A cloud-based dynamic cost control system for engineering projects

Abstract

This invention discloses a cloud computing-based dynamic cost management system for engineering projects, relating to the field of data analysis technology. The system includes: determining the upper limit of available funds, the project completion deadline, and the project's application requirements; dividing the project's construction progress according to task type and dividing the completion deadline based on task type to obtain the construction time limit for each task type; using the construction time limit for each task type as the basic constraint and the upper limit of available funds as the variable constraint, generating construction methods for each task type; determining whether the construction methods for each task type meet the project's application requirements; if not, re-selecting; if yes, executing normally. The advantages of this invention are: reduced manual intervention costs, shortened construction period, and improved capital utilization management.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a cloud computing-based dynamic cost control system for engineering projects. Background Technology

[0002] Dynamic cost control in engineering projects refers to a methodology based on real-time data collection, intelligent analysis, and dynamic optimization technologies to achieve refined and adaptive management of costs, schedules, and resources throughout the entire project lifecycle. Its core is to use digital means to monitor cost changes, resource consumption, and schedule deviations during construction in real time, and to dynamically adjust construction plans using optimization algorithms to ensure efficient project progress within budget constraints.

[0003] Existing engineering cost control schemes generally suffer from insufficient fund management capabilities, mainly manifested in the following ways: relying on static budgets and manual experience, making it difficult to track fund flows in real time; lacking a dynamic response mechanism to market fluctuations, resulting in a high risk of budget overruns; and a disconnect between fund allocation and construction progress, making it impossible to dynamically optimize resource input based on actual project progress, leading to idle or insufficient funds, which ultimately affects project cost-effectiveness and schedule control. Summary of the Invention

[0004] To address the aforementioned technical problems, a cloud computing-based dynamic cost management system for engineering projects is provided. This technical solution resolves the problems described above.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A cloud computing-based method for dynamic management and control of engineering costs includes:

[0007] Determine the upper limit of available funds for the project, the project completion timeframe, and the project application requirements;

[0008] Based on the construction progress of the project, the completion time limit is divided according to the task type of the construction progress to obtain the construction time limit of each task type of the project.

[0009] Based on the construction time limit of each task type of the project as the basic constraint, and the upper limit of the project's available funds as the variable constraint, the construction method of each task type of the project is generated.

[0010] Determine whether the construction methods for each task type in the project meet the application requirements of the project. If not, determine whether to re-select; if yes, determine whether to execute normally.

[0011] Furthermore, the construction progress of the project information is divided according to task type to obtain the construction progress of several task types of the project.

[0012] Verify the reachability between the construction processes of several task types in the project, and construct a tree-like hierarchical task path dependency matrix for the project;

[0013] Obtain construction time limit parameters for several task types from historical engineering project information, and establish a reference database of construction time limits for task types of historical engineering projects.

[0014] By marking the task type and construction time limit of historical engineering projects and referring to several tasks of the same type as the engineering project in the database, the construction progress and duration samples of several task types of the engineering project are obtained.

[0015] Using construction progress and duration samples of several task types in an engineering project as reference objects for scatter plot estimation, the expected duration vectors of the construction progress of several task types in the engineering project are extracted.

[0016] Furthermore, based on the tree-like hierarchical task path dependency matrix of the project, and using the expected duration vectors of the construction process of several task types in the project, the forward and reverse execution duration paths of the construction process of several task types in the project are verified.

[0017] By utilizing the forward and reverse execution schedule paths of the construction processes for several task types in an engineering project, the allowable floating time for the construction processes of several task types in the engineering project is calculated, and the execution paths for the construction processes of several task types in the engineering project are determined.

[0018] Furthermore, based on the expected duration vector of the construction process of several task types in the project and the allowable floating time of the construction process of several task types in the project, the spatial distance of the construction process of each task type in the tree-like hierarchical task path dependency matrix of the project is calculated.

[0019] Using the k-means clustering algorithm, tasks are partitioned to minimize the spatial distance of construction processes for each task type in the tree-like hierarchical task path dependency matrix of the project, thus obtaining a set of construction processes for parallel and non-parallel task types in the project.

