Project cost allocation method, system and equipment and storage medium

Abstract

The invention belongs to the technical field of project cost, and provides a project cost allocation method, system and device and a storage medium, and the method comprises the steps: obtaining a deviation value between a planned value and an actual value of each project cost in a project construction process; setting different alarm levels according to different deviation values by using a K-means algorithm; and when the deviation value does not exceed the early warning level, calculating the shared cost of the project, and carrying out cost sharing among the departments according to the shared cost of the project. The method solves the problem that there is no early warning when the project cost deviates from the planned value during partnership operation, and the situation that the cost deviates from the planned value cannot be subjected to alarm judgment.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering cost, and in particular, relates to a project cost allocation method, system, equipment and storage medium. Background Art

[0002] In the process of global economic slowdown, market competition has become more intense, and the traditional pattern of independent operation of enterprises has been gradually broken. Cooperation and partnership have become a trend that is more conducive to the development of enterprises. Under the enterprise partnership model, when there are multiple project investments, especially when multiple projects affect each other, how to reasonably and effectively manage the investment funds related data of each project (such as statistical analysis of fund use related data during project investment, income distribution and automatic processing of expense settlement data, etc.) has become an urgent problem to be solved.

[0003] Among them, during the investment process, a general budget for the project's engineering costs is usually formulated in the project proposal and feasibility study report in the early design stage of the project. However, due to the long construction period, uncertainties such as price changes, policy adjustments, and weather delays often occur during the construction period. Human subjective factors dominate the general budget, and the prediction methods are limited. Therefore, during the construction process, it is inevitable that the project costs will deviate from the planned values.

[0004] The existing technology does not have an early warning for situations where project costs deviate from the planned values, and is unable to make warning judgments for situations where costs deviate from the planned values. Summary of the invention

[0005] To solve the above problems, the present invention provides a project cost sharing method, system, device and storage medium, and adopts the K-means algorithm to construct a deviation alarm calculation model, and obtains a preliminary warning threshold interval by dividing the deviation value N1 into multiple threshold intervals representing multiple alarms according to the deviation ratio Yn; each interval in this preliminary warning threshold interval is used as the initial clustering setting of the K-means algorithm; then, the deviation value N2 of the actual value and the planned value under the remaining monitoring nodes is calculated, and the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval is calculated, and the deviation value N2 is classified into the threshold interval with the smallest Euclidean distance therefrom, and a new warning threshold interval is obtained after classification; the clustering results are updated in this way, which can ensure that managers can make effective capital allocation according to the alarm level when correcting costs.

[0006] Specifically, the present invention is preferably implemented by the following scheme:

[0007] The present disclosure provides a project cost allocation method, comprising: obtaining a deviation value between a planned value and an actual value of each project cost during a project construction process;

[0008] The K-means algorithm is used to set different alarm levels according to different deviation values; wherein, setting different alarm levels includes:

[0009] During the project construction process, multiple monitoring nodes are selected, the deviation value N1 between the actual value and the planned value under the multiple monitoring nodes is calculated, and the deviation ratio Yn is obtained according to the deviation value N1; the deviation value N1 is divided into multiple threshold intervals representing multiple alarm conditions according to the deviation ratio Yn, and the preliminary warning threshold interval is obtained;

[0010] Calculate the deviation value N2 between the actual value and the planned value under the remaining monitoring nodes, calculate the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval, classify the deviation value N2 into the threshold interval with the smallest Euclidean distance, and obtain the new warning threshold interval after classification;

[0011] In the new warning threshold interval, each warning threshold interval corresponds to an alarm level;

[0012] When the deviation value does not exceed the warning level, the project common expenses are calculated and the expenses are shared among departments based on the project common expenses.

[0013] Further,

[0014] The determination of the project common weight includes: using statistical analysis to evaluate the working hours required for the same workload in different projects from multiple dimensions to obtain a comparative amount as the project common weight.

