A method, system, and related equipment for predicting engineering costs based on MLP
By acquiring construction phase information and generating detailed project cost tables using the MLP model, the problem of MLP models being unable to display cost changes during the project cost prediction process is solved, thus realizing multi-stage display and accurate prediction of project costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING MUNICIPAL CONSTR ENG CONSTR CONSULTING CO LTD
- Filing Date
- 2023-01-31
- Publication Date
- 2026-05-26
AI Technical Summary
Existing MLP models cannot effectively represent the possible cost changes during the engineering cost prediction process, resulting in limited prediction results that cannot meet the adjustment needs during the engineering process.
By acquiring construction phase information of the target project, the MLP cost prediction model is used to obtain construction variables for different construction phases, and detailed cost tables are generated based on these variables, including cost tables for the research phase, design phase, and extension phase. Combined with construction adjustment information and construction quality, multiple adjustment cost tables are generated to meet the changes in the project process.
It provides a detailed display of possible changes in project cost, facilitating cost control and decision-making for managers, improving the accuracy and diversity of cost forecasts, and meeting the adjustment needs during the project process.
Smart Images
Figure CN116011660B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of engineering cost estimation, and in particular to an MLP-based engineering cost prediction method, system, and related equipment. Background Technology
[0002] MLP models are commonly used neural network training models. For example, they are frequently used in the prediction of engineering costs. By inputting historical data of the project into the MLP model for training, the model needed for engineering cost prediction can be obtained, which can effectively reduce the workload of cost prediction personnel.
[0003] While MLP models can predict project costs, the predictions are usually quite simplistic. During the project, various variables will be adjusted, and a single prediction result cannot represent the possible cost changes that may occur during the project. Summary of the Invention
[0004] In order to better demonstrate the possible cost changes that may occur during the project, this application provides a method, system and related equipment for predicting project costs based on MLP.
[0005] Firstly, this application provides an MLP-based method for predicting engineering costs, employing the following technical solution:
[0006] A method for predicting engineering costs based on MLP includes:
[0007] Obtain construction stage information for the target project, including the scheme study stage, preliminary design stage, and design expansion stage;
[0008] Based on the construction stage information, obtain the corresponding construction variables for each construction stage;
[0009] Obtain a preset MLP cost prediction model, which includes different quantities of project costs corresponding to different construction stages;
[0010] Based on the construction variables and the MLP cost prediction model, cost tables for different construction stages are obtained.
[0011] By adopting the above technical solution, construction variables for different construction stages are obtained based on construction stage information. Then, cost estimates for different construction stages are generated using the MLP cost prediction model and the corresponding construction variables. This allows for the acquisition of cost estimates for different stages and provides a better representation of potential cost changes during the project.
[0012] Preferably, obtaining cost tables for different construction stages based on the construction variables and the MLP cost prediction model includes:
[0013] When the construction stage information is the scheme research stage, the construction variables of the target project in the scheme research stage are obtained as research stage variables;
[0014] All project costs are obtained based on the MLP cost prediction model and the variables of the research phase, and are used as the research phase cost.
[0015] Based on the research phase cost, the research phase cost table is obtained in a preset order;
[0016] When the construction stage information is the preliminary design stage, the construction variables of the target project in the preliminary design stage are obtained and used as design stage variables;
[0017] All project costs are obtained based on the MLP cost prediction model and the design stage variables, and are used as the design stage cost.
[0018] The design phase cost table is obtained based on the design phase cost table and the research phase cost table;
[0019] When the construction stage information is the design extension stage, the construction variables of the target project in the design extension stage are obtained as extension stage variables;
[0020] All project costs are obtained based on the MLP cost prediction model and the extended stage variables, and are used as the extended stage cost.
[0021] The extended stage cost table is obtained based on the extended stage cost and the design stage cost table;
[0022] The number of engineering costs in the research phase cost schedule is greater than the number of engineering costs in the design phase cost schedule, and the number of engineering costs in the design phase cost schedule is greater than the number of engineering costs in the expansion phase cost schedule.
[0023] By adopting the above technical solution, the cost schedule for the research phase is first obtained, then the cost schedule for the design phase is obtained based on the cost schedule for the design phase and the cost schedule for the research phase, and finally the cost schedule for the extended phase is obtained based on the variables of the extended phase and the cost schedule for the design phase. This allows the cost schedule to be refined according to different phases, making it easier for managers to consider the possible costs at different phases and to control the cost of the project.
