Disclosure of Invention
The invention aims to provide a method for early warning and managing data matching and super-approximate calculation of a metallurgical engineering project engineering quantity list so as to overcome the defects in the prior art.
In order to achieve the above purpose, the invention provides the following technical scheme:
the method for early warning and management of the matching of the project quantity list data of the metallurgical engineering project and the super-approximate calculation comprises the following steps:
s1, transmitting an approximate calculation list and a purchase list in a cost management module to a data matching module by taking a project as a unit, and storing the approximate calculation list and the purchase list in a data matching pool in the data matching module;
s2, the data matching pool carries out data matching on the approximate calculation list and the purchase list input into the data matching pool;
s3, the data matching pool carries out primary matching on the approximate calculation list and the purchasing list, if the automatic matching is successful, the approximate calculation list and the purchasing list are stored into a matching success table, if the automatic matching is not successful, an identifier is made and temporarily stored in a matching buffer area, and manual intervention and adjustment are carried out;
and S4, adjusting the approximate calculation list and the purchasing list temporarily stored in the matching buffer area, intervening and performing secondary matching on the adjusted approximate calculation list and purchasing list data until complete matching is successful, establishing a complete corresponding relation between the approximate calculation and the purchasing list, and transmitting the data into the early warning management module.
Furthermore, the cost management module comprises an approximation module and a purchasing module, wherein the approximation module comprises an approximation list unit taking items as units, and the purchasing module comprises a purchasing list unit taking items as units; the data matching module comprises a data matching pool, and the early warning management module comprises a purchase list over-approximate calculation early warning unit and a purchase list over-fee control line early warning unit; the approximate calculation module carries out approximate calculation management on the items and outputs the approximate calculation list details to the data matching module by taking the items as units.
Furthermore, the purchasing module performs purchasing management on the items, and outputs purchasing list details to the data matching module by taking the items as units.
Further, the purchase list and the approximate calculation list are divided into a building installation cost list and an equipment list respectively, and the building installation cost list and the equipment list are matched separately through a data matching pool.
Further, the data matching principle of the data matching pool is as follows:
if the list code exists, carrying out hierarchical matching by using 9 bits, 6 bits and 4 bits in sequence according to the list code;
if no list code exists, accurately matching according to the list name, the item characteristics and the unit, and then carrying out fuzzy matching;
the list code is 12 bits, the stream code is removed from the list code in 12 bits, and the stream code is the last 3 bits.
Further, the data matching pool in S1 has the following data matching relationship:
one purchasing list corresponds to one approximate calculation list;
one purchasing list corresponds to a plurality of approximate calculation lists;
multiple purchase lists correspond to multiple approximate calculation lists or a purchase list has no corresponding approximate calculation list.
Furthermore, names, units and items are required to be added in sequence corresponding to the plurality of approximate calculation lists by the plurality of purchasing lists, and finally, the plurality of pairs of lists are converted into one-to-one or one-to-many successful matching, and standardized processing is firstly performed on characters of measurement units which are different and have the same meaning.
Further, if the equipment list has no standard equipment list number, the matching mode is the same as the list processing mode without the list number in the building installation expense list;
the device inventory matching comprises the following steps:
A. matching the equipment approximate calculation list with the equipment purchase list, accurately matching according to the equipment names, and if only one equipment is matched, successfully matching;
B. if the multiple approximate calculation lists are matched, merging repeated approximate calculation lists for purchase, namely, a one-to-many relationship, and apportioning according to the engineering quantity;
C. and if the precise matching according to the equipment name has no result, performing fuzzy matching according to the equipment name.
Furthermore, matching degree scoring is carried out on the fuzzy matching firstly, and the fuzzy matching is realized through a text similarity scoring algorithm, wherein the text refers to names and projects;
the fuzzy matching comprises the following steps:
a. before calculating the score, segmenting words of the text, and arranging a dictionary base matched with the industry on a background for segmenting words;
b. and giving different weights to each word, adding the corresponding word segmentation weights to obtain the score of each word segmentation, and calculating the weights through the word frequency and the reverse document frequency.
