Rapid settlement method and system based on NLP technology

Abstract

The invention relates to a rapid settlement method and system based on an NLP technology, and the method comprises the steps: obtaining maintenance data in a ship maintenance work order input module, and employing the NLP technology to recognize and extract a maintenance object and a pricing field needed by the settlement of the maintenance object; detecting the integrity of the pricing field, and judging whether a settlement requirement is met or not according to the integrity; if the requirement is met, price calculation is carried out through a settlement formula, and a settlement result is displayed on the terminal, the ship repair price can be rapidly and accurately calculated, and rapid recalculation by changing pricing elements in the midway is supported.

Description

Technical Field

[0001] The present invention relates to a rapid settlement method and system based on NLP technology. Background Art

[0002] The shipbuilding industry is a typical large-scale discrete manufacturing industry. Ship repair is characterized by "short engineering cycle, flat management, and fast response speed". During the repair, there is a large amount of cross-coordination among the hull, electromechanical, painting, and service professions, involving a large number of physical quantity items. In addition, due to the nature of the repair, there are a large number of repair change items in the process, which will generate many corresponding changes in account additions and subtractions.

[0003] The current external settlement method is basically offline settlement. When settling, you need to look through the price manual, manually enter the base price, coefficient, etc., and calculate through Excel or calculator. In most cases, the amount of settlement object data is large, ranging from hundreds to thousands, which is prone to human errors, and there is no traceable record, and it is impossible to verify the correctness of its calculation. The manual calculation process is time-consuming and labor-intensive, and cannot adapt to the fast-paced requirements of ship repair. More importantly, due to the lack of effective control of the settlement process, it is difficult for shipowners to approve the settlement price, and the phenomenon of unsettled accounts is relatively common. Most ship repair companies find it difficult to clear ship accounts. Take a ship repair company with an output value of more than 3 billion as an example. Because there are some settlement disputes between it and the shipowner, there are nearly 500 million accounts receivable. This has also had a great impact on the benefits of ship repair companies. In addition, offline costs can only be attributed to a single ship, and it is impossible to conduct statistical analysis on the costs incurred by various projects to support corporate decision-making.

[0004] Therefore, there is a need for a fast settlement method and system based on NLP technology, which can automatically settle the ship repair price, compare it with the manual price, and modify it midway according to the situation. Summary of the invention

[0005] The purpose of the present invention is to solve the above deficiencies in the prior art and provide a fast settlement method and system based on NLP technology. One purpose of the present invention is to provide a fast settlement method and system based on NLP technology that can quickly and accurately calculate the ship repair price. Another purpose of the present invention is to provide a fast settlement method and system based on NLP technology that can change the pricing elements midway for fast recalculation. The following is a specific solution:

[0006] On the one hand, a fast settlement method based on NLP technology includes the following steps: obtaining work order maintenance data in a ship maintenance work order entry module; identifying and extracting information from the work order maintenance data based on NLP technology, wherein the identifying information includes identifying the maintenance object in the work order maintenance data, and the extracting information includes extracting the pricing fields required for the settlement of the maintenance object; detecting the completeness of the extracted pricing fields, and determining whether the settlement completeness requirements are met; if the pricing fields are complete, settling the pricing fields using the corresponding settlement formula; and displaying the settled price and the corresponding settlement formula on the terminal.

[0007] Furthermore, obtaining the work order maintenance data in the ship maintenance work order entry module includes: obtaining the work order maintenance data in the ship maintenance work order entry module through text input by a user or through voice-to-text input.

[0008] Furthermore, the pricing fields required for the settlement of this maintenance object include: extracting the specifications, construction type, ship area location, purpose and treatment type of the maintenance object, and calculating based on the following formula:

[0009] Among them, S spec represents the specification parameter, σ i is the strength coefficient of the i-th material, V i is the volume of the material, L i is the linearity of the specification size, μ i is the weighting factor under the i-th specification, and α, β, and γ are adjustment indexes.

[0010] Furthermore, the completeness of the extracted pricing fields is checked to determine whether the settlement completeness requirements are met, including: identifying whether the pricing fields of the repair object are complete, and if so, settling the settlement price of the repair object; if it is identified that the pricing fields are incomplete, an alarm will be issued and a prompt will be given for the missing necessary parameters, prompting the repair information to be completed again.

