A dual-path diversion method for R&D cost aggregation

Abstract

This invention discloses a dual-path diversion method for R&D expense collection, comprising the following steps: S1, acquiring basic data, which includes at least project information, sales revenue, and personnel information; S2, determining the R&D expense collection method based on whether R&D expenses are present in the basic data; if no R&D expenses are present, configuring R&D expense ratio requirements in the basic data of S1 and using a budgeted collection method for R&D expense collection; if R&D expenses are present, using a measured collection method for R&D expense collection; S3, outputting the R&D expense collection results. This invention automatically diverts data between two collection paths by first determining whether there are existing R&D expense records after data input. The dual-path design allows the same collection method to cover the entire lifecycle collection needs of an enterprise, from the budget stage to the actual occurrence stage.

Description

Technical Field

[0001] This invention belongs to the field of R&D expense management technology, specifically involving a dual-path diversion method for collecting R&D expenses. Background Technology

[0002] Research and development expense aggregation refers to the process of uniformly collecting and recording various expenses incurred by an enterprise in its research and development activities according to certain classification and accounting rules. This includes expenses related to research, development, testing, and design, so that enterprises can more clearly understand and grasp their R&D investment and provide accurate data support for future scientific research project decisions, financial reporting, and tax declaration.

[0003] Currently, most companies collect R&D expenses manually by compiling various financial data from all R&D financial documents. However, due to the large number of financial documents and the extensive text content, manual retrieval is time-consuming, labor-intensive, and prone to omissions. Therefore, some existing technologies include auxiliary accounting modules based on financial software that can record R&D expenditures in the form of project subsidiary ledgers. Meanwhile, some solutions focus on expense text recognition, knowledge graph modeling, expense classification database establishment, categorized statistics, compliance auditing, or full lifecycle management. Their core functions typically include expense data recognition, expense classification, categorized statistics, approval workflow, audit alerts, or report output. However, the above solutions and existing technologies still have the following shortcomings:

[0004] 1. When enterprises are in the early stages of R&D activities, when new projects are initiated, or when budgets are being prepared, there are no actual R&D expenses to collect. Existing solutions cannot conduct structured collection and expense allocation based on pre-set budgets. They can only be processed after expenses are actually incurred, making it difficult to compare budget and actual data, resulting in a lack of effective support for financial planning.

[0005] 2. The allocation logic among expense categories is rigid and lacks dynamic adaptability. Existing methods often use fixed proportions or simple averages when allocating total R&D expenses to specific categories (such as personnel costs, direct inputs, depreciation and amortization), failing to take into account the actual expenditure characteristics of different expense categories and the dynamic adjustment of project stages. For example, some categories have rigid expenditure characteristics, some are proportionally related to business scale, and some have flexible adjustment space. This allocation method leads to insufficient matching between the allocation results and actual consumption, and weak refined management capabilities.

[0006] Therefore, it is necessary to develop a method for collecting R&D expenses that can meet the diverse needs of enterprises. Summary of the Invention

[0007] The purpose of this invention is to provide a dual-path diversion method and system for collecting R&D expenses, which can meet the diverse R&D expense collection needs of enterprises.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] A method for allocating R&D expenses through a dual-path routing approach includes the following steps:

[0010] S1. Obtain basic data, which includes at least project information, sales revenue, and personnel information;

[0011] S2. Determine the R&D expense collection method based on whether there are R&D expenses in the basic data; if there are no R&D expenses, configure the R&D expense ratio requirement in the S1 basic data and use the budget collection method to collect R&D expenses; if there are R&D expenses, use the actual measurement collection method to collect R&D expenses.

[0012] S3, Output the results of R&D expense aggregation.

[0013] Preferably, the basic data in S1 also includes at least equipment data and material data.

[0014] Preferably, the basic data shown may also include R&D expense data, R&D expense item data, business document data, statistical rules data, or budget structure data.

[0015] Preferably, the project information includes at least the project number, project name, and project start and end period; the personnel information includes at least the name, project, and total attendance days; the equipment data includes at least the equipment name, equipment number, equipment quantity, monthly depreciation, and total equipment usage time; the material data includes at least the material name, material number, material quantity, material unit price, and material amount; and the sales revenue is monthly, quarterly, or annual revenue.