[0020] Based on the set of construction processes for parallel and non-parallel task types in an engineering project, and according to the expected duration vector of the construction process for several task types in the engineering project, the forward execution duration path and the reverse execution duration path of the construction process, the allowed floating time of the construction process, and the execution path of the construction process, a construction time limit vector is assigned to each task type of the engineering project.

[0021] Furthermore, based on several known construction methods for engineering projects, the historical cost change time series parameters of construction resources for each construction method are obtained;

[0022] The time series parameters of historical cost changes in construction resources for each construction method are stabilized.

[0023] Based on AR time series autoregression, a construction cost prediction model for engineering building construction methods is established.

[0024] The historical cost change time series parameters of construction resources for each construction method are substituted into the construction cost prediction model of the construction method of the engineering building. The historical cost change time series parameters of construction resources for each construction method are used as the input of the observed value, and the predicted real-time construction cost of each construction method in the future is used as the output of the predicted value.

[0025] Furthermore, the construction time requirements and real-time construction cost requirements for each construction method are determined, and a database of time-cost requirements for known construction methods is established.

[0026] The objective function is established to minimize the total construction period of the project;

[0027] The constraints are established by using the construction time limits of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint.

[0028] Based on mixed-integer linear programming, using a known construction method time-cost database as input, and under constraints, select several known construction methods that satisfy the objective function to form an initial set of known construction methods for each task type of the selected project.

[0029] Furthermore, based on the decision tree, the construction process of each task type of the project is taken as the root node; the known construction methods of each task type of the project are selected as branch nodes according to the initial screening of the root node of the construction process of each task type; and the information gain of each branch node for minimizing the total construction period of the project is used as the branch decision to obtain the construction methods of each task type of the project.

[0030] Furthermore, based on the construction methods for each task type of the project, the performance indicators of each construction method are quantified, and a scatter plot of the performance of the construction methods for each task type of the project is established.

[0031] Based on the application requirements of the engineering project, mark the construction method performance requirements of each task type of the engineering project;

[0032] Based on the cosine similarity formula, calculate the matching degree between the performance requirements of the construction methods of each task type in the project at each scatter point in the performance scatter plot of the construction methods of each task type in the project.

[0033] The system determines whether the construction methods of each task type in the project meet the application requirements of the project based on the matching degree between the construction methods and performance requirements of each task type in the project corresponding to each scatter point. If not, it determines whether to re-filter; if yes, it determines whether to execute normally.

[0034] Furthermore, a cloud-based dynamic cost management system for engineering projects includes:

[0035] Initial module, construction time limit allocation module, construction method screening module, construction decision module;

[0036] The initial module is used to determine the upper limit of available funds for the project, the project completion time limit, and the project application requirements.

[0037] The construction time limit allocation module is electrically connected to the initial module. The construction time limit allocation module is used to divide the completion time limit according to the task type based on the construction progress of the project, so as to obtain the construction time limit of each task type of the project.

[0038] The construction method screening module is electrically connected to the initial module and the construction time limit allocation module. The construction method screening module is used to generate construction methods for each task type of the project based on the construction time limit of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint.

[0039] The construction decision module is electrically connected to the initial module and the construction method screening module. The construction decision module is used to determine whether the construction method of each task type of the project meets the application requirements of the project. If not, it determines to re-screen; if so, it determines to execute normally.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] This invention proposes a cloud computing-based dynamic cost management scheme for engineering projects. By dynamically integrating funding constraints, schedule requirements, and application needs, it achieves real-time optimization and control of engineering costs. Its core principle lies in leveraging the distributed computing capabilities of cloud computing to rapidly decompose the total project duration into individual task nodes. Combining historical cost time-series forecasts and real-time resource price data, it generates the optimal combination of construction methods under funding constraints through mixed integer programming, and verifies the compliance with requirements using a decision tree algorithm. Its beneficial effects include: 1. Improving the efficiency of multi-task parallel optimization; 2. Dynamically integrating market resource price fluctuations to enhance the accuracy of cost prediction; 3. Reducing manual intervention costs through an automated decision-making chain, thereby shortening the construction period and improving capital utilization. Attached Figure Description

[0042] Figure 1A flowchart of a cloud computing-based dynamic cost control method for engineering projects;

[0043] Figure 2 This is a framework diagram of a cloud-based dynamic cost control system for engineering projects. Detailed Implementation

[0044] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0045] Reference Figure 1 As shown, a cloud computing-based dynamic cost management method for engineering projects includes:

[0046] Step 1: Determine the upper limit of available funds for the project, the project completion timeframe, and the application requirements of the project.

