Material processing cost data analysis method and system for engineering management

Abstract

The invention relates to the technical field of data processing, and provides a material processing cost data analysis method and system for engineering management, and the method comprises the steps: obtaining a plurality of value sets; obtaining first steady-state features of the two value sets according to the difference of the data values corresponding to different data points in the two value sets; obtaining a second steady-state feature of the value set according to the data value difference of the data points in the value set; obtaining steady-state factors of the two value sets according to the first steady-state feature and the second steady-state feature; obtaining similar classes according to the steady-state factors of the two value sets; obtaining the correlation of the value sets which are not in the similar classes from the value sets which are not in the similar classes; obtaining a correlation class based on the correlation; and compressing the steady-state factors in the similar classes and the related classes to obtain compressed data, and completing data analysis according to the compressed data. The method ensures the processing effect under run compression and highlights the feature relationship in each index data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a material processing cost data analysis method and system for engineering management. Background Art

[0002] The cost data of building material processing refers to the cost data generated by the processing and treatment of various building materials in the construction industry. These data usually include the materials, labor, equipment energy and other related cost data required from raw materials to the processing process. They are very important for the analysis and estimation of such data. They provide information about the resources and workload required in the material processing process, helping project managers, architects and other stakeholders to control costs and make decisions. And because these data can involve different types of building materials, such as wood, metal, concrete, stone, etc. The processing process may include steps such as cutting, sawing, milling, welding, spraying, polishing and assembly. Each step will take a certain amount of time and resources, so cost calculation and management are required. In addition, the collection and analysis of building material processing cost data can help evaluate the efficiency and cost-effectiveness of different processing methods, optimize the production process, and provide a reference for the budget preparation of construction projects. These data can also be used to compare the costs of different suppliers or contractors and support contract negotiations and cost control decisions. However, since building materials involve a large number of plate data, the amount of data to be analyzed and stored will be huge. Usually, data compression is used for analysis and storage of such massive data, and the commonly used run-length encoding compression method is very dependent on the data characteristics of the data set to be compressed. That is, the compression effect is ideal only for data sets with high similarity and redundancy, but the compression effect is poor for data sets with low similarity and large differences, and there is even a risk of data inflation.

[0003] Since the traditional run-length encoding compression method is overly dependent on the data characteristics in the data set, and since the data indicators in the construction material processing cost scenario are highly diverse and have large differences in magnitude, the effect of directly performing run-length compression on each indicator of the processing cost data is not stable; at the same time, it is not convenient for analyzing and evaluating the data. Summary of the invention

[0004] The present invention provides a material processing cost data analysis method for engineering management to solve the problem that the run-length compression effect is not stable. The technical solution adopted is as follows: In a first aspect, an embodiment of the present invention provides a material processing cost data analysis method for engineering management, the method comprising the following steps: Get several value sets; Obtain the first steady-state feature of any two value sets according to the difference in data values ​​corresponding to different data points in any two value sets; obtain the second steady-state feature of each value set according to the difference between the data value of each data point in each value set and the data values ​​of the remaining data points; obtain the steady-state factors of any two value sets according to the first steady-state feature of any two value sets and the second steady-state features of the two value sets respectively; Obtain similar classes based on the steady-state factors between all value sets; calculate the correlation coefficient of the value sets that are not in the similar class, and obtain the correlation between any two value sets that are not in the similar class based on the correlation coefficient and entropy of the value sets; obtain the related class based on the correlation between all value sets that are not in the similar class; The stable factors in similar and related classes are compressed to obtain compressed data and store them.

[0005] Preferably, the method for obtaining several value sets is: Statistics are collected for the processing of each building material, and each expense involved in each building material is counted as a cost set of the building material, where the cost set includes the cost items of all building materials. If the cost items of one building material do not include all the cost items, the cost items not included in the building material are marked as 0, and the cost items in each cost set are normalized to obtain the value set of each building material.