[0016] Preferably, in step S2, the specific steps of the budget aggregation method are as follows:

[0017] S201. Calculate the total R&D expenses based on the sales revenue and R&D ratio requirements;

[0018] S202. Determine the R&D expense items to be collected, and divide the R&D expense items into stable structural items, proportional constraint items, and difference balancing items;

[0019] S203. Determine the proportion of the R&D expenses of the i-th project to the total R&D expenses, and obtain the R&D expenses of the i-th project as Ai; where i ≥ 1;

[0020] S204. Based on the data in the information of the i-th project and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item of the R&D expense item in the i-th project;

[0021] Balance item amount Am = Ai - Ah - Ad;

[0022] S205. Allocate the difference balancing amount to each R&D item in the difference balancing item.

[0023] Preferably, in S203, the proportion of the R&D expenses of the i-th project to the total R&D expenses is determined by the proportion of R&D expenses of each project configured by the user or by a weighted model based on project cycle, personnel input, etc.

[0024] Preferably, in step S202, after determining the R&D expense items to be collected, the method further includes: dividing the R&D expenses into multiple categories based on the subcategories within the R&D expense items; these multiple categories include at least statistical, tax, and scientific and technological categories; correspondingly, when outputting the R&D expense collection results, the results are output according to each of the multiple statistical categories. This design allows for the simultaneous generation of expense reports that meet the requirements of different competent authorities in a single collection, avoiding the redundancy of repeatedly collecting expenses by category.

[0025] The R&D expenses referred to in the statistical scope are those that meet the requirements of the National Bureau of Statistics' "Statistical Reporting System for Enterprise R&D Activities" and "Statistical Specifications for Research and Experimental Development (R&D) Input".

[0026] The R&D expenses mentioned above refer to those that meet the requirements of the "Notice of the Ministry of Finance, the State Administration of Taxation and the Ministry of Science and Technology on Improving the Policy of Pre-tax Deduction for Research and Development Expenses" (Cai Shui

[2015] No. 119), the "Announcement of the State Administration of Taxation on Issues Concerning the Scope of Collection of Pre-tax Deduction for R&D Expenses" (No. 40 of 2017), and the "Announcement of the Ministry of Finance and the State Administration of Taxation on Further Improving the Policy of Pre-tax Deduction for R&D Expenses" (No. 7 of 2023).

[0027] The R&D expenses mentioned above refer to the R&D expenses that meet the requirements of the "Administrative Measures for the Recognition of High-tech Enterprises" (Guoke Fa Huo

[2016] No. 32) and the "Guidelines for the Administration of the Recognition of High-tech Enterprises" (Guoke Fa Huo

[2016] No. 195).

[0028] The R&D expense items include one or more of the following: personnel costs, direct input costs, depreciation and long-term amortization costs, intangible asset amortization costs, design and testing costs, other related costs, outsourced R&D costs, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; preferably, in step s202, the industry type is first determined based on basic data, and then the R&D expense items are divided into stable structure items, proportional constraint items, and difference balancing items according to the industry type; the industry type is divided into manufacturing and non-manufacturing; the industry classification is based on the national economic industry classification standard for classifying enterprises.

[0029] If the basic data analysis is for the manufacturing industry, the stable structural items are one or more of the following: personnel costs and intangible asset amortization costs; the proportional constraint items are one or more of the following: depreciation costs and long-term amortization costs, design and testing costs, outsourced research and development costs, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; the difference balancing items are one or more of the following: direct input costs, land and building expenditures, patent and proprietary technology expenditures, or other related costs.

[0030] If the basic data analysis pertains to non-manufacturing industries, the stable structural item is one or more of intangible asset amortization expenses and direct input costs; the proportional constraint item is one or more of depreciation expenses and long-term amortization expenses, design and testing expenses, outsourced research and development expenses, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; the difference balancing item is one or more of personnel labor costs or other related expenses; preferably, the specific method for allocating the difference balancing item amount to each R&D item in the difference balancing item in step 205 is as follows:

[0031] The difference balancing items are divided into primary adjustment R&D expense items and auxiliary adjustment R&D expense items;

[0032] First, the amount of the difference balancing item is allocated using the main adjustment R&D expense account; when the main adjustment R&D expense account does not meet the preset constraints, the remaining difference is transferred to the auxiliary adjustment R&D expense account for further allocation; after each round of allocation, the allocation result is checked for multi-dimensional constraints.