[0024] Preferably, after obtaining the extended stage cost table based on the extended stage cost and the design stage cost table, the method further includes:
[0025] Obtain construction adjustment information;
[0026] Based on the construction information, determine whether there is any information regarding schedule adjustment;
[0027] If such information exists, the contract modification information is obtained based on the aforementioned schedule adjustment information.
[0028] The contract adjustment price is obtained based on the aforementioned contract information;
[0029] A first adjusted cost schedule is obtained based on the extended stage cost schedule and the contract adjustment price.
[0030] By adopting the above technical solution, it is possible to determine whether there is any schedule adjustment information based on the construction adjustment information. If so, the contract adjustment price is obtained, and the first adjustment cost table is obtained based on the extended stage cost table and the contract adjustment price. This allows for better cost prediction and improves the accuracy of cost prediction.
[0031] Preferably, after obtaining the extended stage cost table based on the extended stage cost and the design stage cost table, the method further includes:
[0032] Retrieve historical construction team information for all historical construction teams;
[0033] Based on the historical construction team information, the unit price for adjusting the construction period and the corresponding construction quality are obtained;
[0034] A second adjusted cost table is obtained based on the extended cost table and the adjusted unit price for the construction period;
[0035] A quality cost table is obtained based on the construction quality and the second adjusted cost table.
[0036] By adopting the above technical solution, a second adjusted cost schedule is obtained by expanding the cost schedule and adjusting the unit price for the construction period. A quality cost schedule is then obtained based on the construction quality and the second adjusted cost schedule. This allows the obtained cost schedule to be selected according to the construction quality, ensuring that the obtained cost schedule has more reliable quality as much as possible.
[0037] Preferably, the step of obtaining the quality cost schedule based on the construction quality and the second adjusted cost schedule includes:
[0038] Obtain the cost difference between the second adjusted cost schedules of the historical construction teams;
[0039] Determine whether there is a cost difference exceeding the difference threshold;
[0040] If they exist, then obtain the two second adjusted cost tables corresponding to the cost difference, and use them as the filter cost tables;
[0041] Obtain the selected cost tables corresponding to the historical construction teams with better construction quality from the two selected cost tables, and use them as the retained cost tables;
[0042] The retained cost list is sorted based on the construction quality, and the quality cost list is obtained.
[0043] If it does not exist, the second adjusted cost table is sorted based on the preset priority of the construction quality, and the quality cost table is obtained.
[0044] By adopting the above technical solution, it is possible to determine whether there is a cost difference exceeding the difference threshold, and whether the difference between the two second adjusted cost tables is too large. If there is a difference, it is proven to be too large. In order to ensure construction quality, the screening cost table corresponding to the better quality is selected as the retained cost table, and after sorting, the quality cost table is obtained. If there is no difference, the second adjusted cost table is directly matched according to the construction quality according to the preset priority to obtain the quality cost table, thereby improving the diversity of choices.
[0045] Preferably, after obtaining the screened cost table, the method further includes:
[0046] Obtain the construction quality of the historical construction teams corresponding to the two aforementioned cost estimates, and use it as the screening quality.
[0047] When both screening quality requirements are met, the construction cost of the two screening cost tables is obtained as the screening cost.
[0048] Obtain the screening cost table corresponding to the smaller of the two screening costs, and use it as the retained cost table;
[0049] The retained cost schedule is sorted based on the construction quality, and the quality cost schedule is obtained.
[0050] By adopting the above technical solution, when both screening quality standards meet the preset quality requirements, the screening cost table with the lower screening cost is obtained as the retained cost table, and after sorting, the quality cost table is obtained. Thus, the obtained cost table can meet the quality requirements while saving costs.
[0051] Preferably, after obtaining the cost tables for different construction stages based on the construction variables and the MLP cost prediction model, the method further includes:
[0052] Obtain the preset maximum acceptable construction cost;
[0053] The cost table is filtered based on the maximum acceptable cost to obtain the upper limit cost table.
[0054] By adopting the above technical solution, the cost list is screened based on the highest acceptable cost, and an upper limit cost list is obtained. This reduces the number of costs in the cost list and makes it easier for managers to evaluate and screen costs.
[0055] Secondly, this application provides an MLP-based engineering cost prediction system, which adopts the following technical solution:
[0056] A method for predicting engineering costs based on MLP, comprising:
[0057] The information acquisition module is used to acquire construction stage information of the target project, including the scheme study stage, the preliminary design stage, and the design expansion stage.
