1' the data category r is the data category needing to be adjusted;
if Δ erIf the difference threshold is not exceeded, the difference is judged to be within the rangePlanning data yN+1' data class r is a data class that does not require adjustment.
Optionally, the planning data, the real data, and the prediction data have the same data type composition form, and the total number of the planning data, the real data, and the prediction data is the same.
The invention also provides an information-based project cost management system, which is used for realizing the information-based project cost management method, and comprises the following steps:
the model establishing unit is used for establishing a construction cost period prediction model and predicting the prediction data of the construction cost of the existing project in the future project period;
the planning rationality judging unit is used for carrying out difference comparison on the predicted data of the project cost in the future project period and the planning data of the project cost in the future project period and judging whether the planning data of the future project period is rational or not according to the difference comparison result;
and the planning adjusting unit is used for adjusting unreasonable planning data of the future engineering period until the difference between the prediction data of the future engineering period and the planning data of the future engineering period is within a threshold range.
Compared with the prior art, the invention has the following beneficial effects:
the construction cost period prediction model realizes that the prediction data of the project cost in the future project period is obtained by utilizing the known real data of the project cost of the existing project in the existing project period, and the prediction data of the project cost of the existing project in the future project period is established on the real cost data of the real progress period of the existing project and is more in line with the real cost data of the existing project in the future project period.
The planning data of the project cost of the prior project in the future project period is compared with the prediction data of the project cost of the prior project period in the future project period, so that the rationality of the initial project cost planning can be judged, the unreasonable re-planning of the project cost can be completed, the project cost can be adjusted in stages according to the project progress, the accuracy and the feasibility of the project cost can be improved, and the optimal project cost management can be provided for enterprises.
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 below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
FIG. 1 is a flow chart of a method for managing information-based construction costs according to an embodiment of the present invention;
fig. 2 is a block diagram of a management system according to an embodiment of the present invention.
The reference numerals in the drawings denote the following, respectively:
1-a model building unit; 2-a planning rationality determination unit; and 3, planning and adjusting the unit.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The project cost is to plan each project cost in advance before the project construction, because the project has a plurality of uncertain factors in the actual construction process, the actual project cost data of the project in the corresponding construction period and the project cost in the corresponding construction period have difference, if the difference is smaller, the accuracy and the feasibility of the plan data of the current project cost can be considered to be high, the project can be constructed according to the plan data of the current project cost, if the difference is larger, the accuracy and the feasibility of the plan data of the current project cost can be considered to be low, the project can not be constructed according to the plan data of the current project cost, otherwise, the large-range deviation of the project construction can be caused, the whole construction precision of the project is influenced, therefore, the embodiment provides a model for establishing the project cost period prediction to establish the actual cost data of the project in the current project period, predicting the data of the project cost of the project in the future project period as the criterion for judging whether the planning data of the project cost of the project in the future project period is accurate and reasonable, and adjusting the planning data to adapt to the real project construction process to realize the accurate planning of the project cost.
As shown in fig. 1, the present invention provides an information-based project cost management method, which comprises the following steps:
step S1, constructing a construction cost period prediction model based on the real data of the construction cost of the historical project in the historical project period, and predicting the prediction data of the construction cost of the existing project in the future project period by using the construction cost period prediction model based on the real data of the construction cost of the existing project in the existing project period;
in step S1, the method for constructing the cost cycle prediction model includes:
step S101, quantizing the real data of each engineering period in the historical engineering period into a vector form to be used as a training sample, reserving the period number of the engineering period as the time sequence attribute of the corresponding training sample, and arranging all the training samples according to the time sequence attribute to form a training time sequence sample, wherein the training time sequence sample is { x }t|t∈[1,n]In which xtThe real data of the t-th engineering period in the historical engineering period is obtained, the training sample of the t-th time sequence is obtained, and n is the total period number of the engineering period in the historical engineering period;
specifically, the data categories of the project cost in the real data of the historical project period include, but are not limited to: building steel cost, building timber cost, building brick material cost, decoration material cost, labor cost, bag clearing cost and the like, and the real data of the engineering cost in each engineering period in the historical engineering period can be quantized into a vector form { C }r|r∈[1,p]}，CrNumber indicating data class r in real dataAccording to the cost of building steel, building wood, building brick, decoration material, manpower and bale cleaning respectively C1，C2，C3，C4，C5，C6And x ist＝{Ctr|r∈[1,p],t∈[1,n]I.e. the real data x of the t-th project cycle in the historical project cycletCost C of construction steel characterized as the t-th project cycle in the historical project cyclet1Cost of building Wood Ct2Cost of building brick Ct3Decoration material cost Ct4Cost of labor Ct5Cost of bag cleaning Ct6Form of vector of construction { Ct1,Ct2,Ct3,Ct4,Ct5,Ct6}。
And S102, applying the training time sequence sample to a CNN-LSTM hybrid neural network for model training to obtain a cost period prediction model.
Step S102, the method for applying the training time sequence sample to the CNN-LSTM hybrid neural network for model training comprises the following steps:
inputting the training time sequence sample into a CNN convolutional neural network for feature extraction, and outputting a feature sequence, wherein the training parameter setting of the CNN convolutional neural network comprises the following steps: setting 64 network filters, setting an activation function as a Relu function, setting pooling processing as a max-polling mode, and setting dropout probability as 0.25;
the historical engineering with the same attributes as the existing engineering is used as a training sample, and because the engineering with compatible attributes has certain construction similarity, a cost period prediction model can be constructed by utilizing the trend characteristic of the engineering cost of the model learning historical engineering and can be directly transferred to the existing engineering for use, so that the future trend of the engineering cost can be predicted according to the real data of the engineering cost of the existing engineering in the existing engineering period.
The CNN convolutional neural network is utilized to extract the data characteristics of the training time sequence sample, and a characteristic sequence with time sequence dependence attribute is output as the input of the LSTM long and short term memory network, so that the LSTM long and short term memory network can conveniently dig out various timesThe associated attribute between the data features in the sequence is mapped to the engineering period and is expressed as the associated attribute between the real data of the excavation engineering cost in each engineering period in the historical engineering period, thereby realizing the input of the real data x of the engineering period t with the engineering cost in the fronttObtaining real data x of the project cost in the post-project period t +1t+1Therefore, the function of predicting the project cost in the future project period is realized.
Performing time sequence prediction training on the characteristic sequence input value LSTM long and short term memory network, and outputting prediction data of a future engineering period, wherein the training parameter setting of the LSTM long and short term memory network comprises the following steps: the network layer time step is set as the feature type m of the feature sequence, and the network Attention coefficient is set as α ═ softmax (A)TSigmoid (O), the Attention Value of the network is set to S ═ O αTThe predicted value of the network output is set as out to sigmoid (B)T*S)，A＝{a}，O＝{oi|i∈[1,m]}，α＝{αi|i∈[1,m]}，S＝{si|i∈[1,m]}，B＝{b}，out＝{outputi|i∈[1,m]The sftmax is a softmax operation function body, the sigmoid is a sigmoid operation function body, O is an output predicted value when an Attention mechanism is not introduced into the LSTM long-short term memory network, and O isiWhen the Attention mechanism is not introduced into the LSTM long and short term memory network, the output value of the ith network layer time step is a preset weight of O, alpha is an Attention coefficient vector formed by the Attention coefficients of all the network layer time steps, and alpha isiIs the Attention coefficient of the i-th network, S is the Attention Value vector formed by the Attention values of all network layer time step, SiThe Value of the Attention Value of the ith network layer time step, b is the preset weight of S, out is the output predicted Value when the Attention mechanism is introduced into the LSTM long-short term memory network, outputiWhen an Attention mechanism is introduced to the LSTM long and short term memory network, the output value of the ith network layer time step is T, and T is a transpose symbol.
An Attention mechanism is introduced into the LSTM long-short term memory network, and the action of important features in an input feature sequence can be amplified, so that the expandability and the accuracy of a cost period prediction model are improved.