[0033] If the verification fails, repeat the transfer and allocation process until the error in the balance item amount meets the preset threshold or reaches the preset stop condition.

[0034] The multi-dimensional constraint verification includes at least one of the following: total amount error, project ratio deviation, inventory constraint, integer constraint, continuous precision or caliber ratio limit; through the multi-dimensional verification system, the reliability of the collection results is ensured in multiple compliance dimensions.

[0035] When the difference balancing item includes material costs, discrete materials and continuous materials are identified according to the material unit or user configuration; the discrete materials are materials with integer units (such as units or units) or indivisible materials; the continuous materials are divisible materials; discrete materials are constrained by integers, and continuous materials are compensated for according to a preset precision.

[0036] The main adjustment of R&D expense items is determined based on available amount, divisibility, constraints, proportionality limits, and completeness of evidence;

[0037] Preferably, the primary adjustment R&D expense item is the R&D expense item with the largest R&D expense amount among the difference balancing items. The technical consideration for this priority strategy is that the item with the largest available amount usually has the largest adjustment space, and prioritizing its allocation can absorb the difference to the maximum extent, reduce the number of rounds of allocation to auxiliary adjustment items, and improve allocation efficiency.

[0038] Preferably, in step S2, the specific steps of the measured aggregation method are as follows:

[0039] S211. Confirm whether there are individual project R&D expenses in the R&D expenses, and whether there are R&D expense items and item fees;

[0040] S212. If there is no R&D expense for the i-th project in the R&D expenses, and there is no R&D expense item and item fee, use steps S202-S205 in the budget collection method to collect the R&D expenses.

[0041] S213. If there is R&D expense for the i-th project in the R&D expenses, but there is no R&D expense account and account expense; first determine the R&D expense account to be collected, and determine the stable structure item, proportional constraint item and difference balancing item in the R&D account, and use steps S204-S205 in the budget collection method to allocate the R&D expenses.

[0042] S214. If there is no R&D expense for the i-th project in the R&D expenses, there are R&D expense items and item costs; determine the stable structural items, proportional constraint items and difference balancing items in the R&D items; use steps S203-S205 in the budget collection method to collect R&D expenses.

[0043] The experimental aggregation method, by judging the branch of data completeness, can automatically select the most suitable aggregation strategy for the current data conditions when some actual cost data is missing, so as to maximize the use of existing data to complete the aggregation without waiting for the data to be completely complete.

[0044] Preferably, the R&D expense collection method of the present invention further includes step S4, evidence chain output; specifically including the following steps:

[0045] S41. Extract the association identifier by combining the R&D expense collection results and basic data;

[0046] The associated identifiers include at least the project identifier, business entity identifier, resource identifier, and subject identifier;

[0047] S42. Mapping between project numbering and R&D subject structure;

[0048] In the R&D expense aggregation results, the item number project is used, and the business documents in the basic data are mapped to the corresponding item number and item number to form a mapping relationship;

[0049] S43, Generation of the chain of evidence;

[0050] Based on the mapping relationship, the business data corresponding to each type of expense are associated and processed to generate corresponding supporting documents; including at least: personnel data, equipment data, and material data.

[0051] The personnel data generation steps are as follows: based on the personnel number, project number, and time period, the attendance data and salary data are associated to generate the basis for labor cost allocation. The basis for labor cost allocation includes a working hour percentage allocation table or a labor cost allocation table.

[0052] The equipment data generation steps are as follows: based on the equipment number, project number, and time period, the equipment operation record is associated with the depreciation calculation data to generate the basis for depreciation expense allocation. The basis for depreciation expense allocation includes a depreciation calculation table or a depreciation allocation table.

[0053] The material data generation steps are as follows: based on the material number, project number, and time period, the material requisition form and the material return form are associated to generate a material collection and allocation list, which includes a material allocation detail table.