[0015] Further,

[0016] The method of obtaining the deviation between the planned value and the actual value of each project cost includes:

[0017] Collect the cost deviation values ​​of historical projects of the same type and the cost activity status parameters of the corresponding cost nodes as training sets;

[0018] The BP neural network is trained using the training set to obtain a trained cost deviation calculation model for predicting cost deviation values;

[0019] The cost activity status parameters of each project are input into the cost deviation calculation model, and the cost deviation calculation model outputs the deviation value between the planned value and the actual value.

[0020] Further,

[0021] In the K-means algorithm, the criterion function of the Euclidean distance is to minimize the sum of the squares of the distances from the data object to its cluster center, and its expression is:

[0022]

[0023] Among them, k is the number of clusters, Ci is the center point of the first cluster, dist is the distance from x to C i The distance is , and x is a certain deviation value.

[0024] Further,

[0025] The calculation formula for the common area percentage is:

[0026]

[0027] cq=sq*w

[0028] Among them, percent is the percentage of common area; cq is used to indicate the correction amount of common area standard of the project; It is used to represent the sum of the common pool standard corrections of the n items contained in the corresponding common pool expense account. sq is used to represent the calculated common pool standard of the project under the associated common pool standard. w is used to represent the common pool weight of the project.

[0029] Further,

[0030] Genetic algorithm is used to optimize the weights and thresholds of BP neural network during the training process.

[0031] Further,

[0032] The common expenses of the calculation items include:

[0033] At the end of the project investment period, for each common expense item, the total amount of the corresponding item expense is multiplied by the common percentage of each project under the common expense item to obtain the common expense of each project under the common expense item for each day. The common expense of each project for each day is summed up to obtain the total common expense of each project during the investment period.

[0034] The present invention also provides a project cost sharing system, comprising:

[0035] The data acquisition module is used to obtain the deviation between the planned value and the actual value of each project cost during the project construction process;

[0036] The alarm level setting module is used to set different alarm levels according to different deviation values ​​using the K-means algorithm; the K-means algorithm is used to set different alarm levels according to different deviation values; including:

[0037] During the project construction process, multiple monitoring nodes are selected, the deviation value N1 between the actual value and the planned value under the multiple monitoring nodes is calculated, and the deviation ratio Yn is obtained according to the deviation value N1; the deviation value N1 is divided into multiple threshold intervals representing multiple alarm conditions according to the deviation ratio Yn, and the preliminary warning threshold interval is obtained;

[0038] Calculate the deviation value N2 between the actual value and the planned value under the remaining monitoring nodes, calculate the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval, classify the deviation value N2 into the threshold interval with the smallest Euclidean distance, and obtain the new warning threshold interval after classification;

[0039] In the new warning threshold interval, each warning threshold interval corresponds to an alarm level;

[0040] The calculation module is used to calculate the project common expenses and allocate expenses among departments based on the project common expenses when the deviation value does not exceed the warning level.

[0041] Based on the same inventive concept, the present invention provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any of the project cost sharing methods described above.

[0042] A computer storage medium stores executable instructions, which, when executed by a processor, enable the processor to implement any of the project cost allocation methods described above.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] The present invention constructs a deviation alarm calculation model based on the K-means algorithm, and obtains a preliminary warning threshold interval by dividing the deviation value N1 into multiple threshold intervals representing multiple alarms according to the deviation ratio Yn; each interval in the preliminary warning threshold interval is used as the initial clustering setting of the K-means algorithm; then the deviation value N2 between the actual value and the planned value under the remaining monitoring nodes is calculated, and the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval is calculated, and the deviation value N2 is classified into the threshold interval with the smallest Euclidean distance thereto, and a new warning threshold interval is obtained after classification; the clustering result is updated in this way;

[0045] This method takes into account the uncertainty of cost deviations in complex large-scale construction projects, realizes the continuous division of deviation warnings, and avoids the occurrence of warning blanks. Taking the investment deviation ratio of the cost monitoring points of complex large-scale construction projects as the warning analysis element, the model follows the principle of "minimum sum of squares of internal deviations and maximum sum of squares of inter-class deviations", which can effectively define the positive and negative deviations of cost monitoring points, and ensure that managers can effectively allocate funds according to the warning level when correcting cost deviations.