The training mode of the LSTM long-short term memory network is set as the reverse transmission mode of seq2seq, and the error of the LSTM long-short term memory network is set as
Figure BDA0003271173720000081
Wherein n is the total period number of the engineering period in the historical engineering period, xtTrue data, out, for the project cost in the tth project cycle in the historical project cycletAnd (3) predicted data of the t-th project period of the project cost output when an Attention mechanism is introduced into the LSTM long-short term memory network in the historical project period.
The input of the construction cost period prediction model is real data of a previous construction period of the construction cost of the historical engineering, the output is prediction data of a later construction period of the construction cost of the historical engineering, the prediction precision of the construction cost period prediction model can be judged by comparing the prediction data of the later construction period of the construction cost of the historical engineering with the real data of the later construction period of the construction cost of the historical engineering, namely the difference between the prediction data of the later construction period of the construction cost of the historical engineering and the real data of the later construction period of the construction cost of the historical engineering is smaller, the higher the prediction precision is, the larger the difference between the prediction data of the later construction period of the construction cost of the historical engineering and the real data of the later construction period of the construction cost of the historical engineering is, the lower the prediction precision is, and a loss function is constructed by the difference between the prediction data of the later construction period of the construction cost of the historical engineering and the real data of the later construction cost of the historical engineering in the construction period, the built construction cost period prediction model has high-precision attributes, is migrated to the existing engineering, inputs real data of the construction cost of the existing engineering in the existing engineering period, and outputs the prediction data of the construction cost of the existing engineering in the future engineering period, so that the prediction data of the construction cost of the existing engineering in the future engineering period is closer to the real data of the construction cost of the existing engineering in the future engineering period, and whether the planning data of the construction cost of the existing engineering in the future engineering period deviates from the real trend of the construction progress of the existing engineering can be judged by comparing the planning data of the construction cost of the existing engineering in the future engineering period with the prediction data of the construction cost of the existing engineering in the future engineering period.
Step S2, comparing the difference between the forecast data of the project cost in the future project period and the planning data of the project cost in the future project period, and judging whether the planning data of the future project period is reasonable or not according to the comparison result;
in step S2, the method for comparing the difference between the predicted data of the construction cost in the future construction period and the planned data of the construction cost in the future construction period includes:
the construction cost of the existing engineering is set in the existing engineering period [1, N]True data of { xk|k∈[1,N]Inputting the predicted data out of the project cost in the future project period N +1 into the cost period prediction modelN+1；
Predicted data out for project cost in future project period N +1N+1Planning data y of future engineering period N +1 with engineering costN+1A difference comparison is made, wherein,
if the difference comparison result exceeds the difference threshold range, the planning data y of the project cost in the future project period N +1N+1Unreasonable;
if the difference comparison result does not exceed the difference threshold range, the planning data y of the project cost in the future project period N +1N+1And (4) the method is reasonable.
The calculation method of the difference comparison result comprises the following steps:
calculating predicted data outN+1And planning data yN+1The Euclidean distance of the user is used as a difference comparison result, and the difference comparison result is calculated by a formula:
Figure BDA0003271173720000101
where Δ e is the difference comparison result, T is the transpose operator, WT＝{wr|r∈[1,p]}，WTFor planning data yN+1Of the respective data class, wrFor planning data yN+1Class of dataImportance weight of r, p is planning data yN+1Total number of data categories of (1).
Because the importance degree of the real data, the planning data and the prediction data forming the engineering cost is different, the more important data types and the lower tolerance degree of the difference indicate that the more important data types are in the prediction data outN+1And planning data yN+1The greater the weight occupied in the difference comparison result, the importance degree of the data category is quantified by setting the importance weight for the data category.