[0011] Furthermore, if the pricing field is complete, the pricing field is settled by a corresponding settlement formula and the calculated price and the corresponding formula are displayed by the terminal, and the settlement formula is based on the following formula: Among them, P repair is the final repair settlement price, C i is the basic cost of the ith pricing field, F(S i , L i , T i ) is a function based on specification, regional location and treatment type, α and β are adjustment coefficients, is the change rate of the material volume gradient of the ith item, γ jis the jth weight factor, G(D j , Z j ) is an adjustment function based on the repair time period and usage environment.

[0012] Furthermore, after obtaining the work order maintenance data in the ship maintenance work order entry module, the steps include: encoding the content therein according to the industry price manual, and classifying it according to the catalog, configuring the field of the engineering material quantity under each catalog according to the shipyard pricing standard, and saving it to the price library after the configuration is completed. The encoding is calculated based on the following formula: Among them, C encode Indicates the encoded content, H(ω i ) is the entropy function of the ith content item, E i is the normalization factor of the content item, σ i is the weight coefficient of the content item, η j is the correlation coefficient of the items in the jth directory, ΔS j is the difference between catalog categories, ψ j The level adjustment factor of the directory item.

[0013] Furthermore, the fields in the price library can be modified by the user.

[0014] On the other hand, a rapid settlement system based on NLP technology includes a settlement module, a ship maintenance work order entry module, a terminal, and a price library, wherein the settlement module is used to obtain the work order maintenance data in the ship maintenance work order entry module, and the settlement module is used to identify and extract information from the work order maintenance data based on NLP technology; the settlement module is used to detect the completeness of the extracted pricing fields and determine whether the settlement completeness requirements are met; the settlement module is used to settle the pricing fields through the corresponding settlement formula if the pricing fields are complete; the settlement module is used to display the settled price and the corresponding settlement formula on the terminal; the terminal, the ship maintenance work order entry module and the price library are all used to implement a rapid settlement method based on NLP technology as described in any one of claims 1-7.

[0015] Beneficial effects: The rapid settlement method and system based on NLP technology provided by the present invention can significantly improve settlement efficiency; by automatically processing the text information in the maintenance work order, the present invention can quickly identify and extract the necessary pricing fields, eliminating the tedious steps of manual identification and input, and greatly shortening the settlement time; when processing a large number of maintenance work orders, this efficiency improvement is particularly obvious, which helps enterprises reduce labor costs;

[0016] The present invention ensures the settlement speed and accuracy at the same time. By accurately extracting key information such as the specifications of the maintenance object, the construction type, the ship area location, the purpose and the treatment type, and combining a perfect pricing field detection mechanism, the present invention can accurately calculate the maintenance price, which not only avoids human errors, but also ensures the reliability of the settlement results.

[0017] The present invention has the ability to flexibly adjust pricing elements; during the repair process, pricing elements may change, and the present invention can quickly respond to these changes, perform rapid recalculations, ensure real-time updates of settlement prices, and adapt to various changes during the repair process, thus bringing higher flexibility to the ship repair industry;

[0018] In general, the settlement process of the present invention is automatically operated on a computer, with complete and traceable records; it saves manpower, significantly improves settlement efficiency, and realizes rapid settlement; it avoids human calculation errors, and also avoids problems such as settlement errors caused by loss of records of offline change items; the records are clear, making it easy to obtain the shipowner's approval of the settlement price and realize ship settlement; costs can be attributed to specific engineering categories, and the accumulated structured data can be used by managers to analyze the return on investment of various types of repair engineering items. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of a rapid settlement method based on NLP technology;

[0020] Figure 2 It is a structural diagram of a rapid settlement system based on NLP technology. DETAILED DESCRIPTION

[0021] In order to deepen the understanding of the present invention, the present invention will be further described in detail below in conjunction with embodiments and drawings. The embodiments are only used to explain the present invention and do not constitute a limitation on the protection scope of the present invention.

[0022] In this embodiment:

[0023] Please see Figure 1-Figure 2 :A rapid settlement method based on NLP technology, comprising the following steps: obtaining work order maintenance data in a ship maintenance work order entry module; identifying and extracting information from the work order maintenance data based on NLP technology, wherein the identifying information includes identifying the maintenance object in the work order maintenance data, and the extracting information includes extracting the pricing fields required for the settlement of this maintenance object; detecting the completeness of the extracted pricing fields, and determining whether the settlement completeness requirements are met; if the pricing fields are complete, settling the pricing fields using the corresponding settlement formula; and displaying the settled price and the corresponding settlement formula on the terminal.