[0046] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 A schematic diagram of the method of the present invention is shown. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0050] Figure 1 A schematic diagram of the method according to the present invention is shown, and the specific implementation details of the present invention include:

[0051] Step 1: Budget classification definition: The budget item classification is provided by the finance department to the budget system administrator for entry into the system before budget preparation each year; it includes:

[0052] 1) Planning stage definition: clarify the stage of project budget planning; 2) Planning project definition: define budget projects and facilitate their application in budget planning; 3) Planning: submit a report on the preparation of non-recurrent budget plans; 4) Recurrent expense plan preparation: submit a report on the preparation of recurrent expense budget plans.

[0053] Step 2: Investment flow query: used to provide the function of comparing the planned value, actual value, and execution budget value of the budget project.

[0054] Step 3: Unit price analysis: define unit price analysis and necessary basic information; 1) Unit price analysis detail type definition, used to determine the unit price analysis item type and indicate whether they are direct costs or indirect costs; 2) Unit price analysis details, establish unit price analysis items; 3) Total unit price: the total price calculated based on the unit price analysis items, which is equal to the sum of the unit price analysis items divided by the standard unit; 4) Total direct cost: the total amount of direct costs in the total price calculated based on the unit price analysis items, which is equal to the sum of all direct unit price analysis items divided by the standard unit; 5) Total indirect cost: the total amount of indirect costs in the total price calculated based on the unit price analysis items, which is equal to the sum of all indirect unit price analysis items divided by the standard unit.

[0055] Step 4: Price difference management, including: 1) defining the price difference increase rate; 2) determining the price difference weight, and maintaining the price difference weight according to the cost type and the construction price difference type of the contract / quotation; 3) maintaining the price difference items for the specified year according to the contract / quotation;

[0056] Step 5: Perform budget control, including:

[0057] 1) Establish a static investment conversion coefficient table for the execution estimate and store the corresponding conversion coefficients. For projects without an execution estimate unit price, the system converts the actual amount according to a certain coefficient; 2) Register the execution estimate and obtain the execution estimate from the quotation sheet; the execution estimate does not include price differences. It is based on the information on contracts signed that year provided by the government. The execution estimate can be compared with the contract price to provide investment statistics.

[0058] Step 6: Cost Control Grants: This is the process of defining cost control grants based on investment statistics and is the process of releasing funds to become expenditures.

[0059] Step 7: Plan execution package: used to create a control package or plan execution package for each project.

[0060] Step 8: Determine the project common weight: Use statistical analysis to evaluate the time required for the same workload in different projects from multiple dimensions to obtain a comparative quantity.

[0061] The calculation of the common area weight is:

[0062]

[0063] cq=sq*w

[0064] Among them, percent represents the percentage of public area, and q is used to represent the standard correction of public area of ​​the project. It is used to represent the sum of the common pool standard corrections of the n items contained in the corresponding common pool expense account. sq is used to represent the calculated common pool standard of the project under the associated common pool standard. w is used to represent the common pool weight of the project. * represents multiplication.

[0065] Step 9: Calculate cost deviation: The cost deviation of a construction project refers to the phenomenon that the actual value of each cost deviates from the planned value during the project construction process. The cost deviation control of the project runs through the entire process of the project. The preparation of the feasibility study report in the decision-making stage is the basis for cost control in the later construction and completion acceptance process. Usually, the project cost budget of the project will be formulated in the project proposal and feasibility study report in the early design stage of the project. However, due to the long duration of the project construction, uncertainties such as price changes, policy adjustments, and weather delays during the construction period often occur. The subjective factors in the budget dominate, and the prediction methods are limited. Therefore, during the construction process, it is inevitable that the project cost will deviate from the planned value. As an important part of construction project management, cost control runs through the entire process of project construction, and is the premise for realizing the effective and reasonable use of project financial, human and material resources, thereby ensuring good project returns. As for the selection of cost correction methods, on the one hand, the specific correction object should be considered, and on the other hand, the basic quality of cost control personnel and the special circumstances of the construction project should also be considered. The implementation of cost control in most engineering construction projects is passive, lagging and inefficient. Control is mostly carried out when cost deviations occur and the threat of deviation attracts attention. In order to move from post-control to pre-control, it is extremely urgent and necessary to establish an effective pre-control system for engineering construction costs.