[0058] The variable acquisition module is used to acquire the construction variables of the corresponding construction stage based on the construction stage information.
[0059] The model acquisition module is used to acquire a preset MLP cost prediction model, which includes different quantities of project costs corresponding to different construction stages.
[0060] The cost acquisition module is used to obtain cost tables for different construction stages based on the construction variables and the MLP cost prediction model.
[0061] By adopting the above technical solution, construction variables for different construction stages are obtained based on data transmission between modules and construction stage information. Then, cost estimates for different construction stages are generated according to the MLP cost prediction model and the corresponding construction variables. This allows for the acquisition of cost estimates for different stages and provides a better representation of potential cost changes during the project.
[0062] Thirdly, this application provides a smart terminal, which adopts the following technical solution:
[0063] A smart terminal, comprising:
[0064] Memory is used to store computer programs that can run on a processor;
[0065] The processor, when running the computer program, is capable of performing the steps of any of the methods described above.
[0066] By adopting the above technical solution, the memory can store information, the processor can retrieve the information and issue control instructions, ensuring the orderly execution of the program and achieving the effect of the above solution.
[0067] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:
[0068] A computer-readable storage medium storing a computer program that can be loaded by a processor and executed by any of the methods described above.
[0069] By adopting the above technical solution, when the computer-readable storage medium is loaded into any computer, any computer can execute the engineering cost prediction method based on MLP provided in this application.
[0070] In summary, this application includes at least one of the following beneficial technical effects:
[0071] 1. Based on the construction phase information, obtain the construction variables for different construction phases, and then use the MLP cost prediction model and corresponding construction variables to obtain cost tables for different construction phases. This allows for the acquisition of cost tables for different phases and provides a better representation of potential cost changes during the project.
[0072] 2. First, obtain the cost schedule for the research phase of the scheme research phase. Then, obtain the cost schedule for the design phase based on the cost schedule for the design phase and the cost schedule for the research phase. Next, obtain the cost schedule for the extended phase based on the variables of the extended phase and the cost schedule for the design phase. This allows the cost schedule to be refined according to different phases, making it easier for managers to consider the possible costs at different phases and to control the cost of the project.
[0073] 3. By expanding the cost schedule and adjusting the unit price based on the construction period, a second adjusted cost schedule can be obtained. Based on the construction quality and the second adjusted cost schedule, a quality cost schedule can be obtained. This allows the obtained cost schedule to be selected according to the construction quality, ensuring that the obtained cost schedule has more reliable quality as much as possible. Attached Figure Description
[0074] Figure 1 This is a flowchart illustrating an MLP-based engineering cost prediction method provided in an embodiment of this application.
[0075] Figure 2 This is a flowchart illustrating steps S11 to S19 in one embodiment of this application;
[0076] Figure 3 This is a flowchart illustrating steps S21 to S25 in one embodiment of this application;
[0077] Figure 4 This is a flowchart illustrating steps S31 to S34 in one embodiment of this application;
[0078] Figure 5 This is a flowchart illustrating steps S41 to S46 in one embodiment of this application;
[0079] Figure 6This is a flowchart illustrating steps S51 to S54 in one embodiment of this application;
[0080] Figure 7 This is a flowchart illustrating steps S61 to S62 in one embodiment of this application;
[0081] Figure 8 This is a structural block diagram of an engineering cost prediction system based on MLP provided in an embodiment of this application.
[0082] Explanation of reference numerals in the attached figures:
[0083] 1. Information Acquisition Module; 2. Variable Acquisition Module; 3. Model Acquisition Module; 4. Cost Acquisition Module. Detailed Implementation
[0084] The following is in conjunction with the appendix Figures 1 to 8 This application will be described in further detail.
[0085] This application discloses an engineering cost prediction method based on MLP.
[0086] Reference Figure 1 MLP-based project cost prediction includes:
[0087] S1. Obtain construction phase information for the target project;
[0088] The target project is the project for which the current cost is predicted. The construction stage information includes the scheme study stage, the preliminary design stage, and the design expansion stage. The information can be obtained by recording the start signal of the design at the end of each stage and collecting information from each stage to determine the construction stage information.