[0047] Step 2: Based on the construction progress of the project, divide the completion time into tasks according to the task type to obtain the construction time limit for each task type of the project.

[0048] Step two includes the following:

[0049] Step 201: Based on the construction progress of the project information, divide it according to task type to obtain the construction progress of several task types of the project.

[0050] Verify the reachability between the construction processes of several task types in the project, and construct a tree-like hierarchical task path dependency matrix for the project;

[0051] Obtain construction time limit parameters for several task types from historical engineering project information, and establish a reference database of construction time limits for task types of historical engineering projects.

[0052] By marking the task type and construction time limit of historical engineering projects and referring to several tasks of the same type as the engineering project in the database, the construction progress and duration samples of several task types of the engineering project are obtained.

[0053] Using construction progress and duration samples of several task types within an engineering project as a reference for scatter plot estimation, the expected duration vectors of the construction progress for several task types within the engineering project are extracted as follows:

[0054] ,

[0055] in, Let be the expected duration vector of the construction process for the i-th task type in the engineering project. The shortest construction period for the i-th task type of the project. Let be the possible duration of the construction process for the i-th task type of the project. Let be the longest construction period for the i-th task type of the project. For calibration coefficients, This is a sample of the construction progress and duration for several task types within an engineering project.

[0056] Step 202: Based on the tree-like hierarchical task path dependency matrix of the project, and using the expected duration vectors of the construction process of several task types in the project, verify the forward and reverse execution duration paths of the construction process of several task types in the project.

[0057] By utilizing the forward and reverse execution schedule paths of the construction processes for several task types within an engineering project, the allowable floating time for the construction processes of these task types is calculated, and the execution paths for the construction processes of these task types are determined, as follows:

[0058] ,

[0059] in, This indicates that task type i is a prerequisite task for task type j. This represents the earliest start time of the subsequent construction process for the j-th task type in the project. This refers to the latest completion time of the preliminary construction process for the i-th task type in the project. This represents the latest start time of the subsequent construction process for the j-th task type in the project. Allow floating time for the pre-construction process of the i-th task type in the project.

[0060] Step 203: Based on the expected duration vector of the construction process of several task types in the project and the allowable floating time of the construction process of several task types in the project, calculate the spatial distance of the construction process of each task type in the tree-like hierarchical task path dependency matrix of the project.

[0061] The k-means clustering algorithm is used to partition tasks by minimizing the spatial distance of construction processes for each task type in the hierarchical task path dependency matrix of the engineering project, thereby obtaining a set of construction processes for parallel and non-parallel task types in the engineering project, as follows:

[0062] ,

[0063] in, This refers to a set of construction processes for engineering projects, categorized as either parallel or non-parallel tasks. For the k-th parallel-non-parallel task type cluster, This represents the spatial distance between the preceding and succeeding processes of the i-th task type and the succeeding process of the j-th task type in the hierarchical task path dependency matrix of the engineering project. Let be the expected duration vector of the subsequent construction process for the j-th task type in the engineering project. For indicator functions (if) Right now yes (Construction progress) For floating time weighting coefficients, Allow floating time for the subsequent construction process of the j-th task type in the project. For dependency penalty conditions (if) (This forces i and j to not be assigned to the same cluster).