[0006] Preferably, the method for obtaining the first steady-state characteristics of any two value sets according to the difference in data values ​​corresponding to different data points in any two value sets is: Record any value set as the current compression set, record any set except the current compression set as the matching set, obtain the maximum data value and the minimum data value in the current compression set and the matching set, and obtain the first steady-state characteristics of the current compression set and the matching set based on the difference between each data value in the current compression set and the matching set and the ratio of the minimum and maximum values ​​of the current compression set to the minimum and maximum values ​​of the matching set.

[0007] Preferably, the method for obtaining the first steady-state characteristics of the current compression set and the matching set according to the difference between each data value in the current compression set and the matching set and the ratio of the minimum value to the maximum value of the current compression set to the minimum value to the maximum value of the matching set is: Subtract the data value of the data point at each corresponding position of the current compression set from that of the matching set, calculate the absolute value of the difference, and accumulate the absolute values ​​of the difference to obtain the first parameter; The reciprocal of the difference between the ratio of the minimum data value to the maximum data value of the current compression set and the ratio of the minimum data value to the maximum data value of the matching set is taken, and the absolute value of the reciprocal is recorded as the second parameter; The first parameter and the second parameter are weightedly summed to obtain the first steady-state characteristics of the current compression set and the matching set.

[0008] Preferably, the method for obtaining the second steady-state feature of each value set by subtracting the data value of each data point in each value set from the data values ​​of the remaining data points is: In the formula, represents the data value of the i-th data point in the value set S, represents the data value of the jth data point in the value set S, n represents the number of data points in the value set, represents an exponential function with a natural constant as base, Represents the second steady-state characteristics of the value set S.

[0009] Preferably, the method for obtaining the steady-state factors of any two value sets according to the first steady-state characteristics of the any two value sets and the second steady-state characteristics of the two value sets respectively is: The first steady-state feature of the current compression set and the matching set is weighted with the second steady-state feature of the current compression set to obtain a first weight, the first steady-state feature of the current compression set and the matching set is weighted with the second steady-state feature of the matching set to obtain a second weight, and the maximum value of the first weight and the second weight is used as the steady-state factor of the current compression set and the matching set.

[0010] Preferably, the method for obtaining similarity classes according to the steady-state factors between all value sets is: For the current compression set, obtain the steady-state factors of the current compression set and all other value sets, put the value sets with steady-state factors greater than the preset threshold and the current compression set in a class, calculate the steady-state factors of each value set and all other value sets in the class, put the value sets with steady-state factors greater than the preset threshold into the class, until all value sets are calculated, and then save the value sets with a preset number of steady-state factors greater than the preset threshold in the class, and record the saved classes as similar classes.

[0011] Preferably, the method of calculating the correlation coefficient of the value sets that are not in the similar class and obtaining the correlation between any two value sets that are not in the similar class according to the correlation coefficient and entropy of the value sets is: Any set that is not in the similar class is recorded as the first set, and any value set that is not in the similar class except the first set is recorded as the second set. The Pearson correlation coefficient and the Spearman correlation coefficient are calculated for the first set and the second set, and the largest one of the Pearson correlation coefficient and the Spearman correlation coefficient is taken as the correlation coefficient between the first set and the second set. Calculate the difference between the entropy values ​​of the first set and the second set to obtain the difference entropy between the first set and the second set; The correlation coefficient and the difference entropy between the first set and the second set are weighted to obtain the correlation between the first set and the second set.

[0012] Preferably, the method of compressing the steady-state factors in the similarity class and the related class to obtain compressed data and store the compressed data is: For the value sets in similar classes, run-length encoding is directly used for compression processing. For the value sets in related classes, one of the value sets is retained, and the correlation coefficients between the remaining value sets in the related classes and the value set are used as compression values. The retained value sets and compression values ​​are used as compression results of the related classes.