[0066] Therefore, the present invention adopts the following method to calculate the cost deviation:

[0067] 1) Cost deviation warning calculation model based on K-means algorithm:

[0068] Construction of cost deviation alarm model based on improved K-means clustering model. In order to grasp the negative impact of cost deviation of complex large-scale construction projects on the normal operation of projects, the boundary mean operator was introduced based on the idea of ​​clustering management to construct an improved K-means clustering model. Considering the uncertainty characteristics of cost deviation of complex large-scale construction projects, the model realizes the continuous division of deviation alarms to avoid the occurrence of alarm blanks. Taking the investment deviation ratio of cost monitoring points of complex large-scale construction projects as the alarm analysis element, the model follows the principle of "minimum sum of squares of internal deviations and maximum sum of squares of inter-class deviations", which can effectively define the positive and negative deviations of cost monitoring points and ensure the effective allocation of resources for managers when correcting cost deviations.

[0069] The clustering theory algorithm has a fixed division process, requires fewer samples, is simple to operate, and is easy to implement. In addition, its objective characteristics can largely avoid the arbitrariness of human settings, and can be used to define the cost deviation warning thresholds at various stages of engineering construction projects. As a classic unsupervised clustering algorithm, the K-means algorithm can ensure that the ordered samples are not destroyed, and the data points are divided with the minimization of the sum of squared deviations within the level and the maximization of the sum of squared deviations between levels as the guide, completing the dimensionality reduction clustering of the sample data. The advantages of the K-means algorithm can be summarized as follows: ① There is no requirement for data distribution, and the low data dependence makes it widely used; ② In the face of samples with a high degree of mean collinearity, the algorithm can rely on the discriminant function to simplify the calculation amount; ③ Relying on dimensionality reduction, the algorithm can achieve visual judgment.

[0070] In view of the clustering idea of ​​K-means algorithm, it is feasible to apply it to the discrimination of cost deviation warning and threshold determination of engineering construction projects. By analyzing historical data, the optimal threshold segmentation solution can be effectively found. There are a large number of historical projects in engineering construction projects. Based on the historical database of engineering projects, the cost situation of each activity link in the project can be obtained, and a relatively complete cost deviation database can be formed, which provides sufficient basic data for the warning division behavior using K-means as a clustering method. Based on the potential laws of data, the K-means method can give full play to the driving advantages of "minimizing the deviation between classes and maximizing the deviation between classes", so that the cost deviation data can cluster by itself and continuously update and adjust based on this principle.

[0071] The K-means algorithm is a dynamic clustering algorithm. Its basic idea is to randomly select a number of cluster centers and accumulate various samples to the center points according to the agreed criteria to obtain the initial classification. The final reasonable classification is obtained by judging and updating the rationality of the initial classification. Based on the distance formula to express the similarity between two different objects, the K-means algorithm will stop when the criterion function reaches the optimal value or the number of iterations reaches the maximum value. This paper uses the Euclidean distance as the similarity measurement paradigm. Its criterion function can be described as minimizing the sum of the squares of the distances from the data object to its cluster center, that is:

[0072]

[0073] Where k is the number of clusters, C i is the center point of the first cluster, dist is the distance from x to C i distance.

[0074] The cost deviation early warning model based on K-means clustering theory can classify the cost situation of the same cost monitoring point of multiple similar engineering projects in history into different deviation warning categories, and obtain the final early warning interval by discriminating and analyzing the threat levels of different categories. The construction steps of the early warning model are as follows:

[0075] (1) Determination of initial warning threshold

[0076] Based on the set of deviation monitoring nodes {K1, K2, …, Kn}, the deviation values ​​of the cost monitoring nodes at a certain stage in similar engineering construction projects are collected and preprocessed to obtain the deviation ratio Yn, and Yn is divided into positive and negative groups and arranged according to the specified order. After clarifying the alarm category, the processed data is used to determine the interval threshold according to the idea of ​​the classic optimal segmentation calculation model to obtain the deviation value range under different alarm conditions, but it should be noted that the warning threshold interval is discrete at this time.