[0054] S44, Data consistency verification;

[0055] Consistency checks are performed on personnel data, equipment data, and material data respectively; the checks include consistency checks on attendance data and wage data, and consistency checks on work hour allocation and labor costs;

[0056] Consistency verification between equipment operation records and depreciation data, and consistency verification between depreciation allocation and equipment costs;

[0057] Consistency verification between inventory data and inbound / outbound documents, and consistency verification between material requisition documents and material collection results;

[0058] S45, Output the chain of evidence;

[0059] The verified supporting documents are packaged to generate an evidence chain file package; the evidence chain file package includes at least:

[0060] Supporting documents for personnel expenses;

[0061] Supporting documents for equipment costs;

[0062] Supporting documents for material costs;

[0063] List of evidence chains;

[0064] A report linking the collected results with the original business data.

[0065] This invention's method automatically splits data collection between two paths by first determining whether existing R&D expense records exist after data input. When the company is in the budgeting stage and no actual expenses have occurred, the method uses a budget collection mode by acquiring collection control parameters (such as R&D expense ratio requirements) and combining them with revenue data. This generates structured, subject-level budget collection results, enabling the company to obtain usable expense estimates early in R&D activities, supporting financial planning and project application. When actual expenses have occurred, the method uses a measured collection mode, collecting data based on actual expense records. This dual-path design allows the same collection method to cover the company's full-cycle collection needs from the budgeting stage to the actual expense stage.

[0066] This invention fundamentally changes the traditional fixed-proportion allocation method by dividing R&D expense items into three categories based on their expenditure attributes: stable structural items, proportional constraint items, and difference balancing items. Rigid expenditure items are determined based on actual project data, related expenditure items are estimated based on business indicators, and the remaining difference is systematically solved in flexible items. This hierarchical processing strategy ensures that the aggregation results accurately reflect the company's true expense structure, solving the core problems of rigid allocation logic and lack of flexibility in existing technologies.

[0067] The relay solution mechanism for primary and secondary adjustment R&D expense accounts introduced in the differential balancing allocation, combined with multi-dimensional constraint verification and iterative convergence control, systematically solves the complex constraint problem in multi-account differential allocation. This method can automatically complete precise allocation at the account level under multiple conditions, including total amount constraints, integer constraints, and proportional restrictions, significantly improving the data quality and compliance reliability of the collected results.

[0068] Meanwhile, by decoupling the subject attribute mapping rules from the industry type identifier, this invention allows the three-category architecture to flexibly adapt to the cost structure characteristics of different industries, improving the method's versatility and configurability. The multi-caliber simultaneous aggregation design enables a single calculation to meet the reporting requirements of various calibers, such as statistics, taxation, and science and technology, avoiding redundancy from repeated aggregation. The automatic generation and consistency verification of the evidence chain based on business process data constructs a complete traceability link from the aggregation results to the original business vouchers for auditing and compliance checks. Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0070] Figure 2 This is a schematic diagram of the budget collection method of the present invention;

[0071] Figure 3 This is a schematic diagram of the process for allocating the difference balancing item amount in this invention;

[0072] Figure 4 This is a schematic diagram of the measured data collection method of the present invention;

[0073] Figure 5 This is a schematic diagram illustrating the generation of the evidence chain in this invention. Detailed Implementation

[0074] The technical solution of the present invention will now be clearly and completely described in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0075] like Figure 1 The proposed dual-path cost allocation method includes the following steps:

[0076] S1. Obtain basic data, which includes at least project information, sales revenue, personnel information, equipment data, and material data; the project information includes at least project number, project name, and project start and end dates; the personnel information includes at least name, project affiliation, and total attendance days; the equipment data includes at least equipment name, equipment number, equipment quantity, monthly depreciation, and total equipment usage time; the material data includes at least material name, material number, material quantity, material unit price, and material cost; the sales revenue is monthly, quarterly, or annual revenue.

[0077] S2. Determine the R&D expense collection method based on whether there are R&D expenses in the basic data; if there are no R&D expenses, configure the R&D expense ratio requirement in the S1 basic data and use the budget collection method to collect R&D expenses; if there are R&D expenses, use the actual measurement collection method to collect R&D expenses.

[0078] S3, Output the results of R&D expense aggregation;

[0079] S4, Output of the chain of evidence;

[0080] like Figure 2 As shown, the specific steps of the budget collection method are as follows:

[0081] S201. Calculate the total R&D expenses based on the sales revenue and R&D ratio requirements;

[0082] S202. Determine the R&D expense items to be collected, determine the industry type based on basic data, and classify the R&D expense items into stable structure items, proportional constraint items, and difference balancing items according to the industry type; classify the R&D expenses into multiple categories according to the subcategories in the R&D expense items; the multiple categories include at least statistical categories, tax categories, and scientific and technological categories.