[0089] S2. Obtain the corresponding construction variables for each construction stage based on the construction stage information;
[0090] In other words, the construction variables are different for each stage in the construction phase information. By determining the current construction phase, the corresponding construction variables can be determined.
[0091] S3. Obtain the preset MLP cost prediction model;
[0092] The MLP cost prediction model is trained using historical project data. It includes different project costs for different construction stages. In other words, each construction stage can have multiple project costs depending on the specific construction variables.
[0093] S4. Obtain cost tables for different construction stages based on construction variables and the MLP cost prediction model;
[0094] Finally, the cost schedules for different construction stages are predicted using construction variables and the MLP cost prediction model. The construction variables for the scheme study stage include those for the preliminary design stage, and the construction variables for the preliminary design stage include those for the design extension stage.
[0095] Therefore, the amount of project cost in the cost schedule obtained at each stage is different, so different cost schedules can be provided to managers at different construction stages, and multiple project costs can be obtained for reference at each stage, so as to better show the cost changes that may occur during the project.
[0096] Reference Figure 2 To facilitate the acquisition of cost estimates, in another embodiment, cost estimates for different construction stages are obtained based on construction variables and an MLP cost prediction model, including:
[0097] S11. When the construction stage information is at the scheme study stage, obtain the construction variables of the target project in the scheme study stage as the study stage variables;
[0098] When the construction phase information is determined to be the scheme study phase, because the design details are not sufficient at this time, there are the most construction variables. Therefore, the construction variables of the target project in the scheme study phase are obtained as the study phase variables.
[0099] S12. Based on the MLP cost prediction model and research stage variables, obtain all project costs as the research stage cost;
[0100] Then, based on the MLP cost prediction model and the variables of the research stage, all the engineering costs are obtained as the research stage costs. For example, in the scheme research stage, the variables that can be determined are divided into three categories: A, B, and C. Under each category, there are three intermediate categories for the preliminary design stage, and under each intermediate category, there are three subcategories for the design extension stage. At this time, twenty-seven engineering costs will be generated, which are the research stage costs.
[0101] When these construction variables are added to the MLP cost prediction model, different project costs can be generated based on each different variable. In other words, each sub-category corresponds to a project cost, and the project costs of all sub-categories constitute the cost of the research phase.
[0102] S13. Obtain the research phase cost table in a preset order based on the research phase cost;
[0103] Then, the research phase costs are sorted according to a preset order, which yields the research phase cost list. The preset order can be sorted from high to low prices. Of course, in other implementations, it can also be sorted from low to high prices.
[0104] S14. When the construction stage information is in the preliminary design stage, obtain the construction variables of the target project in the preliminary design stage and use them as design stage variables;
[0105] When the construction stage information is in the preliminary design stage, the major category of variables has been determined, but the minor category of variables has not yet been determined. Therefore, the construction variables of the target project in the preliminary design stage are obtained as design stage variables. For example, when the major category of variables is determined to be A, the corresponding minor category of variables can be set as A1, A2 and A3, and there will be nine corresponding minor categories. The variables corresponding to these nine minor categories are the design stage variables.
[0106] S15. Obtain all project costs based on the MLP cost prediction model and design stage variables, and use them as the design stage cost;
[0107] Then, based on the MLP cost prediction model and design stage variables, all project costs are obtained as design stage costs, using the same method as the research stage costs.
[0108] S16. Obtain the design phase cost table based on the design phase cost table and the research phase cost table;
[0109] Then, based on the cost estimates for the design phase and the cost estimates for the research phase, the cost estimate for the design phase is obtained. In other words, based on the cost estimate for the research phase, only the cost estimate for the design phase is retained, and the resulting cost estimate is the cost estimate for the design phase.
[0110] S17. When the construction stage information is the design extension stage, obtain the construction variables of the target project in the design extension stage as the extension stage variables;
[0111] When the construction phase information is in the design extension phase, the class of the variable has been determined, but the subclass of the variable has not yet been determined. Therefore, the construction variables of the target project in the design extension phase are obtained at this time and used as extension phase variables. For example, when the class of the variable is determined to be A1, the corresponding subclass of the variable can be set to A11, A12 and A13, that is, there are three remaining variables, which are the extension phase variables.
[0112] S18. Obtain all project costs based on the MLP cost prediction model and extended stage variables, and use them as the extended stage costs;
[0113] Then, based on the MLP cost prediction model and extended stage variables, all project costs are obtained as extended stage costs, and the method of obtaining these costs is the same as that of obtaining costs in the design stage.