Further, the early warning management module automatically early warns or manually triggers early warning inspection.
In the technical scheme, the method for managing the engineering quantity list data matching and the overgross calculation early warning of the metallurgical engineering project solves the problem of association loss of the purchasing engineering quantity list and the overgross calculation engineering list data through a matching algorithm, reconstructs association relation, realizes the comparative analysis of purchasing winning bid engineering quantity, unit price and the overgross calculation engineering quantity list, and achieves real-time and dynamic overgross calculation early warning and overlimit early warning, thereby achieving the aim of fine management of cost control.
Detailed Description
In order to make the technical solutions of the present invention better understood, those skilled in the art will now describe the present invention in further detail with reference to the accompanying drawings.
As shown in figure 1, the system corresponding to the method for managing the engineering quantity list data matching and the super-approximate calculation early warning of the metallurgical engineering project comprises a construction cost management module, a data matching module and an early warning management module,
the cost management module comprises an approximation module and a purchasing module,
the approximate calculation module carries out approximate calculation management on the items and outputs an approximate calculation list detail to the data matching module by taking the items as units;
the summary list may also be collected from other cost or project management systems. The fields of the rough calculation list in the metallurgical industry comprise WBS codes, physical quantity codes, list names, project characteristics, units, engineering quantities, construction expenses, equipment expenses, installation expenses, price and the like;
the purchasing module performs purchasing management on the items and outputs purchasing list details to the data matching module by taking the items as units;
the purchase list can also be collected from other procurement management systems. The fields of the purchase list comprise the fields of list codes, list names, item characteristics, units, engineering quantities, safety fee construction, equipment fees, price combination and the like, the fields are consistent with the approximate calculation list base, but the related fees are the sum of money in the purchase stage;
the system comprises an estimation module, a purchasing module and a control module, wherein the estimation module stores an estimation list taking items as units, and the purchasing module stores a purchasing list taking the items as units;
the data matching module comprises a data matching pool, the early warning management module comprises a purchase list over-approximate calculation early warning unit and a purchase list over-fee control line early warning unit, and the early warning management module can automatically early warn or manually trigger early warning check;
the invention can realize one-to-one comparative analysis of bid winning engineering quantity, unit price and estimated engineering quantity lists, automatically displays the engineering quantity and data exceeding the estimated calculation in red, and realizes real-time and dynamic early warning of the excess estimated calculation.
The data matching and super-approximate calculation early warning management method comprises the following steps:
s1, transmitting an approximate calculation list and a purchase list in a cost management module to a data matching module by taking a project as a unit, and storing the approximate calculation list and the purchase list in a data matching pool in the data matching module;
s2, the data matching pool carries out data matching on the approximate calculation list and the purchase list input into the data matching pool;
the data matching principle of the data matching pool is as follows:
if the list code exists, carrying out hierarchical matching by using 9 bits, 6 bits and 4 bits in sequence according to the list code, and firstly carrying out accurate matching;
if no list code exists, accurately matching according to the list name, the item characteristics and the unit, and then carrying out fuzzy matching;
fuzzy matching is carried out matching degree scoring firstly, and is realized through a text similarity scoring algorithm, wherein texts refer to list names and project characteristics;
the fuzzy matching comprises the following steps:
a. before calculating the score, segmenting words of the text, and arranging a dictionary base matched with the industry on a background for segmenting words;
b. giving different weights to each word, adding the weights of the corresponding participles to obtain the score of each participle, and calculating the weights through the word frequency and the reverse document frequency;
the list code is 12 bits, the stream code is removed from the list code in the 12 bits, and the stream code is the last 3 bits;
s3, the data matching pool carries out primary matching on the approximate calculation list and the purchase list, if the automatic matching is successful, the approximate calculation list and the purchase list are stored in a matching success table, if the automatic matching is not successful, an identifier is made and temporarily stored in a matching buffer area, and manual intervention and adjustment are carried out;
and S4, manually intervening and adjusting the approximate calculation list and the purchasing list which are temporarily stored in the matching buffer area, and performing secondary matching on the data of the approximate calculation list and the purchasing list after intervening and adjusting until the complete matching is successful, establishing a complete corresponding relation between the approximate calculation and the purchasing list, and transmitting the data into the early warning management module.