[0077] (2) Modify the warning threshold range

[0078] For the cost deviation points of construction projects that are not in the preliminary warning interval, the distance between the point and the representative value of the adjacent warning interval is determined based on the Euclidean distance calculation rule, and the discontinuous points with positive and negative values ​​are classified and determined respectively, and the deviation value interval is optimized to achieve the energy efficiency of continuous warning. If the cost deviation value of the power grid project cost monitoring node K0 (KO∈{K1,K2,…,Kn}) being implemented is p0, pO∈{c3~c2}, then the warning status of the cost monitoring node K0 is "rectification".

[0079] 2) The traditional BP neural network is improved by using bionic algorithm and successfully applied to cost deviation prediction. By simulating the evolutionary process of "survival of the fittest", the initial weights and thresholds in the BP neural network are encoded, and the model is trained based on the existing historical database of cost deviations of complex large-scale construction projects, and finally a cost deviation prediction model for complex large-scale construction projects with fast computing speed and high prediction accuracy is obtained.

[0080] Artificial neural networks, as a collection of artificial neuron connection units, are frameworks based on various machine learning algorithms. By simulating the information processing method of the human brain, the model has realized the process of computing and analyzing according to the logic of the human brain. Inspired by the biological neural network of the animal brain, artificial neural networks simulate the activity patterns of neurons in the biological brain. Each connection can transmit signals between neurons like the synapses in the brain. At the same time, neurons also have information processing functions and send signals to other neurons.

[0081] The BP neural network has a relatively simple structure, an easy-to-understand algorithm, and can be used in many fields. It has strong generalization capabilities and the following advantages: ① Globality. When artificially designing a neural network, it can be set to either linear or nonlinear. A neural network formed by connecting nonlinear neurons is also nonlinear and has strong generalization capabilities; ② Learning and adaptability. The weights in the BP neural network can be adjusted through the back propagation of errors, so it can adapt to changes in the external environment; it can also repeatedly adjust and update parameters during sample data training, and ultimately store the acquired useful information in the network; ③ Mapping function. Neural networks can essentially realize the function of mapping input data to output data, especially after the network structure has been trained, it can well complete various mappings from input to output, and can well solve various "input-output" problems.

[0082] The learning process of BP neural network can be divided into two stages: forward propagation of working signal and reverse propagation of error signal. For the former, the working signal is transmitted from the input layer to the output layer through the hidden layer, during which it will undergo two operations of the activation function, and finally generate an output value in the output layer. For the latter, the error signal is propagated forward layer by layer from the output end to the input end, and the weight of the network is continuously corrected according to the signal error feedback to minimize the difference between the network output result and the expected output. Specifically, the general algorithm of BP neural network can be summarized as follows:

[0083] (1) Set variables and parameters;

[0084] (2) Initialize the network and randomly assign values ​​to the network weights and thresholds;

[0085] (3) Input training samples and calculate the input and output values ​​of neurons in each layer;

[0086] (4) Determine the error between the actual output and the network output value If the deviation meets the requirements, the training is terminated and the network is recorded. If it does not meet the requirements, the next step is performed.

[0087] (5) Determine whether the number of learning steps reaches the maximum number of steps. If it is greater, end the training and record the network. Otherwise, calculate the local gradient of each layer of neurons.

[0088] (6) Calculate the weight correction amount according to the following formula and go to step (3);

[0089] △w(n)=no”(n)v; (n),t=1,2,…,T; u=1,2,…,U

[0090] △w(n)=nw / (n)v}(n),s=1,2,…,S; t=1,2,…,T

[0091] w(n+1)=Wn(n)+△w(n),t=1,2,…,T; u=1,2,…,U

[0092] (7) Determine whether all training samples have been learned. If so, the training ends; otherwise, return to step (3).