[0013] In the second aspect, an embodiment of the present invention also provides a material processing cost data analysis system for engineering management, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0014] The beneficial effects of the present invention are as follows: the present invention proposes a material processing cost data analysis method and system for engineering management; by analyzing and calculating the redundant similarity of each material processing index data, a highly correlated index data set is obtained and classified and personalized compression processing is performed; data sets with correlation or similarity are integrated and synchronously compressed, ensuring the processing effect under run-length compression while facilitating highlighting the characteristic relationship among each index data, which is helpful for the final cost control decision. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention 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 only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 A schematic flow chart of a material processing cost data analysis method for engineering management provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 , which shows a material processing cost data analysis method for engineering management provided by an embodiment of the present invention, the method comprising the following steps: Step S001, obtaining several value sets.

[0019] The construction material processing cost data contains multiple sections and data indicators: for example, it involves material costs; it covers concrete, bricks, steel bars, wood, glass, paint, etc. The cost of labor, equipment and other resources involved in processing construction materials. This includes processing processes such as cutting, grinding, spraying, painting, laying, and subsequent transportation cost indicators, installation cost indicators, technical service cost indicators, etc. For each different building material, multiple indicator data are designed, that is, for a building material, an indicator data set is obtained. If a material does not include the indicators involved, then the item is set to 0. For example, there is no cutting cost for paint materials, so this item is 0 in the indicator data set corresponding to paint.

[0020] Usually, the cost of each building material has a maximum and minimum cost value according to its different types. In order to reduce the impact of the difference in magnitude between data and facilitate better data compression storage and feature analysis, the data values ​​in the indicator data set corresponding to each material are normalized. The formula is as follows: In the formula, represents the data value of the i-th data point in building material A, Indicates the smallest data value in building material A. Indicates the maximum data value in building material A. Represents the normalized data value of the i-th data point in building material A.

[0021] The data values ​​in the indicator data set corresponding to each building material are replaced with the normalized values ​​to obtain the value set of each building material.

[0022] At this point, the value set of each building material is obtained.

[0023] Step S002, obtaining the first steady-state characteristic of any two value sets based on the difference between the data values ​​corresponding to different data points in any two value sets; obtaining the second steady-state characteristic of each value set by subtracting the data value of each data point in each value set from the data values ​​of the remaining data points; obtaining the steady-state factors of any two value sets based on the first steady-state characteristic of any two value sets and the second steady-state characteristics of the two value sets respectively.

[0024] Since different value sets may have the same distribution characteristics, similar or highly correlated material indicator data sets are obtained from the diverse material cost data indicators for classification and compression.

[0025] For run-length coding compression, if the similarity and redundancy of the data values ​​within the value set are high, it will have a positive impact on the final compression effect. However, for the huge material cost index, there are certain differences in the internal characteristic distribution of each index data information, that is, the similarity and redundancy cannot be guaranteed. Therefore, this step analyzes the internal characteristics of each value set. Obtain multiple groups of index data with similar data similarity and distribution rules (that is, similar steady-state factors), classify and integrate them, and obtain the steady-state factors of the two value sets for the steady-state factor evaluation model, and classify and integrate the data sets with higher steady-state factors for compression.

[0026] Specifically, the maximum data value and the minimum data value in each value set are obtained, any value set is recorded as the current compression set, and the difference between each data value of the current compression set and each of the other value sets and the amplitude difference of the two sets are used to obtain the first steady-state characteristics of the current compression set and the value set. The formula is as follows: In the formula, represents the data value of the i-th data point in the current compressed set S, represents the data value of the i-th data point in the value set B, Indicates the minimum data value of the current compressed set S, represents the minimum data value of value set B, Indicates the maximum data value of the current compressed set S, Represents the maximum data value of value set B, represents an exponential function with a natural constant as the base, n represents the number of data points in the value set, and is the empirical weight. In this embodiment is 0.6, is 0.4, Represents the first steady-state characteristics of the current compression set S and the value set B.