[0064] Based on the set of construction processes for parallel and non-parallel task types in an engineering project, and according to the expected duration vectors, forward and reverse execution duration paths, allowed float times, and execution paths of the construction processes for several task types within the engineering project, a construction time limit vector is assigned to each task type, as follows:

[0065] ,

[0066] in, Let be the construction time limit vector for the i-th task type of the engineering project;

[0067] When using it, refer to the content in sections 201 to 203:

[0068] As a further step, the expected project duration is calculated by decomposing structured tasks and modeling dependencies, combined with the historical data-driven PERT three-point estimation method, and the task execution path and floating time are verified by using critical path analysis (CPM). On this basis, by integrating the spatial distance metric of project duration, resources and dependency constraints and the k-means clustering algorithm, the parallel and non-parallel task sets are intelligently divided, and finally a reasonable construction time limit vector is generated.

[0069] Beneficial effects include:

[0070] 1. Historical data-driven project duration forecasting significantly improves accuracy;

[0071] 2. Coupling analysis of dependency matrix and CPM ensures the reliability of the critical path;

[0072] 3. Clustering algorithms are used to optimize the grouping of construction tasks. While ensuring the logical flow of the work processes, parallel scheduling can shorten the total construction period and improve resource utilization.

[0073] Step 3: Based on the construction time limit of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint, generate the construction method for each task type of the project.

[0074] Step three includes the following:

[0075] Step 301: Based on several known construction methods for engineering buildings, obtain the time series parameters of historical cost changes of construction resources for each construction method;

[0076] The time series parameters of historical cost changes in construction resources for each construction method are stabilized.

[0077] Based on AR time series autoregression, a construction cost prediction model for engineering building construction methods is established.

[0078] The historical cost change time series parameters of construction resources for each construction method are substituted into the construction cost prediction model of the construction method of the engineering building. The historical cost change time series parameters of construction resources for each construction method are used as the input of the observed value, and the predicted real-time construction cost of each construction method in the future is used as the output of the predicted value.

[0079] Step 302: Determine the construction time requirement and real-time construction cost requirement for each construction method, and establish a database of time-cost requirements for known construction methods.

[0080] The objective function is established to minimize the total construction period of the project;

[0081] The constraints are established by using the construction time limits of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint.

[0082] Based on mixed-integer linear programming, using a known construction method time-cost database as input, and under constraints, several known construction methods that satisfy the objective function are selected to form an initial set of known construction methods for each task type of the selected project, as follows:

[0083] ,

[0084] in, Let v be the construction period for the i-th task type and v-th construction method of the project. The v-th construction method selected for the i-th task type of the project. Let V be the construction cost of the v-th construction method for the i-th task type in the engineering project. The total construction period of the project. Given the time-cost requirements database for known construction methods, consider the v-th construction method for the i-th task type. The total number of task types. To satisfy the total number of construction methods for the i-th task type;

[0085] Step 303: Based on the decision tree, take the construction process of each task type of the project as the root node; use the known construction methods of each task type of the project as the branch node to satisfy the initial screening of the construction process root node of each task type; and use the information gain of each branch node for minimizing the total construction period of the project as the branch decision to obtain the construction methods of each task type of the project.

[0086] When using this, refer to the content in sections 301 to 303:

[0087] As a further step, by integrating time-series analysis of historical cost data of construction resources and AR prediction models, a time-cost database of construction methods is established. Combined with mixed integer linear programming, an initial set of construction methods is selected under the constraints of funds and time. Then, based on the decision tree algorithm, the combination of construction methods is optimized according to the information gain criterion, so as to achieve the minimization of project duration and efficient allocation of funds.

[0088] The beneficial effects are:

[0089] 1. Dynamic cost modeling: Real-time prediction of construction resource costs based on AR time series, and construction of time-cost database to ensure the timeliness of cost data;

[0090] 2. Mixed Integer Programming (MILP): With construction time and budget constraints as constraints, and minimizing the total construction period as the objective, it precisely selects the set of feasible construction methods.

[0091] 3. Decision tree optimization: By quantifying the impact of each construction method on the total project duration through information gain, the optimal branch is selected, which significantly reduces the total project duration.

[0092] Step 4: Determine whether the construction methods for each task type of the project meet the application requirements of the project. If not, determine to re-screen; if yes, determine to execute normally.