[0093] Construction of cost deviation calculation model based on improved neural network model:

[0094] There are many cost nodes in complex large-scale construction projects, and they have extremely complex linkage relationships with each other. The impact of this uncertainty leads to nonlinear characteristics of cost deviation. Complex large-scale construction projects go through multiple stages such as planning, implementation, and operation and maintenance. The work content involves feasibility studies, surveys and designs, bidding, contract signing, construction, acceptance settlement, operation and maintenance, etc. The huge workload and complex work structure cause the cost amount of complex large-scale construction projects to be different from the prior plan in actual use. Considering the one-time and procedural characteristics of engineering projects, there is a phenomenon that the cost of complex large-scale construction projects is transmitted from the early stage to the later stage. The deviation of costs in the transmission process may lead to a "butterfly effect" in engineering costs. Therefore, building a cost deviation prediction model is of great significance for the cost management of complex large-scale construction projects.

[0095] The neural network forms a system with adaptive and self-learning capabilities based on the input layer, hidden layer and output layer. Through the transmission and training of information between neurons, it can effectively simulate the potential laws of things. The macro-cost deviation of complex large-scale construction projects is affected by the cost activities of the entire life cycle. Combined with the characteristics of neural networks, the cost deviation measurement values ​​occurring in each cost activity stage can be regarded as neural network input elements, and the actual total deviation can be used as the output value to train the neural network model to obtain a calculation model that can achieve deviation prediction.

[0096] The basic data required for training the neural network is obtained by collecting the cost deviation of a large number of similar complex large-scale construction projects and the cost activity status of the corresponding cost nodes. The basic parameters of the neural network and genetic algorithm are set, and the total cost deviation of complex large-scale construction projects is simulated with the prediction error minimization as the optimization guide of the neural network and genetic algorithm.

[0097] The improved neural network cost deviation prediction framework includes two parts: cost deviation data collection and network training. Taking complex large-scale construction projects as the object, by inputting a large number of cost deviation measurement values ​​of the same type of cost activities, relying on the neural network optimized by genetic algorithm to learn and simulate the potential relationship between data, the cost deviation prediction function is realized. The model introduces machine learning theory and uses bionic algorithm to optimize the neural network, so that the generalization ability and learning ability of the model are further improved. The performance of the improved model is significantly optimized in terms of training time and prediction accuracy, reducing the negative impact of the network initial random weights and thresholds on network learning, and can effectively adapt to the multi-dimensional and complex environment of cost deviation prediction of complex large-scale construction projects. Based on the above ideas, the deviation prediction of each cost node of complex large-scale construction projects can also be carried out. By analyzing the importance and correlation of cost influencing factors of key monitoring points, relying on the good performance of neural network nonlinear simulation, the deviation prediction of the next node is completed by changing the input of influencing factors.

[0098] Step 10: Determine the apportioned expenses: Based on the cost deviation forecast results, at the end of the investment period, the public expenses incurred by each project in the current period when using the head office service platform are allocated to the project as the cost of each project based on the standard principle.

[0099] Investment sharing is based on the principle of "matching benefits with costs" to ensure fairness and rationality, while striving to maximize investment benefits. The principles of sharing standards include:

[0100] (1) First, determine the dedicated investment for each function, and then apportion the shared investment after excluding the dedicated investment. Therefore, the investment apportioned to each function is the sum of the dedicated investment and the shared investment.

[0101] (2) When a project facility serves only a few individual functions, the total cost can be allocated by treating those functions as a whole and then allocating the investment according to the beneficiaries.

[0102] (3) The investment cost allocated to each function should not be greater than the investment of its alternative plan, and should be less than the benefit it brings. The benefit shall be the present value of the benefit of each function in the national economic evaluation.

[0103] (4) If certain functions are damaged after the completion of the project, the costs required to restore the original effects through remediation shall be borne by the benefiting functions that caused the damage.

[0104] Cost sharing methods include:

[0105] 1) Index allocation coefficient method: Certain design indicators of the construction project are used as the basis for calculating the allocation coefficient according to each function.