Further, the purchase list and the approximate calculation list are divided into a building installation cost list and an equipment list, and the building installation cost list and the equipment list are separately matched by a data matching pool;
the matching mode of the equipment list without the standard equipment list number is the same as the processing mode of the list without the standard equipment list number in the building installation expense list;
the device inventory matching comprises the following steps:
A. matching the equipment approximate calculation list with the equipment purchase list, accurately matching according to the equipment names, and if only one equipment is matched, successfully matching;
B. if the plurality of approximate calculation lists can be matched, judging that the approximate calculation lists are merged for purchase, namely the one-to-many relation is shared according to the engineering quantity;
C. and if the precise matching according to the equipment name has no result, performing fuzzy matching according to the equipment name.
Further, the data matching pool has the following data matching relationship:
one purchasing list corresponds to one approximate calculation list;
one purchasing list corresponds to a plurality of approximate calculation lists;
multiple purchase lists correspond to multiple approximate calculation lists or a purchase list has no corresponding approximate calculation list.
The multiple purchasing lists corresponding to the multiple approximate calculation lists need to sequentially increase the list names, units and project characteristics, multiple-to-multiple is finally converted into one-to-one or one-to-multiple successful matching under more accurate conditions, and standardized processing is firstly carried out on the words of measurement units which are different and have the same meaning, such as ton, T, T and 1000KG, which are processed according to ton.
Example 1
The data matching and super-approximate calculation early warning management method comprises the following steps:
s1, transmitting an approximate calculation list and a purchase list in a cost management module to a data matching module by taking a project as a unit, and storing the approximate calculation list and the purchase list in a data matching pool in the data matching module;
s2, the data matching pool carries out data matching on the approximate calculation list and the purchase list input into the data matching pool;
the data matching pool has the following data matching relationship:
one purchasing list corresponds to one approximate calculation list;
one purchasing list corresponds to a plurality of approximate calculation lists;
multiple purchasing lists correspond to multiple approximate calculation lists;
one purchase list has no corresponding approximate calculation list.
The data matching principle of the data matching pool is as follows:
if the list code exists, carrying out hierarchical matching by using 9 bits, 6 bits and 4 bits in sequence according to the list code, and firstly carrying out accurate matching;
if no list code exists, accurately matching according to the list name, the item characteristics and the unit, and then carrying out fuzzy matching;
fuzzy matching is firstly carried out matching degree grading, and is realized through a text similarity grading algorithm, wherein texts refer to list names and project characteristics;
the fuzzy matching comprises the following steps:
a. before calculating the score, segmenting words of the text, and arranging a dictionary base matched with the industry on a background for segmenting words;
b. giving different weights to each word, adding the weights of the corresponding participles to obtain the score of each participle, and calculating the weights through the word frequency and the reverse document frequency;
the list code is 12 bits, the stream code is removed from the list code in the 12 bits, and the stream code is the last 3 bits;
s3, the data matching pool carries out primary matching on the approximate calculation list and the purchasing list, if the automatic matching is successful, the approximate calculation list and the purchasing list are stored into a matching success table, if the automatic matching is not successful, an identifier is made and temporarily stored in a matching buffer area, and manual intervention and adjustment are carried out;
and S4, manually intervening and adjusting the approximate calculation list and the purchase list temporarily stored in the matching buffer area, and performing secondary matching on the data of the approximate calculation list and the purchase list after intervening and adjusting until complete matching is successful, establishing a complete corresponding relation between the approximate calculation and the purchase list, and transmitting the data into the early warning management module.