[0083] The R&D expense items include one or more of the following: personnel costs, direct input costs, depreciation and long-term amortization costs, intangible asset amortization costs, design and testing costs, other related costs, outsourced R&D costs, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; the industry types are divided into manufacturing and non-manufacturing.

[0084] If the basic data analysis is for the manufacturing industry, the stable structural items are one or more of the following: personnel labor costs and intangible asset amortization costs; the proportional constraint items are one or more of the following: depreciation costs and long-term amortization costs, design and testing costs, outsourced research and development costs, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; and the difference balancing items are one or more of the following: direct input costs and other related costs.

[0085] If the basic data analysis is for non-manufacturing industries, the stable structural item is one or more of intangible asset amortization expenses and direct input expenses; the proportional constraint item is one or more of depreciation expenses and long-term amortization expenses, design and testing expenses, commissioned external research and development expenses, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; the difference balancing item is one or more of personnel labor costs or other related expenses; S203, determine the proportion of the R&D expenses of the i-th project to the total R&D expenses, and obtain the R&D expenses of the i-th project as Ai; where i≥1;

[0086] S204. Based on the data in the information of the i-th project and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item of the R&D expense item in the i-th project;

[0087] Balance item amount Am = Ai - Ah - Ad;

[0088] S205. Allocate the difference balancing amount to each R&D item in the difference balancing item.

[0089] like Figure 3 As shown, the specific method for allocating the balance sheet amount to each R&D item within the balance sheet is as follows:

[0090] The difference balancing items are divided into primary adjustment R&D expense items and auxiliary adjustment R&D expense items;

[0091] First, the amount of the difference balancing item is allocated using the main adjustment R&D expense account; when the main adjustment R&D expense account does not meet the preset constraints, the remaining difference is transferred to the auxiliary adjustment R&D expense account for further allocation; after each round of allocation, the allocation result is checked for multi-dimensional constraints.

[0092] If the verification fails, repeat the transfer and allocation process until the error in the balance item amount meets the preset threshold or reaches the preset stop condition.

[0093] The multi-dimensional constraint verification includes at least one of the following: total amount error, project proportion deviation, inventory constraint, integer constraint, and continuous precision or caliber proportion limit.

[0094] The main adjustment R&D expense item is the R&D expense item with the largest R&D expense amount in the difference balancing item.

[0095] like Figure 4 As shown, the specific steps of the measured aggregation method are as follows:

[0096] S211. Confirm whether there are individual project R&D expenses in the R&D expenses, and whether there are R&D expense items and item fees;

[0097] S212. If there is no R&D expense for the i-th project in the R&D expenses, and there is no R&D expense account or account fee, use steps S202-S205 in the budget collection method to collect the R&D expenses; S213. If there is R&D expense for the i-th project in the R&D expenses, but there is no R&D expense account or account fee, first determine the R&D expense account to be collected, and determine the stable structure item, proportional constraint item, and difference balancing item in the R&D account, and use steps S204-S205 in the budget collection method to allocate the R&D expenses;

[0098] S214. If there is no R&D expense for the i-th project in the R&D expenses, but there are R&D expense items and item costs; determine the stable structural items, proportional constraint items, and difference balancing items in the R&D items; and collect R&D expenses using steps S203-S205 in the budget collection method.

[0099] like Figure 5 As shown, the S4 evidence chain output includes the following steps:

[0100] S41. Extract the association identifier by combining the R&D expense collection results and basic data;

[0101] The associated identifiers include at least the project identifier, business entity identifier, resource identifier, and subject identifier;

[0102] S42. Mapping between project numbering and R&D subject structure;

[0103] In the R&D expense aggregation results, the item number projects are mapped to the corresponding project numbers and item numbers in the basic data, forming a mapping relationship.

[0104] S43, Generation of the chain of evidence;

[0105] Based on the mapping relationship, the business data corresponding to each type of expense are associated and processed to generate corresponding supporting documents; including at least: personnel data, equipment data, and material data.