[0027] Among them, for the first term, the smaller the mean of the absolute value of the difference between the internal data values ​​of the current indicator data is, the stronger the first steady-state characteristics of the current two sets are. Under the effect of , the closer the final calculation result is to 1, the stronger the first steady-state feature is. For the second term, the fraction represents the difference ratio of internal data values. The smaller the difference ratio of internal data values ​​in any pair of data sets, the stronger the first steady-state feature is. Similarly, in the inverse proportional normalization function , the second term approaches 1.

[0028] For the current compressed set, the difference between the data value of each data point and the data values ​​of the remaining data points is used to obtain the second steady-state characteristic of the current compressed set. The formula is as follows: In the formula, represents the data value of the i-th data point in the current compressed set S, represents the data value of the jth data point in the current compressed set S, n represents the number of data points in the value set, represents an exponential function with a natural constant as base, Represents the second steady-state characteristic of the current compression set S.

[0029] Among them, the absolute value represents the difference between the current data value and the value in the indicator data set; the second steady-state feature describes the mean of the difference between the values ​​of each item in the current compressed set and the mean; the smaller the value, the stronger the second steady-state feature, and the smaller the inverse proportional normalization function. , the result value tends to 1.

[0030] The steady-state factor of the current compression set and the remaining value sets is obtained according to the second steady-state characteristic of the current compression set and the first steady-state characteristics of the current compression set and the remaining value sets. The formula is as follows: In the formula, represents the first steady-state characteristics of the current compression set S and value set B, represents the second steady-state characteristic of the current compressed set S, represents the second steady-state characteristic of the value set B, represents the maximum value function, , is the weight value. In this embodiment is 0.6, is 0.4, Represents the steady-state factor between the current compression set S and the value set B.

[0031] At this point, the steady-state factors of any two value sets are obtained.

[0032] Step S003, obtaining similar classes based on the steady-state factors between all value sets; calculating the correlation coefficients of the value sets that are not in the similar classes and combining the entropy of the value sets to obtain the correlation between any two value sets that are not in the similar classes; obtaining related classes based on the correlations between all value sets that are not in the similar classes.

[0033] According to the above method, the steady-state factors of the two value sets are obtained, and a steady-state threshold is set. If the steady-state factors of the two value sets are greater than the steady-state threshold, it means that the two value sets are similar. If the steady-state factors of the two value sets are less than the steady-state threshold, it means that the two value sets are not similar. In this embodiment, the steady-state threshold is 0.7.

[0034] Calculate the steady-state factor with any value set and the remaining value sets, mark the value set with a steady-state factor greater than the steady-state threshold as similar, put it into a class, calculate the steady-state factor for each value set in the class and each of the remaining value sets, and similarly put the value set with a steady-state factor greater than the steady-state threshold into the class, until all value sets have calculated the steady-state factor with the remaining value sets. In the class, for each value set, if the value set has a steady-state factor calculated with 5 or more value sets greater than the steady-state threshold, it will be retained in the class, otherwise it will be excluded from the class, thus obtaining a similar class.

[0035] The similarity and stability of the value sets in the similar class are high, and theoretically there will be a lot of data redundancy when performing run-length encoding compression. For the value sets that are not in the similar class, their internal data characteristics are theoretically large differences in data values ​​and large differences in mean values ​​from other class indicator data sets. Although they have no similarity, different value sets may be correlated. Therefore, the correlation of the value sets that are not in the similar class is calculated to obtain the correlation coefficient of the two value sets and combine the entropy value difference of the two value sets to obtain the correlation of the two value sets. The formula is as follows: In the formula, represents the value set S, represents the value set B, represents the Pearson correlation coefficient between value set S and value set B, represents the Spearman correlation coefficient between the value set S and the value set B, It means to select the maximum value of two correlation coefficients. Represents the correlation coefficient between value set S and value set B. Represents the entropy value of all data values ​​in the value set S, Represents the entropy value of all data values ​​in the value set B, represents an exponential function with a natural constant as base, Represents the difference entropy between value set S and value set B. and Respectively represent weight values. Since the correlation coefficient has a greater impact on the correlation, in this embodiment .