[0114] S19. Obtain the extended stage cost schedule based on the extended stage cost schedule and the design stage cost schedule;
[0115] Finally, the extended stage cost table is obtained based on the extended stage cost table and the design stage cost table. The method of obtaining the extended stage cost table is the same as that of obtaining the design stage cost table, that is, only the extended stage cost is retained. In other words, the extended stage cost table only includes the cost corresponding to construction variables A11, A12 and A13.
[0116] Among them, the number of engineering costs in the cost schedule of the research phase is greater than the number of engineering costs in the cost schedule of the design phase, and the number of engineering costs in the cost schedule of the design phase is greater than the number of engineering costs in the cost schedule of the expansion phase.
[0117] Therefore, the above methods can display the project cost at each different stage, making it convenient for managers to select the project cost based on the predicted project cost and preset cost requirements, and to make certain adjustments to the cost requirements and construction requirements, thus enabling decision-makers to better manage project costs.
[0118] Reference Figure 3 In the actual design process, there may be temporary adjustments to construction requirements, which may lead to changes in the contract and thus affect the project cost. Therefore, in another embodiment, after obtaining the extended stage cost schedule based on the extended stage cost and the design stage cost schedule, the following steps are also included:
[0119] S21. Obtain construction adjustment information;
[0120] Construction adjustment information refers to data sent to the system via terminal devices. This information includes the number of days for which the construction period is adjusted, as well as price adjustment information.
[0121] S22. Determine whether there is any schedule adjustment information based on construction information;
[0122] In other words, it determines whether the number of days for schedule adjustment in the construction adjustment information is zero. If it is zero, it means that there is no schedule adjustment information; otherwise, it means that there is.
[0123] S23. If it exists, obtain the contract modification information based on the schedule adjustment information;
[0124] If it does not exist, no cost adjustment is needed. If it does exist, then the contract modification information is obtained based on the schedule adjustment information. That is, the construction adjustment information includes the schedule adjustment information and the corresponding contract modification information.
[0125] S24. Obtain contract adjustment prices based on contract information;
[0126] The contract modification information includes the contract adjustment price corresponding to the adjustment of the construction period, which can be obtained when the contract information is acquired.
[0127] S25. Obtain the first adjusted cost schedule based on the extended phase cost schedule and the contract adjustment price;
[0128] In other words, the contract adjustment price is added to or subtracted from each cost in the extended stage cost schedule, forming a new cost schedule, which is the first adjusted cost schedule. If the contract adjustment price is a decrease, it is subtracted; if the contract adjustment price is an increase, it is added.
[0129] Therefore, based on the above methods, the project cost can be adjusted according to the construction adjustment information, so that the obtained project cost table can be more accurate, enabling decision-makers to make better selections and analyses, and thus better manage project costs.
[0130] Reference Figure 4 In order to better manage project costs, it is also necessary to have requirements for construction quality. Therefore, in another embodiment, after obtaining the extended stage cost table based on the extended stage cost table and the design stage cost table, it also includes:
[0131] S31. Obtain historical construction team information for all historical construction teams;
[0132] In other words, the system stores construction information of construction teams that have cooperated with the company in the past, which is the historical construction information of the construction teams.
[0133] S32. Adjust the unit price and corresponding construction quality based on historical construction team information to determine the construction period;
[0134] In other words, the historical construction team information includes the unit price for adjusting the construction period and the construction quality of each construction team. The unit price for adjusting the construction period is the cost that needs to be adjusted for each day the construction period is increased or decreased. The construction quality is the construction quality determined after a comprehensive evaluation of the quality of the project delivered by the construction teams that have cooperated with us in the past. That is, the construction quality is data that is marked and stored for each construction team based on historical data.
[0135] S33. Obtain the second adjusted cost schedule based on the extended cost schedule and the unit price adjusted for the construction period;
[0136] In other words, the project cost is obtained by adding or subtracting the unit price for the construction period adjustment on the basis of the expanded cost table. Continuing with the example above, each historical construction team has three adjusted project costs, that is, each historical construction team has a cost table, which is the second adjusted cost table.
[0137] S34. Obtain the quality cost schedule based on construction quality and the second adjusted cost schedule;
[0138] Finally, the second adjusted cost table is sorted according to the construction quality from high to low, and the resulting cost table is the quality cost table. Of course, in other embodiments, the construction quality can also be sorted in order from low to high.