Example 2
The device inventory matching comprises the following steps:
A. matching the equipment approximate calculation list with the equipment purchase list, accurately matching according to the equipment names, and if only one equipment is matched, successfully matching;
B. if the plurality of approximate calculation lists can be matched, judging that the approximate calculation lists are merged for purchase, namely the one-to-many relation is shared according to the engineering quantity;
C. and if the precise matching according to the equipment name has no result, performing fuzzy matching according to the equipment name.
Example 3
Data matching module algorithm model
And the data matching module is used for performing data matching on the approximate calculation list and the purchase list input into the data matching pool. First-stage matching is carried out, if automatic matching is successful, the matching is stored in a matching success table, if the matching is not successful, identification is carried out and temporarily stored in a matching buffer area, manual intervention adjustment is carried out, data after the matching is successful again enters second-stage matching, until complete matching is achieved, the complete corresponding relation between the approximate calculation and the purchase list is established, and the data are transmitted to an early warning management module.
The general calculation list and the purchasing list of the construction project in the metallurgical industry are respectively detailed in a building installation expense list (building installation expense list for short) and an equipment list and separately purchased. The list of the building and installation expenses is generally prepared by using 'valuation rules of a metallurgical industry construction engineering quantity list (2013 edition)', and the specification of the list is a national industry standard and is an industry standard observed by design houses, construction units and construction units. In the pricing rule of the metallurgy list, the serial number of the standard list is 12-bit code, and the equipment list has no relevant national standard.
Based on the analysis of the actual situation of the list data, the list of the installation cost of the building and the list of the equipment are considered separately when a matching algorithm is designed.
General principle of building installation fee list matching: if the list code exists, the hierarchical matching is carried out by using 9 bits (the stream code is removed from 12 bits, and the last 3 bits are the stream code), 6 bits and 4 bits in sequence according to the list code, and the accurate matching is firstly carried out, and then the fuzzy matching is carried out. If no list code exists, the list is matched accurately according to the list name, the item characteristics and the unit, and then fuzzy matching is carried out. The exact match according to the list name, item feature and unit means that the list name, item feature and unit are completely equal. The fuzzy matching means that the word segmentation scores according to the weight through a fuzzy matching algorithm, and the highest scoring pair is used as a matching item. See in particular the description of the fuzzy matching algorithm below.
When matching, matching the approximate calculation list with the purchase list, wherein logically, four matching relations exist: one procurement list corresponds to one approximate calculation list (one-to-one for short), one procurement list corresponds to a plurality of approximate calculation lists (one-to-many for short), a plurality of procurement lists correspond to a plurality of approximate calculation lists (many-to-many for short), and one procurement list does not find the corresponding approximate calculation list (no matching for short). The corresponding standard is as follows: if the first 9 digits of the numbering are completely the same, the correspondence is judged.
If the matching is one-to-one or one-to-many, the program judges that the matching is successful.
If the condition is many-to-many, the list name, unit and item characteristics are required to be added in sequence, so that the condition of more accuracy is used for finally converting the many-to-many into one-to-one or one-to-many successful matching.
The characters of the measurement units are different and have the same meaning, and standardized treatment is firstly carried out, such as treatment according to ton, T, T and 1000 KG.
The method for managing the engineering quantity list data matching and the overgross calculation early warning of the metallurgical engineering project solves the problem of the association loss of the procurement engineering quantity list and the gross calculation engineering list data through a matching algorithm, reconstructs the association relation, realizes the comparative analysis of the procurement winning engineering quantity, the unit price and the gross calculation engineering quantity list, and achieves the real-time and dynamic overgross calculation early warning and overlimit early warning, thereby achieving the aim of fine management of cost control.
While certain exemplary embodiments of the present invention have been described above by way of illustration only, it will be apparent to those of ordinary skill in the art that the described embodiments may be modified in various different ways without departing from the spirit and scope of the invention. Accordingly, the drawings and description are illustrative in nature and are not to be construed as limiting the scope of the invention.