[0106] First, apportion the water consumption between flood control and benefit-producing departments according to reservoir capacity, and then apportion the water consumption between benefit-producing departments. It can also be apportioned by reservoir capacity. When allocating by reservoir capacity, it is necessary to clarify the requirements of each department for reservoir capacity, and also pay attention to the use of reservoir capacity by each department. Different considerations and treatment methods will also result in different apportionment results. This method can clearly reflect the water use characteristics of each function.

[0107] 2) Best equivalent alternative allocation method: used for comprehensive water conservancy projects with hydropower generation tasks.

[0108] The best equivalent alternative project refers to a project that is built separately with a certain benefit project indicator of a multi-objective development water conservancy project, which can replace the benefit project and is economically optimal. This method does not share the cost of the common project, but shares all costs in proportion. This method is often used in situations where the economic benefits of certain benefiting departments are difficult to calculate, as a reference for the calculation results of other cost sharing.

[0109] 3) Allocation according to the proportion of the present value of benefits obtained by each function: Since the basic purpose of building a comprehensive utilization hydraulic project is to achieve the expected economic benefits, it is reasonable to allocate the benefits according to the proportion of the present value of benefits obtained by each function, but there are also some contradictions. For example, when the project has both industrial and agricultural water use, the industry will share more costs because of the greater benefits, causing the analysis results to deviate from the actual situation.

[0110] SCRB method: also known as the separable cost-residual benefit allocation method, it regards cost savings as residual benefits. First, separate the dedicated investment of each function from the total project, and then allocate the common project investment according to the residual benefits of each function. The sum of the dedicated investment and the shared investment after the residual benefit is the investment cost amount of the benefiting function. This method is rarely used in my country because it has a large workload and requires a large amount of data.

[0111] Primary and secondary status allocation method: This method can be used when the hub's mission status is obvious and the benefits of the main function account for a large proportion. This method is relatively simple, but it has great limitations because many projects have clear primary and secondary status, but the secondary function does not bear the common engineering investment.

[0112] Based on the method of the present invention, the embodiment of the present disclosure also provides a prediction system corresponding to the above method, which includes: a data acquisition module, which is used to obtain the deviation value between the planned value and the actual value of each project cost during the project construction process; an alarm level setting module, which is used to use the K-means algorithm to set different alarm levels according to different deviation values; using the K-means algorithm to set different alarm levels according to different deviation values; including: selecting multiple monitoring nodes during the project construction process, calculating the deviation value N1 between the actual value and the planned value under the multiple monitoring nodes, and obtaining the deviation ratio Yn according to the deviation value N1; dividing the deviation value N1 into multiple threshold intervals representing multiple alarms according to the deviation ratio Yn, and obtaining a preliminary warning threshold interval; calculating the deviation value N2 between the actual value and the planned value under the remaining monitoring nodes, calculating the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval, classifying the deviation value N2 into the threshold interval with the smallest Euclidean distance thereto, and obtaining a new warning threshold interval after classification; in the new warning threshold interval, each warning threshold interval corresponds to an alarm level. The calculation module is used to calculate the project common expenses and allocate expenses among departments based on the project common expenses when the deviation value does not exceed the warning level.

[0113] Based on the same inventive concept as the above disclosed content, an embodiment of the present disclosure also provides a prediction device corresponding to the above method, which includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0114] It should be noted that the electrical connection between the above-mentioned units does not necessarily mean the connection between lines. An indirect connection method can be applied to the embodiments of the present disclosure as long as the purpose of the present disclosure is achieved.

[0115] Based on the same inventive concept, the present disclosure also provides a computer storage medium on which executable instructions are stored. When the instructions are executed by a processor, the processor performs the above method.