[0139] Therefore, by using the above methods, when there are high requirements for project quality, the project cost can be compared and selected based on the quality cost list, which can further increase the range of choices and reduce the difficulty of selection.
[0140] Reference Figure 5 To better adjust the cost schedule and meet selection requirements, in another embodiment, a quality cost schedule is obtained based on construction quality and a second adjusted cost schedule, including:
[0141] S41. Obtain the cost difference between the second adjusted cost schedules of the historical construction teams;
[0142] A specific method could be to obtain the total cost of each historical construction team's project, and then obtain the difference between the total costs of each historical construction team, which is the cost difference.
[0143] S42. Determine if there is a cost difference exceeding the difference threshold;
[0144] This involves comparing the cost difference with a threshold value. For example, subtracting the threshold value from the cost difference: if the result is greater than zero, it indicates a cost difference exceeding the threshold; if the result is less than or equal to zero, it indicates no cost difference exceeding the threshold. Of course, the threshold value can be set according to actual needs.
[0145] S43. If it exists, then obtain the two second adjustment cost tables corresponding to the cost difference, and use them as the screening cost tables;
[0146] If there is a cost difference exceeding the difference threshold, it proves that the difference between the two cost tables is too large. In this case, obtain the two second adjustment cost tables corresponding to the cost difference as the screening cost tables.
[0147] S44. Obtain the filter cost table corresponding to the historical construction team with better construction quality from the two filter cost tables, and keep it as the retain cost table;
[0148] This means obtaining the historical construction team with higher construction quality from the two selected cost tables and keeping it as the retained cost table. For example, if the construction quality of historical construction teams is sorted into A, B, C, and D, then A corresponds to the highest construction quality and D corresponds to the lowest construction quality.
[0149] Assuming that the ranking level of historical construction team B is A, the ranking level of historical construction team A is B, and the cost tables corresponding to the strong construction team A and the historical construction team B are screening cost tables, then the cost table to be retained is the screening cost table corresponding to historical construction team B.
[0150] S45. Sort the retained cost schedules based on construction quality and obtain the quality cost schedule;
[0151] Finally, all the retained cost estimates are sorted again according to the construction quality from highest to lowest. The resulting cost estimates are the quality cost estimates, which provides multiple cost estimates that meet the requirements, thereby improving the accuracy and convenience of selection.
[0152] S46. If it does not exist, sort the second adjustment cost table based on the preset priority of construction quality and obtain the quality cost table;
[0153] If there is no cost difference exceeding the difference threshold, it proves that the difference between the two cost tables is small. Therefore, no other screening steps are needed. Instead, the second adjustment cost table can be sorted directly according to the preset priority of construction quality to obtain the quality cost table.
[0154] The preset priority means sorting the costs from highest to lowest based on construction quality. Of course, in other implementations, the priority can also be sorted from lowest to highest. This results in a cost list sorted by quality, making it easier for decision-makers to make selections.
[0155] Reference Figure 6 In order to better save costs while ensuring quality, in another embodiment, after obtaining the screening cost list, the following steps are also included:
[0156] S51. Obtain the construction quality of the historical construction teams corresponding to the two filter cost tables, and use it as the filter quality;
[0157] Construction quality data is pre-stored. Each historical construction team corresponds to a construction quality. Once the historical construction team corresponding to the selected cost table is determined, the construction quality of the corresponding historical construction team can be obtained, i.e., the selection quality.
[0158] S52. When both screening quality requirements are met, the construction cost of the two screening cost tables is obtained and used as the screening cost.
[0159] During the screening process of the quality cost list, if both screened quality standards meet the preset quality requirements, that is, the construction quality of both historical construction teams meets the minimum requirements, the preset quality requirements can be set according to the actual situation. For example, it can be set to ranking level B. That is, when the construction quality is greater than or equal to ranking level B, it proves that the quality requirements are met; otherwise, it is not met.
[0160] Therefore, when both screening quality standards meet the preset quality requirements, the construction cost of the two screening cost tables is obtained as the screening cost. The screening cost can be obtained by obtaining the total value of the project cost of each historical construction team, which is the screening cost.
[0161] S53. Obtain the screening cost table corresponding to the smaller screening cost among the two screening costs, and use it as the retained cost table;
[0162] Then, the two screening costs are compared, and the screening cost table with the smaller screening cost is obtained as the retained cost table. That is, when the construction quality of both historical construction teams meets the requirements, the cost table corresponding to the smallest project cost is obtained, so as to reduce costs while ensuring construction quality.