The difference comparison result can realize the low-difference priority discrimination of the data category with high importance degree, namely the planning data yN+1And predicted data outN+1The data classes of high medium importance can be screened out in the presence of low level differences, with high sensitivity for the data classes of high importance, and then the planning data y can be filteredN+1And adjusting to ensure the high-importance data category maintaining accuracy, and finally realizing that the engineering cost maintains the important data planning accuracy in the engineering construction process, thereby maintaining the engineering construction accuracy at a high level.
In step S3, plan data y for unreasonable future engineering cyclesN+1' the method of making the adjustment includes:
planning data yN+1′＝{C(N+1)r′|r∈[1,p]And predicted data outN+1＝{C(N+1)r|r∈[1,p]Sequentially carrying out data weight difference operation on each data type, and carrying out data weight difference operation result { delta e) of each data typer|r∈[1,p]Comparing with a difference threshold range, wherein,
Δer＝wr*(C(N+1)r′-C(N+1)r)，C(N+1)r' data of data class r in planning data of future project period N +1 for project cost of existing project, C(N+1)rAnd the data of the data type r in the planning data of the future engineering period N +1 is the engineering cost of the existing engineering.
If Δ erIf the difference threshold is exceeded, the planning data yN+1' the data category r is the data category needing to be adjusted;
if Δ erIf the difference threshold is not exceeded, the planning data yN+1' data class r is a data class that does not require adjustment.
Step S3, adjusting the unreasonable planning data of the future engineering period until the difference between the prediction data of the engineering cost in the future engineering period and the planning data of the future engineering period is within the threshold range;
the step can realize the progress supervision of the construction cost of the existing engineering, and the planning data of the construction cost in each engineering period can be adjusted in real time according to the actual situation so as to avoid the deviation of the planning and the actual situation.
In step S3, the planning data yN+1' where the data category to be adjusted is re-planned for the project cost to obtain new planning data yN+1", until predicted data outN+1And planning data yN+1"is within a threshold range.
The planning data, the real data and the prediction data have the same data type composition form, and the total number of the planning data, the real data and the prediction data is the same.
Because the predicted data of the project cost of the existing project in the future project period is equivalent to the real data of the project cost of the existing project in the future project period, the plan data of the unreasonable future project period is adjusted until the difference between the predicted data of the future project period and the plan data of the future project period is within the threshold range, and the unreasonable plan data of the future project period can be adjusted to a reasonable state, so that the existing project can be ensured to continue to carry out construction execution according to the plan data, and the plan accuracy of the project cost is ensured.
As shown in fig. 2, based on the above method for managing the cost of the informatization project, the present invention provides a management system, which includes:
the model building unit 1 is used for building a construction cost period prediction model and predicting prediction data of the construction cost of the existing project in the future project period;
the planning rationality judging unit 2 is used for carrying out difference comparison on the predicted data of the project cost in the future project period and the planning data of the project cost in the future project period, and judging whether the planning data of the future project period is rational or not according to the difference comparison result;
and the planning adjusting unit 3 is used for adjusting the planning data of the unreasonable future engineering period until the difference between the prediction data of the future engineering period and the planning data of the future engineering period is within the threshold range.
The invention constructs the construction cost period prediction model to obtain the prediction data of the construction cost in the future engineering period by utilizing the known real data of the construction cost of the prior engineering in the prior engineering period, the prediction data of the construction cost of the prior engineering in the future engineering period is established on the real expense data of the prior engineering in the real progress period, which is more in line with the real expense data of the prior engineering in the future engineering period, the rationality of the initial construction cost planning can be judged by comparing the planning data of the construction cost of the prior engineering in the future engineering period with the prediction data of the construction cost of the prior engineering period in the future engineering period, and the unreasonable re-planning of the construction cost is completed, and the construction cost is adjusted in stages according to the construction progress, so that the accuracy and the feasibility of the construction cost are improved, and the optimal construction cost management can be provided for enterprises.
The above embodiments are only exemplary embodiments of the present application, and are not intended to limit the present application, and the protection scope of the present application is defined by the claims. Various modifications and equivalents may be made by those skilled in the art within the spirit and scope of the present application and such modifications and equivalents should also be considered to be within the scope of the present application.