[0036] Among them, since the correlation coefficients tend to 1, it indicates a strong positive correlation. Tend to -1 for a strong negative correlation; and tend to 0 for a low correlation, so the absolute values ​​are taken respectively; so that they all meet the requirement of tending to 1 for a strong correlation. Since it is sufficient to meet one of the strong correlations, the maximum value of the correlation coefficient is taken to represent the correlation coefficient between the current two indicator data sets. The entropy values ​​of the two value sets reflect the degree of chaotic discreteness of their internal data. If the difference value is smaller, it means that the distribution characteristics of the internal data of the current two indicator data sets are relatively similar, reflecting that their correlation is strong. Therefore, based on these two, the correlation of the two value sets is obtained.

[0037] A correlation threshold is set. In this embodiment, the correlation threshold is 0.7. If the correlation between two value sets is greater than the correlation threshold, it means that the two value sets are related. For all value sets that are not in the similar class, the value sets with correlation are obtained. If there are two or more value sets that are related, a related class is formed.

[0038] So far, similar classes and several related classes have been obtained.

[0039] Step S004, compress the steady-state factors in the similar class and the related class to obtain compressed data, and complete data analysis based on the compressed data and the value set that is not in the similar class and the related class.

[0040] For the value sets in similar classes, run-length encoding is directly used for compression processing, and for the value sets in related classes, one of the value sets is retained, and the correlation coefficients between the remaining value sets in the related classes and the value set are used as compression values. The retained value sets and all compressed values ​​are the compression results of the related classes; decompression is completed according to the retained value sets and compressed value sets, and the value sets that are not in the related classes and similar classes are not compressed, thereby reducing the amount of data information on the processing cost and saving storage space. The compressed data is transmitted to the data analysis module, and the compressed data in the data analysis module is decompressed to obtain the original data. So far, this embodiment has completed the efficient compression and storage of data by analyzing the data on the processing cost of building materials.

[0041] Based on the same inventive concept as the above-mentioned method embodiment, the present invention provides a material processing cost data analysis system for engineering management, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the above-mentioned computer program, when executed by the processor, implements the steps of a material processing cost data analysis method for engineering management.

[0042] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A material processing cost data analysis method for engineering management, characterized in that: The method comprises the following steps: Get several value sets; Obtain the first steady-state feature of any two value sets according to the difference in data values ​​corresponding to different data points in any two value sets; obtain the second steady-state feature of each value set according to the difference between the data value of each data point in each value set and the data values ​​of the remaining data points; obtain the steady-state factors of any two value sets according to the first steady-state feature of any two value sets and the second steady-state features of the two value sets respectively; Obtain similar classes based on the steady-state factors between all value sets; calculate the correlation coefficient of the value sets that are not in the similar class, and obtain the correlation between any two value sets that are not in the similar class based on the correlation coefficient and entropy of the value sets; obtain the related class based on the correlation between all value sets that are not in the similar class; The stable factors in similar and related classes are compressed to obtain compressed data and store them.

2. A material processing cost data analysis method for engineering management according to claim 1, characterized in that: The method for obtaining several value sets is: Statistics are collected for the processing of each building material, and each expense involved in each building material is counted as a cost set of the building material, where the cost set includes the cost items of all building materials. If the cost items of one building material do not include all the cost items, the cost items not included in the building material are marked as 0, and the cost items in each cost set are normalized to obtain the value set of each building material.

3. The material processing cost data analysis method for engineering management according to claim 1 is characterized in that: The method for obtaining the first steady-state features of any two value sets according to the difference in data values ​​corresponding to different data points in any two value sets is: Record any value set as the current compression set, record any set except the current compression set as the matching set, obtain the maximum data value and the minimum data value in the current compression set and the matching set, and obtain the first steady-state characteristics of the current compression set and the matching set based on the difference between each data value in the current compression set and the matching set and the ratio of the minimum and maximum values ​​of the current compression set to the minimum and maximum values ​​of the matching set.