[0116] Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A project cost allocation method, characterized in that: The method comprises: During the project construction process, obtain the deviation between the planned value and the actual value of each project cost; The K-means algorithm is used to set different alarm levels according to different deviation values; wherein, setting different alarm levels includes: During the project construction process, multiple monitoring nodes are selected, the deviation value N1 between the actual value and the planned value under the multiple monitoring nodes is calculated, and the deviation ratio Yn is obtained according to the deviation value N1; the deviation value N1 is divided into multiple threshold intervals representing multiple alarm conditions according to the deviation ratio Yn, and the preliminary warning threshold interval is obtained; Calculate the deviation value N2 between the actual value and the planned value under the remaining monitoring nodes, calculate the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval, classify the deviation value N2 into the threshold interval with the smallest Euclidean distance, and obtain the new warning threshold interval after classification; In the new warning threshold interval, each warning threshold interval corresponds to an alarm level; When the deviation value does not exceed the warning level, the project common expenses are calculated and the expenses are shared among departments based on the project common expenses.

2. A project cost sharing method according to claim 1, characterized in that: The determination of the project common weight includes: using statistical analysis to evaluate the working hours required for the same workload in different projects from multiple dimensions to obtain a comparative amount as the project common weight.

3. A project cost sharing method according to claim 1, characterized in that: The method of obtaining the deviation between the planned value and the actual value of each project cost includes: Collect the cost deviation values ​​of historical projects of the same type and the cost activity status parameters of the corresponding cost nodes as training sets; The BP neural network is trained using the training set to obtain a trained cost deviation calculation model for predicting cost deviation values; The cost activity status parameters of each project are input into the cost deviation calculation model, and the cost deviation calculation model outputs the deviation value between the planned value and the actual value.

4. A project cost sharing method according to claim 1, characterized in that: In the K-means algorithm, the criterion function of the Euclidean distance is to minimize the sum of the squares of the distances from the data object to its cluster center, and its expression is: Among them, k is the number of clusters, C i is the center point of the first cluster, dist is the distance from x to C i The distance is , and x is a certain deviation value.

5. A project cost sharing method according to claim 1, characterized in that: The calculation formula for the common area percentage is: cq=sq*w Among them, percent is the percentage of common area; cq is used to indicate the correction amount of common area standard of the project; It is used to represent the sum of the common pool standard corrections of the n items contained in the corresponding common pool expense account. sq is used to represent the calculated common pool standard of the project under the associated common pool standard. w is used to represent the common pool weight of the project.

6. A project cost sharing method according to claim 1, characterized in that: Genetic algorithm is used to optimize the weights and thresholds of BP neural network during the training process.

7. A project cost sharing method according to claim 1, characterized in that: The common expenses of the calculation items include: At the end of the project investment period, for each common expense item, the total amount of the corresponding item expense is multiplied by the common percentage of each project under the common expense item to obtain the common expense of each project under the common expense item for each day. The common expense of each project for each day is summed up to obtain the total common expense of each project during the investment period.

8. A project cost sharing system, characterized in that: The system comprises: The data acquisition module is used to obtain the deviation between the planned value and the actual value of each project cost during the project construction process; The alarm level setting module is used to set different alarm levels according to different deviation values ​​using the K-means algorithm; wherein, setting different alarm levels includes: During the project construction process, multiple monitoring nodes are selected, the deviation value N1 between the actual value and the planned value under the multiple monitoring nodes is calculated, and the deviation ratio Yn is obtained according to the deviation value N1; the deviation value N1 is divided into multiple threshold intervals representing multiple alarm conditions according to the deviation ratio Yn, and the preliminary warning threshold interval is obtained; Calculate the deviation value N2 between the actual value and the planned value under the remaining monitoring nodes, calculate the Euclidean distance between the deviation value N2 and the interval representative value in the preliminary warning threshold interval, classify the deviation value N2 into the threshold interval with the smallest Euclidean distance, and obtain the new warning threshold interval after classification; In the new warning threshold interval, each warning threshold interval corresponds to an alarm level; The calculation module is used to calculate the project common expenses and allocate expenses among departments based on the project common expenses when the deviation value does not exceed the warning level.

9. An electronic device, comprising: at least one processor; and a memory in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, characterized in that: The instructions are executed by the at least one processor to enable the at least one processor to perform the project cost allocation method as described in any one of claims 1 to 7.

10. A computer storage medium having executable instructions stored thereon, characterized in that: When the instructions are executed by a processor, the processor implements the project cost allocation method according to any one of claims 1 to 7.

Images