[0093] Step four includes the following:

[0094] Step 401: Based on the construction methods of each task type in the project, quantify the performance indicators of each construction method and establish a scatter plot of the performance of the construction methods of each task type in the project.

[0095] Based on the application requirements of the engineering project, mark the construction method performance requirements of each task type of the engineering project;

[0096] Based on the cosine similarity formula, calculate the matching degree between the performance requirements of the construction methods of each task type in the project at each scatter point in the performance scatter plot of the construction methods of each task type in the project.

[0097] The system determines whether the construction methods of each task type in the project meet the application requirements of the project based on the matching degree between the construction methods and performance requirements of each task type in the project corresponding to each scatter point. If not, it determines whether to re-filter; if yes, it determines whether to execute normally.

[0098] Reference Figure 2 As shown, a cloud-based dynamic cost control system for engineering projects includes:

[0099] Initial module, construction time limit allocation module, construction method screening module, construction decision module;

[0100] The initial module is used to determine the upper limit of available funds for the project, the project completion time limit, and the project application requirements.

[0101] The construction time limit allocation module is electrically connected to the initial module. The construction time limit allocation module is used to divide the completion time limit according to the task type based on the construction progress of the project, so as to obtain the construction time limit of each task type of the project.

[0102] The construction method screening module is electrically connected to the initial module and the construction time limit allocation module. The construction method screening module is used to generate construction methods for each task type of the project based on the construction time limit of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint.

[0103] The construction decision module is electrically connected to the initial module and the construction method screening module. The construction decision module is used to determine whether the construction method of each task type of the project meets the application requirements of the project. If not, it determines to re-screen; if so, it determines to execute normally.

[0104] In summary, the advantages of this invention are: real-time optimization and control of project costs through dynamic integration of funding constraints, schedule requirements, and application needs. Its core principle lies in: utilizing the distributed computing capabilities of cloud computing to rapidly decompose the total project duration into individual task nodes; combining historical cost time-series forecasts and real-time resource price data; generating the optimal combination of construction methods under funding constraints through mixed integer programming; and verifying its compliance with requirements using a decision tree algorithm. Its beneficial effects include: 1. improving the efficiency of multi-task parallel optimization; 2. dynamically integrating market resource price fluctuations to enhance the accuracy of cost prediction; 3. reducing manual intervention costs through an automated decision-making chain, thereby shortening the construction period and improving capital utilization management.

[0105] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for dynamic control of engineering costs based on cloud computing, characterized in that, include: Determine the upper limit of available funds for the project, the project completion timeframe, and the project application requirements; Based on the construction progress of the project, the completion time limit is divided according to the task type, and the construction time limit of each task type of the project is obtained. Based on the construction time limit of each task type of the project as the basic constraint, and the upper limit of the project's available funds as the variable constraint, the construction method of each task type of the project is generated. Determine whether the construction methods for each task type of the project meet the application requirements of the project. If not, determine to re-screen; if yes, determine to execute normally. The construction methods for each task type of the generated engineering project include: Based on several known construction methods for engineering buildings, obtain the time series parameters of historical cost changes of construction resources for each construction method; The time series parameters of historical cost changes in construction resources for each construction method are stabilized. Based on AR time series autoregression, a construction cost prediction model for engineering building construction methods is established. The historical cost change time series parameters of construction resources for each construction method are substituted into the construction cost prediction model of the construction method of the engineering building. The historical cost change time series parameters of construction resources for each construction method are used as the input of the observed value, and the predicted real-time construction cost of each construction method in the future is used as the output of the predicted value. Determine the construction time requirements and real-time construction cost requirements for each construction method, and establish a time-cost requirement database for known construction methods. The objective function is established to minimize the total construction period of the project; The constraints are established by using the construction time limits of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint. Based on mixed-integer linear programming, using a known construction method time-cost database as input, and under constraints, select several known construction methods that satisfy the objective function to form an initial set of known construction methods for each task type of the selected project. Based on the decision tree, the construction process of each task type in the project is taken as the root node; the known construction methods of each task type in the project are taken as the branch nodes according to the initial screening of the root node of the construction process of each task type; and the branch decision is based on the information gain of each branch node in minimizing the total construction period of the project, so as to obtain the construction methods of each task type in the project.