[0024] Acquiring the work order maintenance data in the ship maintenance work order entry module includes: acquiring the work order maintenance data in the ship maintenance work order entry module through text input by the user or through voice-to-text input; extracting the pricing fields required for the settlement of this maintenance object includes: extracting the specifications, construction type, ship area location, purpose and processing type of the maintenance object, and calculating based on the following formula:

[0025] Among them, S spec represents the specification parameter, σ i is the strength coefficient of the i-th material, V i is the volume of the material, L i is the linearity of the specification size, μ i is the weighting factor under the i-th specification, α, β, γ are adjustment indexes; the completeness of the extracted pricing fields is detected, and whether the settlement completeness requirements are met is determined, including: identifying whether the pricing fields of the repair object are complete, and if complete, settling the settlement price of the repair object; if it is identified that the pricing fields are incomplete, an alarm will be issued and the necessary parameters that are missing will be prompted to re-complete the repair information; if the pricing fields are complete, the pricing fields will be settled through the corresponding settlement formula, and the following steps are also included: the calculated price and the corresponding formula are displayed through the terminal; the settlement formula is based on the following formula:

[0026] Among them, P repair is the final repair settlement price, C i is the basic cost of the ith pricing field, F(S i , L i , T i ) is a function based on specification, regional location and treatment type, α and β are adjustment coefficients, is the change rate of the material volume gradient of the ith item, γ j is the jth weight factor, G(D j , Z j ) is an adjustment function based on the repair time period and the use environment. After obtaining the work order maintenance data in the ship maintenance work order entry module, the steps also include: encoding the content according to the industry price manual, and classifying it according to the catalog, configuring the field of the engineering quantity under each catalog according to the shipyard pricing standard, and saving it to the price library after the configuration is completed; the encoding is calculated based on the following formula:

[0027] Among them, C encode Indicates the encoded content, H(ω i ) is the entropy function of the ith content item, E i is the normalization factor of the content item, σ i is the weight coefficient of the content item, ηj is the correlation coefficient of the items in the jth directory, ΔS j is the difference between catalog categories, ψ j is the hierarchical adjustment coefficient of the catalog item; the fields in the price library can be modified by the user; it also includes a settlement module, a ship maintenance work order entry module, a terminal, and a price library. The settlement module is used to obtain the work order maintenance data in the ship maintenance work order entry module. The settlement module is used to identify and extract information from the work order maintenance data based on NLP technology; the settlement module is used to detect the completeness of the extracted pricing fields and determine whether the settlement completeness requirements are met; the settlement module is used to settle the pricing fields through the corresponding settlement formula if the pricing fields are complete; the settlement module is used to display the settled price and the corresponding settlement formula on the terminal; the terminal, the ship maintenance work order entry module and the price library are all used for a fast settlement method based on NLP technology.

[0028] Here is how it works:

[0029] Step 1: Obtain the work order maintenance data in the ship maintenance work order entry module;