Claims (10)

1. A method for managing construction costs of an information engineering, comprising:
constructing a construction cost period prediction model based on real data of the construction cost of the historical project in the historical project period, and predicting the prediction data of the construction cost of the existing project in the future project period by using the construction cost period prediction model based on the real data of the construction cost of the existing project in the existing project period;
the forecast data of the project cost in the future project period is compared with the planning data of the project cost in the future project period in a difference mode, and whether the planning data of the future project period is reasonable or not is judged according to the difference comparison result;
and adjusting unreasonable planning data of the future engineering period until the difference between the prediction data of the engineering cost in the future engineering period and the planning data of the future engineering period is within a threshold range.
2. The method as claimed in claim 1, wherein the step of comparing the difference between the predicted data of the construction cost in the future construction period and the planned data of the construction cost in the future construction period and determining whether the planned data of the future construction period is reasonable or not according to the comparison result comprises:
the construction cost of the existing engineering is set in the existing engineering period [1, N]True data of { xk|k∈[1,N]Inputting the predicted data out of the project cost in the future project period N +1 into the project cost period prediction modelN+1；
The predicted data out of the project cost in the future project period N +1N+1Planning data y of the project cost in the future project period N +1N+1A difference comparison is made, wherein,
if the difference comparison result exceeds the difference threshold range, the planning data y of the project cost in the future project period N +1N+1Unreasonable;
if the difference comparison result does not exceed the difference threshold range, the planning data y of the project cost in the future project period N +1N+1And (4) the method is reasonable.
3. The information project cost management method according to claim 2, wherein the difference comparison result calculation method includes:
calculating the prediction data outN+1With the planning data yN+1As a difference comparison result, the difference comparison result is calculated by the formula:
Figure FDA0003271173710000011
where Δ e is the difference comparison result, T is the transpose operator, WT＝{wr|r∈[1,p]}，WTFor planning data yN+1Of the respective data class, wrFor planning data yN+1P is planning data yN+1Total number of data categories of (1).
4. The method of claim 1, wherein the constructing of the construction period prediction model based on the real data of the construction cost of the historical construction in the historical construction period comprises:
quantizing the real data of each engineering period in the historical engineering period into a vector form to serve as a training sample, reserving the period number of the engineering period as the time sequence attribute of the corresponding training sample, and sequentially arranging all the training samples according to the time sequence attribute to form a training time sequence sample { x }t|t∈[1,n]}; wherein x istThe real data of the t-th engineering period in the historical engineering period is obtained, the training sample of the t-th time sequence is obtained, and n is the total period number of the engineering period in the historical engineering period;
and applying the training time sequence sample to the CNN-LSTM hybrid neural network for model training to obtain a cost period prediction model.
5. The method as claimed in claim 4, wherein the applying the training time sequence samples to the CNN-LSTM hybrid neural network for model training to obtain the cost cycle prediction model comprises:
inputting the training time sequence sample into a CNN convolutional neural network for feature extraction, and outputting a feature sequence; wherein the setting of the training parameters of the CNN convolutional neural network comprises: setting 64 network filters, setting an activation function as a Relu function, setting pooling processing as a max-polling mode, and setting dropout probability as 0.25;
performing time sequence prediction training on the characteristic sequence input value LSTM long and short term memory network, and outputting prediction data of a future engineering period;
the training parameter setting of the LSTM long-short term memory network comprises the following steps:
setting the network layer time step as a characteristic type m of the characteristic sequence;
the network Attention coefficient is set as alpha ═ softmax (A)T*sigmoid(O))；