[0163] S54. Sort the retained cost schedules based on construction quality and obtain the quality cost schedule.
[0164] Finally, all the retained cost estimates are sorted in descending order of construction quality to obtain a quality cost estimate. This not only ensures that the cost estimates meet the requirements, but also reduces the number of cost estimates that meet the requirements, thus simplifying the selection process.
[0165] Reference Figure 7 To enable faster filtering of cost estimates, in another embodiment, after obtaining cost estimates for different construction stages based on construction variables and the MLP cost prediction model, the following steps are also included:
[0166] S61. Obtain the preset maximum acceptable cost;
[0167] S62. Filter the cost schedule based on the highest acceptable cost and obtain the upper limit cost schedule.
[0168] Specifically, the system pre-stores the maximum acceptable cost, which is the maximum price that can be paid during the construction process. Then, the cost table is filtered according to the maximum acceptable cost to obtain the upper limit cost table. In other words, all project costs in the cost table that are greater than the maximum acceptable cost are deleted. The remaining project costs form the upper limit cost table.
[0169] Once the cost schedule is obtained, a preliminary screening can be conducted based on the highest acceptable cost, which facilitates the selection and management of the predicted project cost in the future.
[0170] The implementation principle of the MLP-based engineering cost prediction method in this application is as follows: First, the construction stage information of the target project is obtained. Then, construction variables for different construction stages are obtained based on the construction stage information. Next, a preset MLP cost prediction model is obtained. Finally, cost tables for different construction stages are obtained based on the MLP cost prediction model and the corresponding construction variables. This allows for the acquisition of cost tables for different stages and provides a better representation of potential cost changes during the project.
[0171] This application also discloses an MLP-based engineering cost prediction system, which can achieve the same technical effect as the MLP-based engineering cost prediction method described above.
[0172] Reference Figure 8 MLP-based engineering cost prediction systems include:
[0173] Information acquisition module 1 is used to acquire construction stage information of the target project, including the scheme study stage, preliminary design stage, and design expansion stage.
[0174] Variable acquisition module 2 is used to acquire construction variables for the corresponding construction stage based on construction stage information;
[0175] Model acquisition module 3 is used to acquire a preset MLP cost prediction model, which includes different quantities of project costs corresponding to different construction stages.
[0176] Module 4, which is used to obtain cost tables for different construction stages based on construction variables and the MLP cost prediction model.
[0177] Specifically, information acquisition module 1 acquires construction stage information of the target project and sends it to variable acquisition module 2, which is connected to it. The construction stage information includes the scheme study stage, preliminary design stage, and design expansion stage. Then, variable acquisition module 2 acquires the corresponding construction variables for the construction stage based on the construction stage information and sends them to cost acquisition module 4, which is connected to it.
[0178] Next, the model acquisition module 3 acquires a preset MLP cost prediction model and sends it to the connected cost acquisition module 4. The MLP cost prediction model includes different quantities of project costs corresponding to different construction stages. Finally, the cost acquisition module 4 obtains cost tables for different construction stages based on construction variables and the MLP cost prediction model.
[0179] This allows for the acquisition of cost estimates at different stages, and provides a better view of potential cost changes during the project.
[0180] This application also discloses an intelligent terminal, including a memory and a processor. The memory stores an intelligent computer program. The processor, when running the intelligent computer program, is capable of executing the steps of the MLP-based engineering cost prediction method described above. The intelligent computer program can use known processing procedures to perform a series of steps such as querying, comparing, and judging data, thereby achieving the prediction of engineering costs.