[0030] Step 2: Identify and extract information from the work order maintenance data based on NLP technology, where the identification information includes identifying the maintenance object in the work order maintenance data, and the extraction information includes extracting the pricing fields required for the settlement of this maintenance object; for example, when the maintenance object is a steel pipe, it is necessary to extract the steel pipe diameter, construction type (replacement, disassembly, cleaning, etc.), the location of the ship where the maintenance object is located (engine room, cargo hold, side tank, etc.), purpose (hydraulic pipe, high-pressure pipe, heating pipe, sanitary pipe, etc.), treatment type (inside pipe cleaning, etc.), etc. Washing, pickling, etc.); in the pricing field, the steel pipe diameter belongs to the base price factor for the settlement of the steel pipe maintenance object, and the others are called coefficients; for example, the steel pipe diameter is 0-25mm, the steel pipe base price is 21 yuan, the casing base price is 16.8 yuan, the flange base price is 6 yuan, and the elbow base price is 8 yuan; the diameter is 26-40mm, the steel pipe base price is 30 yuan, the casing base price is 24 yuan, the flange base price is 9 yuan, and the elbow base price is 11 yuan, and so on. The range in the schematic diagram lists the price corresponding to the diameter of 800mm. The coefficients include construction coefficient, location coefficient, usage coefficient, material coefficient, treatment coefficient, etc. Among the construction coefficients, the renewal coefficient is 1, that is, the price is the base price multiplied by 1, the disassembly coefficient is 0.8, that is, the price is the base price multiplied by 0.8, and the same applies to the following, and the cleaning coefficient is 0.6; among the location coefficients, the coefficients of the engine room, cargo hold, and side tank are 1.25, and the double bottom, pipe tunnel, and bow tank are 1.3; among the usage coefficients, the hydraulic pipes and high-pressure pipes are 1.5, the heating pipes are 1.3, the sanitary pipes are 1.35, the cable pipes are 1.1, etc.; among the material coefficients, the galvanized pipes are 1.25, the stainless steel pipes are 3, etc.; among the treatment coefficients, the pipe cleaning is 1.35, the pickling is 1.15, etc.; the pricing quantity is the number of meters or weight, which varies from shipyard to shipyard;

[0031] Step 3: Check the completeness of the extracted pricing fields and determine whether they meet the settlement completeness requirements; the specific verification process is: first identify the repair object, and then identify whether the pricing fields involved in the price database of the repair object are complete. If they are complete, it will enter the settlement module to settle the price of the repair item; if it is identified that the parameters are incomplete, an alarm will be issued to prompt the lack of necessary parameters, requiring the repair personnel to complete the repair information again;

[0032] An example of the pricing field integrity verification process, for example, the input of the hull repair content is: 20 steel pipes with a diameter of DN15 in the engine room area are replaced, and the length of the pipes is 2.5 meters. First, identify the repair object as a steel pipe, and then perform corresponding identification and extraction based on its pricing parameters in the price library. For example, the pricing parameters under the steel pipe include wall thickness, diameter, length, quantity, material, construction type, location, and purpose. Then, "Replace 20 steel pipes with a diameter of DN15 in the engine room area, and the length of the pipes is 2.5 meters." This sentence includes the diameter (DN15), length (2.5), quantity (20), location (engine room), construction type (replacement), and lacks wall thickness, material, and purpose. At this time, the system will issue an alarm prompt, and the price cannot be calculated, requiring the repair personnel to re-fill the missing repair information;

[0033] Step 4: If the pricing field is complete, the pricing field will be settled through the corresponding settlement formula; Specific implementation process: First, a ship repair work order platform was developed, on which multiple price libraries were configured. Generally, shipyards will configure multiple price libraries according to different types of customers, such as high-quality customers or ordinary customers, domestic customers or foreign customers, etc. Before each ship is repaired, it will be confirmed with the shipyard in advance which price library to use.

[0034] The implementation process of the price library is as follows: (1) Encode the industry price manual, classify it by catalog, and assign a code to each project; (2) Configure the fields of the project quantities under each catalog according to the shipyard pricing standards; the configuration rules are as follows: configure the corresponding specific base price and coefficient for the different pricing parameters involved in each quantity, and other influencing factors will also be considered in the total price formula calculation, such as high temperature increase, discount, etc. The base price, coefficient, quantity, increase, discount, etc. are collectively referred to as settlement factors. The so-called settlement factors are understood as parameters that affect the calculation of the final settlement amount; when setting, all factors that may affect the settlement amount need to be considered. Parameters, such as material, specification, quantity, operation method, operation difficulty coefficient, customer factors, discount coefficient (if it is a high-quality customer), weather factors (high temperature / cold weather), etc. The above example is only for the price library configuration rules of the steel pipe. There are hundreds of parameters involved in the quantity pricing of various ship repair professions including hull, electromechanical, painting, services, etc., so it is very convenient to settle quickly after the price library is configured; in addition, if the shipyard wants to modify the base price or coefficient value in the configured price library, it can modify the value in the relevant price library. After the modification, it will be updated and released, and the ships being repaired and repaired later will be settled according to the upgraded price library standards;