2. The method for dynamic control of engineering costs based on cloud computing according to claim 1, characterized in that: The construction progress of engineering projects is divided according to task type to obtain the construction progress of several task types of the engineering project. Verify the reachability between the construction processes of several task types in the project, and construct a tree-like hierarchical task path dependency matrix for the project; Obtain construction time limit parameters for several task types from historical engineering project information, and establish a reference database of construction time limits for task types of historical engineering projects. By marking the task type and construction time limit of historical engineering projects and referring to several tasks of the same type as the engineering project in the database, the construction progress and duration samples of several task types of the engineering project are obtained. Using construction progress and duration samples of several task types in an engineering project as reference objects for scatter plot estimation, the expected duration vectors of the construction progress of several task types in the engineering project are extracted.

3. The method for dynamic control of engineering costs based on cloud computing according to claim 2, characterized in that: Based on the tree-like hierarchical task path dependency matrix of the project, and using the expected duration vectors of the construction process of several task types in the project, we can verify the forward and reverse execution duration paths of the construction process of several task types in the project. By utilizing the forward and reverse execution schedule paths of the construction processes for several task types in an engineering project, the allowable floating time for the construction processes of several task types in the engineering project is calculated, and the execution paths for the construction processes of several task types in the engineering project are determined.

4. The method for dynamic control of engineering costs based on cloud computing according to claim 3, characterized in that: Based on the expected duration vector of the construction process of several task types in the project and the allowable floating time of the construction process of several task types in the project, calculate the spatial distance of the construction process of each task type in the tree-structured task path dependency matrix of the project. Using the k-means clustering algorithm, tasks are partitioned to minimize the spatial distance of construction processes for each task type in the tree-like hierarchical task path dependency matrix of the project, thereby obtaining a set of construction processes for parallel and non-parallel task types in the project. Based on the set of construction processes for parallel and non-parallel task types in an engineering project, and according to the expected duration vector of the construction process for several task types in the engineering project, the forward execution duration path and the reverse execution duration path of the construction process, the allowed floating time of the construction process, and the execution path of the construction process, a construction time limit vector is assigned to each task type of the engineering project.

5. The method for dynamic control of engineering costs based on cloud computing according to claim 4, characterized in that: Based on the construction methods of various task types in the project, the performance indicators of each construction method are quantified, and a scatter plot of the performance of the construction methods of various task types in the project is established. Based on the application requirements of the engineering project, mark the construction method performance requirements of each task type of the engineering project; Based on the cosine similarity formula, calculate the matching degree between the performance requirements of the construction methods of each task type in the project at each scatter point in the performance scatter plot of the construction methods of each task type in the project. The system determines whether the construction methods of each task type in the project meet the application requirements of the project based on the matching degree between the construction methods and performance requirements of each task type in the project corresponding to each scatter point. If not, it determines whether to re-filter; if yes, it determines whether to execute normally.

6. A cloud computing-based dynamic cost control system for engineering projects, characterized in that, A method for implementing a cloud-based dynamic cost management and control method for engineering projects as described in any one of claims 1-5 includes: Initial module, construction time limit allocation module, construction method screening module, construction decision module; The initial module is used to determine the upper limit of available funds for the project, the project completion time limit, and the project application requirements. The construction time limit allocation module is electrically connected to the initial module. The construction time limit allocation module is used to divide the completion time limit according to the task type based on the construction progress of the project, so as to obtain the construction time limit of each task type of the project. The construction method screening module is electrically connected to the initial module and the construction time limit allocation module. The construction method screening module is used to generate construction methods for each task type of the project based on the construction time limit of each task type of the project as the basic constraint and the upper limit of the project's available funds as the variable constraint. The construction decision module is electrically connected to the initial module and the construction method screening module. The construction decision module is used to determine whether the construction method of each task type of the project meets the application requirements of the project. If not, it determines to re-screen; if so, it determines to execute normally.

Images