[0106] The personnel data generation steps are as follows: based on the personnel number, project number, and time period, the attendance data and salary data are associated to generate the basis for labor cost allocation. The basis for labor cost allocation includes a working hour percentage allocation table or a labor cost allocation table.

[0107] The equipment data generation steps are as follows: based on the equipment number, project number, and time period, the equipment operation record is associated with the depreciation calculation data to generate the basis for depreciation expense allocation. The basis for depreciation expense allocation includes a depreciation calculation table or a depreciation allocation table.

[0108] The material data generation steps are as follows: based on the material number, project number, and time period, the material requisition form and the material return form are associated to generate a material collection and allocation list, which includes a material allocation detail table.

[0109] S44, Data consistency verification;

[0110] Consistency checks are performed on personnel data, equipment data, and material data respectively; the checks include consistency checks on attendance data and wage data, and consistency checks on work hour allocation and labor costs;

[0111] Consistency verification between equipment operation records and depreciation data, and consistency verification between depreciation allocation and equipment costs;

[0112] Consistency verification between inventory data and inbound / outbound documents, and consistency verification between material requisition documents and material collection results;

[0113] S45, Output the chain of evidence;

[0114] The verified supporting documents are packaged to generate an evidence chain file package; the evidence chain file package includes at least:

[0115] Supporting documents for personnel expenses;

[0116] Supporting documents for equipment costs;

[0117] Supporting documents for material costs;

[0118] List of evidence chains;

[0119] A report linking the collected results with the original business data.

[0120] Example 1

[0121] This embodiment uses a manufacturing company as an example to explain the budget collection method in detail.

[0122] A method for allocating R&D expenses through a dual-path routing approach includes the following steps:

[0123] S1. Obtain basic data; The company's sales revenue in 2025 is 80 million yuan; It includes 3 R&D projects: RD01, RD02, and RD03; The start and end dates of all projects are from January 2025 to December 2025; Personnel information is shown in Table 1; Equipment data is shown in Table 2; Material data is shown in Table 3.

[0124] Table 1 contains personnel information.

[0125] Personnel Number Name Project Attendance days Project working hours percentage P001 Researcher A RD01 240 100% P002 Researcher B RD01 245 100% P003 Researcher C RD01 238 100% P004 Researcher D RD02 242 100% P005 Researcher E RD02 240 100% P006 Researcher F RD03 235 100% P007 Researcher G RD03 230 100% P008 Researcher H RD03 228 100%

[0126] Table 2 shows the equipment data.

[0127] Equipment Number Equipment Name Project Depreciation for the year (in ten thousand yuan) Device Attributes Operating hours (h) EQ001 R&D test bench RD01 8.5 Special equipment 320 EQ002 Precision measuring instruments RD01 12.3 Special equipment 280 EQ003 Simulation workstation RD02 9.8 Special equipment 240 EQ004 3D printer RD02 7.2 Special equipment 180 EQ005 Material testing instrument RD03 14.2 Special equipment 260

[0128] Table 3 shows the material data.

[0129] Material Number Material Name unit quantity Unit price (yuan) Amount (RMB) M001 integrated circuit indivual 500 600 300000 M002 Sensor module indivual 200 1200 240000 M003 connector set 400 800 320000 M004 PCB board piece 80 1500 120000 M005 copper wire kilogram 1000 40 40000

[0130] S2. Since there are no R&D expenses in the basic data, configure the R&D expense ratio requirement in the S1 basic data, and use the budget aggregation method to aggregate R&D expenses; the specific steps are as follows:

[0131] S201. Calculate the total R&D expenses based on sales revenue and the required R&D ratio. The company's sales revenue in 2025 is 80 million yuan, and the required R&D expense ratio is no less than 4%. Therefore, the total R&D expenses = 80 million × 4% = 3.2 million yuan.

[0132] S202. The R&D expense items are determined as personnel costs, depreciation and long-term amortization expenses, and direct input costs. The company is a manufacturing enterprise. The stable structure item is personnel costs; the proportional constraint item is depreciation and long-term amortization expenses; and the difference balancing item is direct input costs. According to the subcategories in the R&D expense items, they are divided into statistical, tax, and scientific and technological categories.