The Attention Value of the network is set to S ═ O alphaT；
The predicted value of the network output is set as out which is sigmoid (B)T*S)；
Wherein, A is { a }, O is { O ═ O }i|i∈[1,m]}，α＝{αi|i∈[1,m]}，S＝{si|i∈[1,m]}，B＝{b}，out＝{outputi|i∈[1,m]}；
softmax is a softmax operation function body, sigmoid is a sigmoid operation function body, O is an output predicted value when an Attention mechanism is not introduced into the LSTM long and short term memory network, and O isiWhen the Attention mechanism is not introduced into the LSTM long and short term memory network, the output value of the ith network layer time step is a preset weight of O, alpha is an Attention coefficient vector formed by the Attention coefficients of all the network layer time steps, and alpha isiIs the Attention coefficient of the i-th network, S is the Attention Value vector formed by the Attention values of all network layer time step, SiThe Value of the Attention Value of the ith network layer time step, b is the preset weight of S, out is the output predicted Value when the Attention mechanism is introduced into the LSTM long-short term memory network, outputiWhen an Attention mechanism is introduced to the LSTM long and short term memory network, the output value of the ith network layer time step is T, and T is a transpose symbol.
6. The information project cost management method according to claim 5,
the training mode of the LSTM long-short term memory network is set as a reverse transmission mode of seq2 seq;
error settings of the LSTM long and short term memory network
Figure FDA0003271173710000031
Wherein n is the total period number of the engineering period in the historical engineering period, xtTrue data, out, for the project cost in the tth project cycle in the historical project cycletAnd (3) predicted data of the t-th project period of the project cost output when an Attention mechanism is introduced into the LSTM long-short term memory network in the historical project period.
7. The method as claimed in claim 6, wherein the step of adjusting the unreasonable planning data of the future engineering period until the difference between the predicted data of the engineering cost in the future engineering period and the planning data of the future engineering period is within a threshold value comprises:
planning data yN+1' where the data category to be adjusted is re-planned for the project cost to obtain new planning data yN+1", until the predicted data outN+1With the planning data yN+1"is within a threshold range.
8. The method of claim 7, wherein adjusting the projected data of the future project period that is unreasonable until the difference between the predicted data of the project cost in the future project period and the projected data of the future project period is within a threshold range comprises:
the planning data yN+1' and the prediction data outN+1Sequentially performing data weight difference operation on each data category, and weighting the data of each data categoryWeighted difference value calculation result [ Delta e ]r|r∈[1,p]Comparing with the difference threshold range; wherein the content of the first and second substances,
if Δ erIf the difference is within the range of the threshold value, the planning data yN+1' the data category r is the data category needing to be adjusted;
if Δ erIf the difference threshold is not exceeded, the planning data yN+1' data class r is a data class that does not require adjustment.
9. The method as claimed in claim 8, wherein the planning data, the real data, and the forecast data have the same data type configuration, and the planning data, the real data, and the forecast data have the same total data type.
10. An information-oriented construction cost management system for realizing the information-oriented construction cost management method according to any one of claims 1 to 9, comprising:
the model building unit (1) is used for building a construction cost period prediction model and predicting prediction data of the construction cost of the existing project in the future project period;
the planning rationality judging unit (2) is used for carrying out difference comparison on the predicted data of the project cost in the future project period and the planning data of the project cost in the future project period, and judging whether the planning data of the future project period is reasonable or not according to the difference comparison result;
and the planning adjusting unit (3) is used for adjusting the unreasonable planning data of the future engineering period until the difference between the prediction data of the future engineering period and the planning data of the future engineering period is within a threshold range.
CN202111112821.0A 2021-09-18 2021-09-18 Management method and system for informatization project cost Active CN113807596B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111112821.0A CN113807596B (en) 2021-09-18 2021-09-18 Management method and system for informatization project cost