[0035] Step 5: The settled price and the corresponding settlement formula are displayed on the terminal; the settlement object is broad, and can be the total price after the final repair is completed, or the price of any repair quantity, any profession or any repair part / stage. It allows the management and management to view the repair expenses and costs as needed at any stage. These data are also convenient for enterprises to conduct statistical analysis on the costs incurred by various projects to support corporate decision-making; in addition, the shipowner representative will track the ship repair process throughout, so it is also convenient for him to view the repair amount and other data at any time, which is conducive to improving the shipowner representative's trust in the final settlement quotation.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A fast settlement method based on NLP technology, characterized in that: The following steps are involved: Obtain the work order maintenance data in the ship maintenance work order entry module; identify and extract information from the work order maintenance data based on NLP technology, wherein the identification information includes identifying the maintenance object in the work order maintenance data, and the extraction information includes extracting the pricing field required for the settlement of the maintenance object; detect the completeness of the extracted pricing field, and determine whether it meets the settlement completeness requirements; If the pricing field is complete, the pricing field will be settled using the corresponding settlement formula; the settled price and the corresponding settlement formula will be displayed on the terminal.

2. According to claim 1, a rapid settlement method based on NLP technology is characterized in that: The obtaining of the work order maintenance data in the ship maintenance work order entry module comprises: obtaining the work order maintenance data in the ship maintenance work order entry module through text input by a user or through voice-to-text input.

3. According to claim 1, a rapid settlement method based on NLP technology is characterized in that: The pricing fields required for the settlement of the maintenance object include: extracting the specifications, construction type, ship area location, purpose and treatment type of the maintenance object, and calculating based on the following formula: Among them, S spec represents the specification parameter, σ i is the strength coefficient of the i-th material, V i is the volume of the material, L i is the linearity of the specification size, μ i is the weighting factor under the i-th specification, and α, β, and γ are adjustment indexes.

4. According to claim 1, a rapid settlement method based on NLP technology is characterized in that: The detection of the completeness of the extracted pricing fields and the determination of whether the settlement completeness requirements are met include: identifying whether the pricing fields of the repair object are complete, and if so, settling the settlement price of the repair object; if it is identified that the pricing fields are incomplete, an alarm will be issued and a prompt will be given for the missing necessary parameters, prompting the user to complete the repair information again.

5. The rapid settlement method based on NLP technology according to claim 1, characterized in that: If the pricing field is complete, the pricing field is settled using a corresponding settlement formula and the step of: displaying the calculated price and the corresponding formula through a terminal, wherein the settlement formula is based on the following formula: Among them, P repair is the final repair settlement price, C i is the basic cost of the ith pricing field, F(S i , L i , T i ) is a function based on specification, regional location and treatment type, α and β are adjustment coefficients, is the change rate of the material volume gradient of the ith item, γ j is the jth weight factor, G(D j , Z j ) is an adjustment function based on the repair time period and usage environment.

6. The rapid settlement method based on NLP technology according to claim 1, characterized in that: After obtaining the work order maintenance data in the ship maintenance work order entry module, the steps also include: encoding the content therein according to the industry price manual, and classifying it according to the catalog, configuring the field of the engineering material quantity under each catalog according to the shipyard pricing standard, and saving it to the price library after the configuration is completed. The encoding is calculated based on the following formula: Among them, C encode Indicates the encoded content, H(ω i ) is the entropy function of the ith content item, E i is the normalization factor of the content item, σ i is the weight coefficient of the content item, η j is the correlation coefficient of the items in the jth directory, ΔS j is the difference between catalog categories, ψ j The level adjustment factor of the directory item.

7. A rapid settlement method based on NLP technology according to claim 6, characterized in that: The fields in the price library can be modified by the user.

8. A fast settlement system based on NLP technology, characterized in that: It includes a settlement module, a ship maintenance work order entry module, a terminal and a price library, wherein the settlement module is used to obtain the work order maintenance data in the ship maintenance work order entry module, and the settlement module is used to identify and extract information from the work order maintenance data based on NLP technology; the settlement module is used to detect the completeness of the extracted pricing fields and determine whether the settlement completeness requirements are met; the settlement module is used to settle the pricing fields through the corresponding settlement formula if the pricing fields are complete; the settlement module is used to display the settled price and the corresponding settlement formula on the terminal; the terminal, the ship maintenance work order entry module and the price library are all used to implement a fast settlement method based on NLP technology as described in any one of claims 1-7.

Images