[0133] S203. Determine the proportion of the R&D expenses of the i-th project to the total R&D expenses, and obtain the R&D expenses of the i-th project as Ai; where i ≥ 1;

[0134] The budget allocations for the three projects are as follows: RD01 50%, RD02 30%, and RD03 20%. Therefore, the R&D costs for each project are:

[0135] Project RD01: A1 = 320 × 50% = 1.6 million yuan

[0136] Project RD02: A2 = 320 × 30% = 960,000 yuan

[0137] Project RD03: A3 = 320 × 20% = 640,000 yuan

[0138] S204. Based on the data in the information of the i-th project and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item of the R&D expense item in the i-th project;

[0139] Balance item amount Am = Ai - Ah - Ad;

[0140] The system reads information on 8 R&D personnel from the human resources module and automatically allocates labor costs based on project relationships and attendance records; attendance data is linked to personnel numbers and project numbers to ensure that salary allocation is traceable.

[0141] Calculate the allocation of labor costs for each project based on attendance records:

[0142] Labor costs for Project RD01: 640,000 yuan

[0143] Labor costs for Project RD02: 480,000 yuan

[0144] Labor costs for Project RD03: 560,000 yuan

[0145] The total labor cost for the three projects is Ah = 64 + 48 + 56 = 1.68 million yuan;

[0146] Read the operation records of 5 R&D equipment in the fixed assets module, allocate R&D man-hours according to the budget ratio, calculate depreciation expenses, and calculate depreciation expenses:

[0147] Depreciation expense for project RD01: 85,000 + 123,000 = 208,000 yuan

[0148] Depreciation expense for RD02 project: 9.8 + 7.2 = 170,000 yuan

[0149] Depreciation expense for RD03 project: 142,000 yuan

[0150] The total depreciation expense for the three projects is Ad = 20.8 + 17 + 14.2 = 520,000 yuan;

[0151] The amount of the difference balancing item Am = T - Ah - Ad = 320 - 168 - 52 = 1 million yuan;

[0152] S205. Allocate the balance sheet amount to each R&D item within the balance sheet;

[0153] In this embodiment, the difference balancing item is the direct input cost (material cost), and the allocation method includes the following steps:

[0154] The material can be divided to identify discrete materials (e.g., units, pieces). Integer solutions are used for discrete materials, while continuous materials (e.g., kg, meters) can be divided. A base value of 980,000 yuan is allocated for integer materials.

[0155] The base allocation amount for materials is 980,000 yuan, and the target material cost is 1,000,000 yuan. Therefore, the difference Δ = 1,000,000 - 980,000 = 20,000 yuan.

[0156] Select copper wire with a unit price of 40 yuan / kg from the divisible materials. Its unit price and the difference value are relatively easy to match; the quantity to make up the difference x = Δ ÷ unit price = 20000 ÷ 40 = 500kg.

[0157] Total material cost = integer base + difference = 980,000 + 500 × 40 = 980,000 + 20,000 = 1,000,000 yuan, which is exactly equal to the target value.

[0158] The material costs for each project are allocated as follows:

[0159] Material costs for Project RD01: 500,000 yuan

[0160] Material costs for Project RD02: 300,000 yuan

[0161] Material costs for Project RD03: 200,000 yuan

[0162] In this embodiment, Ah + Ad + Am = 168 + 52 + 100 = 3.2 million yuan, and the total error E = |320 -320| = 0, which satisfies the convergence condition, and the iteration stops.

[0163] S4, Output of the chain of evidence;

[0164] S41. The system combines the results of R&D expense collection with basic data to extract related identifiers, including project identifiers (RD01, RD02, RD03), business entity identifiers, personnel identifiers (P001-P008), equipment identifiers (EQ001-EQ005), material identifiers (M001-M004), and subject identifiers.

[0165] S42. Project number and R&D subject structure mapping: Match the subject number in the R&D expense aggregation results with the project number, and map business documents (attendance records, equipment operation records, material requisition forms) to the corresponding project number and subject number.

[0166] S43. Generation of the chain of evidence, including:

[0167] Personnel Data: Generate a work hour allocation table, linking attendance data and salary data.

[0168] Equipment Data: Generates a depreciation allocation table, linking equipment operation records with depreciation calculation data.

[0169] Material Data: Generates a detailed material allocation table, linking material requisition forms with material collection results.

[0170] S44. Data consistency verification: Verify that the data consistency in each dimension has passed.