4. A material processing cost data analysis method for engineering management according to claim 3, characterized in that: The method for obtaining the first steady-state characteristics of the current compression set and the matching set according to the difference between each data value in the current compression set and the matching set and the ratio of the minimum value to the maximum value of the current compression set to the minimum value to the maximum value of the matching set is: Subtract the data value of the data point at each corresponding position of the current compression set from that of the matching set, calculate the absolute value of the difference, and accumulate the absolute values ​​of the difference to obtain the first parameter; The reciprocal of the difference between the ratio of the minimum data value to the maximum data value of the current compression set and the ratio of the minimum data value to the maximum data value of the matching set is taken, and the absolute value of the reciprocal is recorded as the second parameter; The first parameter and the second parameter are weightedly summed to obtain the first steady-state characteristics of the current compression set and the matching set.

5. The material processing cost data analysis method for engineering management according to claim 1 is characterized in that: The method for obtaining the second steady-state feature of each value set by subtracting the data value of each data point in each value set from the data values ​​of the remaining data points is: In the formula, represents the data value of the i-th data point in the value set S, represents the data value of the jth data point in the value set S, n represents the number of data points in the value set, represents an exponential function with a natural constant as base, Represents the second steady-state characteristics of the value set S.

6. The material processing cost data analysis method for engineering management according to claim 3 is characterized in that: The method for obtaining the steady-state factors of any two value sets according to the first steady-state characteristics of any two value sets and the second steady-state characteristics of the two value sets respectively is: The first steady-state feature of the current compression set and the matching set is weighted with the second steady-state feature of the current compression set to obtain a first weight, the first steady-state feature of the current compression set and the matching set is weighted with the second steady-state feature of the matching set to obtain a second weight, and the maximum value of the first weight and the second weight is used as the steady-state factor of the current compression set and the matching set.

7. The material processing cost data analysis method for engineering management according to claim 3 is characterized in that: The method for obtaining similarity classes based on the steady-state factors between all value sets is: For the current compression set, obtain the steady-state factors of the current compression set and all other value sets, put the value sets with steady-state factors greater than the preset threshold and the current compression set in a class, calculate the steady-state factors of each value set and all other value sets in the class, put the value sets with steady-state factors greater than the preset threshold into the class, until all value sets are calculated, and then save the value sets with a preset number of steady-state factors greater than the preset threshold in the class, and record the saved classes as similar classes.

8. The material processing cost data analysis method for engineering management according to claim 1 is characterized in that: The method of calculating the correlation coefficient of the value sets that are not in the similar class and obtaining the correlation between any two value sets that are not in the similar class according to the correlation coefficient and entropy of the value sets is: Any set that is not in the similar class is recorded as the first set, and any value set that is not in the similar class except the first set is recorded as the second set. The Pearson correlation coefficient and the Spearman correlation coefficient are calculated for the first set and the second set, and the largest one of the Pearson correlation coefficient and the Spearman correlation coefficient is taken as the correlation coefficient between the first set and the second set. Calculate the difference between the entropy values ​​of the first set and the second set to obtain the difference entropy between the first set and the second set; The correlation coefficient and the difference entropy between the first set and the second set are weighted to obtain the correlation between the first set and the second set.

9. The material processing cost data analysis method for engineering management according to claim 1, characterized in that: The method for compressing the steady-state factors in the similarity class and the related class to obtain compressed data and store the compressed data is: For the value sets in similar classes, run-length encoding is directly used for compression processing. For the value sets in related classes, one of the value sets is retained, and the correlation coefficients between the remaining value sets in the related classes and the value set are used as compression values. The retained value sets and compression values ​​are used as compression results of the related classes.

10. A material processing cost data analysis system for engineering management, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images