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111112821.0A CN113807596B (en) 2021-09-18 2021-09-18 Management method and system for informatization project cost

Publications (2)

Publication Number Publication Date
CN113807596A true CN113807596A (en) 2021-12-17
CN113807596B CN113807596B (en) 2022-12-23

Family

ID=78940252

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111112821.0A Active CN113807596B (en) 2021-09-18 2021-09-18 Management method and system for informatization project cost

Country Status (1)

Country Link
CN (1) CN113807596B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994681A (en) * 2023-03-24 2023-04-21 江苏苏港智能装备产业创新中心有限公司 Engineering cost real-time analysis management method and management system thereof
CN117196544A (en) * 2023-11-07 2023-12-08 恒实建设管理股份有限公司 Intelligent management method and system for engineering information

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105930931A (en) * 2016-04-22 2016-09-07 国网浙江省电力公司经济技术研究院 Electric power engineering cost management method
CN105956788A (en) * 2016-05-24 2016-09-21 国家电网公司 Dynamic management control method for cost of power transmission and transformation project
CN108009674A (en) * 2017-11-27 2018-05-08 上海师范大学 Air PM2.5 concentration prediction methods based on CNN and LSTM fused neural networks
CN110059878A (en) * 2019-04-15 2019-07-26 中国计量大学 Based on CNN LSTM photovoltaic power generation power prediction model and its construction method
CN113011633A (en) * 2021-02-05 2021-06-22 国网湖南省电力有限公司 Overhead transmission line engineering cost prediction method, device and medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105930931A (en) * 2016-04-22 2016-09-07 国网浙江省电力公司经济技术研究院 Electric power engineering cost management method
CN105956788A (en) * 2016-05-24 2016-09-21 国家电网公司 Dynamic management control method for cost of power transmission and transformation project
CN108009674A (en) * 2017-11-27 2018-05-08 上海师范大学 Air PM2.5 concentration prediction methods based on CNN and LSTM fused neural networks
CN110059878A (en) * 2019-04-15 2019-07-26 中国计量大学 Based on CNN LSTM photovoltaic power generation power prediction model and its construction method
CN113011633A (en) * 2021-02-05 2021-06-22 国网湖南省电力有限公司 Overhead transmission line engineering cost prediction method, device and medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
宋帅: "《桥梁系统地震易损性研究》", 30 September 2019, 中国铁道出版社 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994681A (en) * 2023-03-24 2023-04-21 江苏苏港智能装备产业创新中心有限公司 Engineering cost real-time analysis management method and management system thereof
CN117196544A (en) * 2023-11-07 2023-12-08 恒实建设管理股份有限公司 Intelligent management method and system for engineering information
CN117196544B (en) * 2023-11-07 2024-01-30 恒实建设管理股份有限公司 Intelligent management method and system for engineering information

Also Published As

Publication number Publication date
CN113807596B (en) 2022-12-23

Similar Documents

Publication Publication Date Title
Shukla Neuro-genetic prediction of software development effort
CN113807596B (en) Management method and system for informatization project cost
Cheng et al. Artificial intelligence approaches to achieve strategic control over project cash flows
CN112966954B (en) Flood control scheduling scheme optimization method based on time convolution network
CN110751318A (en) IPSO-LSTM-based ultra-short-term power load prediction method
Bala et al. A computer-based cost prediction model for institutional building projects in Nigeria: An artificial neural network approach
CN112947300A (en) Virtual measuring method, system, medium and equipment for processing quality
CN111915073A (en) Short-term prediction method for intercity passenger flow of railway by considering date attribute and weather factor
JPH0636061A (en) Learning system of hierarchic neural network
CN111652425A (en) River water quality prediction method based on rough set and long and short term memory network
CN113449919B (en) Power consumption prediction method and system based on feature and trend perception
CN112308298B (en) Multi-scenario performance index prediction method and system for semiconductor production line
CN111461404A (en) Short-term load and hydropower prediction method based on neural network prediction interval
CN110689162A (en) Bus load prediction method, device and system based on user side classification
CN103778466B (en) Converter steel-making temperature modeling forecasting method based on vector error and system thereof
Ye An algorithm for construction project cost forecast based on particle swarm optimization-guided BP neural network
CN115310782A (en) Power consumer demand response potential evaluation method and device based on neural turing machine
CN110310199B (en) Method and system for constructing loan risk prediction model and loan risk prediction method
Chen Estimating job cycle time in a wafer fabrication factory: A novel and effective approach based on post-classification
CN114757330A (en) Urban instantaneous water consumption prediction method based on LSTM
CN109635008B (en) Equipment fault detection method based on machine learning
CN107491841A (en) Nonlinear optimization method and storage medium
Zhang et al. RSVRs based on feature extraction: a novel method for prediction of construction projects’ costs
CN116151581A (en) Flexible workshop scheduling method and system and electronic equipment
Liu et al. Investment decision support for engineering projects based on risk correlation analysis

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
GR01 Patent grant
GR01 Patent grant