[0181] This application also discloses a computer-readable storage medium that stores a computer program capable of being loaded and executed by a processor for MLP-based engineering cost prediction as described above. The computer-readable storage medium includes, for example, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0182] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A method for predicting engineering costs based on MLP, characterized in that, include: Obtain construction stage information for the target project, including the scheme study stage, preliminary design stage, and design expansion stage; Based on the construction stage information, obtain the corresponding construction variables for each construction stage; Obtain a preset MLP cost prediction model, which includes different quantities of project costs corresponding to different construction stages; Based on the construction variables and the MLP cost prediction model, cost tables for different construction stages are obtained; specifically, the following sub-steps are included: When the construction stage information is the scheme research stage, the construction variables of the target project in the scheme research stage are obtained as research stage variables; All project costs are obtained based on the MLP cost prediction model and the variables of the research phase, and are used as the research phase cost. Based on the research phase cost, the research phase cost table is obtained in a preset order; When the construction stage information is the preliminary design stage, the construction variables of the target project in the preliminary design stage are obtained and used as design stage variables; All project costs are obtained based on the MLP cost prediction model and the design stage variables, and are used as the design stage cost. The design phase cost table is obtained based on the design phase cost table and the research phase cost table; When the construction stage information is the design extension stage, the construction variables of the target project in the design extension stage are obtained as extension stage variables; All project costs are obtained based on the MLP cost prediction model and the extended stage variables, and are used as the extended stage cost. The extended stage cost table is obtained based on the extended stage cost and the design stage cost table; The number of engineering costs in the research phase cost schedule is greater than the number of engineering costs in the design phase cost schedule, and the number of engineering costs in the design phase cost schedule is greater than the number of engineering costs in the expansion phase cost schedule.
2. The engineering cost prediction method based on MLP according to claim 1, characterized in that, After obtaining the extended stage cost table based on the extended stage cost and the design stage cost table, the method further includes: Obtain construction adjustment information; Based on the construction stage information, determine whether there is any schedule adjustment information; If such information exists, the contract modification information is obtained based on the aforementioned schedule adjustment information. The contract adjustment price is obtained based on the aforementioned contract modification information; A first adjusted cost schedule is obtained based on the extended stage cost schedule and the contract adjustment price.
3. The engineering cost prediction method based on MLP according to claim 1, characterized in that, After obtaining the extended stage cost table based on the extended stage cost and the design stage cost table, the method further includes: Retrieve historical construction team information for all historical construction teams; Based on the historical construction team information, the unit price for adjusting the construction period and the corresponding construction quality are obtained; A second adjusted cost table is obtained based on the extended stage cost table and the adjusted unit price for the construction period; A quality cost table is obtained based on the construction quality and the second adjusted cost table.
4. The engineering cost prediction method based on MLP according to claim 3, characterized in that, The process of obtaining a quality cost schedule based on the construction quality and the second adjusted cost schedule includes: Obtain the cost difference between the second adjusted cost schedules of the historical construction teams; Determine whether there is a cost difference exceeding the difference threshold; If they exist, then obtain the two second adjusted cost tables corresponding to the cost difference, as the screening cost tables; Obtain the selected cost tables corresponding to the historical construction teams with better construction quality from the two selected cost tables, and use them as the retained cost tables; The retained cost list is sorted based on the construction quality, and the quality cost list is obtained. If it does not exist, the second adjusted cost table is sorted based on the preset priority of the construction quality, and the quality cost table is obtained.
5. The engineering cost prediction method based on MLP according to claim 4, characterized in that, After obtaining the filtered cost table, the process also includes: Obtain the construction quality of the historical construction teams corresponding to the two aforementioned cost estimates, and use it as the screening quality. When both screening quality requirements are met, the construction cost of the two screening cost tables is obtained as the screening cost. Obtain the screening cost table corresponding to the smaller of the two screening costs, and use it as the retained cost table; The retained cost schedule is sorted based on the construction quality, and the quality cost schedule is obtained.
6. The engineering cost prediction method based on MLP according to claim 1, characterized in that, After obtaining the cost tables for different construction stages based on the construction variables and the MLP cost prediction model, the method further includes: Obtain the preset maximum acceptable construction cost; The cost table is filtered based on the maximum acceptable cost to obtain the upper limit cost table.
7. A system applicable to the MLP-based engineering cost prediction method according to any one of claims 1-6, characterized in that, include: Information acquisition module (1) is used to acquire construction stage information of the target project, including scheme study stage, preliminary design stage and design expansion stage; The variable acquisition module (2) is used to acquire the construction variables of the corresponding construction stage based on the construction stage information; The model acquisition module (3) is used to acquire a preset MLP cost prediction model, which includes different quantities of project costs corresponding to different construction stages. The cost acquisition module (4) is used to obtain cost tables for different construction stages based on the construction variables and the MLP cost prediction model.
8. A smart terminal, characterized in that, include: Memory is used to store computer programs that can run on a processor; The processor, when running the computer program, is capable of performing the steps of the method as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that: The computer program is stored that can be loaded by a processor and executed according to any one of claims 1 to 6.