[0171] S45. Output evidence chain file package, including personnel expense proof materials, equipment expense proof materials, material expense proof materials, evidence chain list and a report on the correlation between the collection results and the original business data.

[0172] Example 2

[0173] This embodiment uses a software company as an example to collect R&D expenses.

[0174] A method for allocating R&D expenses through a dual-path routing approach includes the following steps:

[0175] S1. Obtain basic data: the company's operating data for the first quarter of 2025, with sales revenue of 15 million yuan and total R&D expenses of 900,000 yuan; also including project information, personnel information, equipment data, and material data; project information is shown in Table 4; personnel information is shown in Table 5; equipment data is shown in Table 6.

[0176] Table 4 Project Information

[0177] Project Number Project Name Project cycle Project Phase RD04 Intelligent Algorithm Optimization System 2025.01-2025.06 Development phase RD05 Data analysis platform 2025.01-2025.09 Development phase

[0178] Table 5 Personnel Information

[0179] Personnel Number Name Corresponding project Monthly salary (ten thousand yuan) P101 R&D Engineer A RD04 1.8 P102 Senior Engineer B RD04 2.5 P103 Algorithm Engineer C RD04 2.2 P104 Test Engineer D RD04 1.6 P105 Data Engineer E RD05 2.0 P106 Architect F RD05 2.8 P107 Developer G RD05 1.9 P108 Operations Engineer H RD05 1.8

[0180] Table 6 Equipment Data

[0181] Equipment Number Equipment Name Corresponding project Monthly depreciation (in ten thousand yuan) EQ101 server cluster RD04 5.5 EQ102 Algorithm training GPU RD04 8.2 EQ103 cloud service platform RD05 6.8

[0182] S2. If R&D expenses are included in the basic data, then the actual measurement and aggregation method is used to aggregate the R&D expenses; the specific steps are as follows:

[0183] S211. Confirm whether there are individual project R&D expenses in the R&D expenses, and whether there are R&D expense items and item fees;

[0184] S212. There is no R&D expense for the i-th project in the R&D expenses, and there is no R&D expense account and account fee. First, determine the R&D expense account to be collected. The R&D expense account to which it belongs is intangible asset amortization, depreciation expense and long-term amortization expense, design and testing expenses, personnel labor costs, and other related expenses.

[0185] Based on the basic data, the industry is identified as non-manufacturing (software and information technology services). The stable structural item is the amortization of intangible assets; the proportional constraint items are depreciation expenses and long-term amortization expenses, design and testing expenses, etc.; the difference balancing item is personnel labor costs and other related expenses. According to the subcategories in the R&D expense account, R&D expenses are divided into multiple calibers; the multiple calibers include at least statistical caliber, tax caliber, and scientific and technological caliber.

[0186] S203. Determine the proportion of the R&D expenses of the i-th project to the total R&D expenses, and obtain the R&D expenses of the i-th project as Ai; i≥1; RD04 accounts for 60%, with a specific amount of 540,000 yuan; RD05 accounts for 40%, with a specific amount of 360,000 yuan;

[0187] S204. Based on the data in the information of the i-th project and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item of the R&D expense item in the i-th project;

[0188] The amount of the difference balancing item is Am = Ai - Ah - Ad; Specifically:

[0189] Calculate the amortization expense of intangible assets based on the information in the basic data; as shown in Table 7.

[0190] Table 7 Amortization Expense of Intangible Assets

[0191] Intangible asset number Intangible asset name Corresponding project Monthly amortization (ten thousand yuan) IA001 Image recognition algorithm patent RD04 3.2 IA002 Data Processing Software V2.0 RD04 2.8 IA003 Intelligent analysis model RD05 4.5

[0192] The total amount of stable structure items for the two projects is Ah = 6 + 4.5 = 10.5 million yuan;

[0193] Based on the equipment data and design experiment cost records, confirm the cost amount for the proportional constraint item; the design experiment cost records are shown in Table 8:

[0194] Table 8. Record of Design and Experiment Costs

[0195] Fee type Corresponding project Amount (ten thousand yuan) Algorithm testing fee RD04 4.5 Data labeling fee RD05 3.2

[0196] Project RD04 proportional constraint item: 5.5 + 8.2 + 4.5 = 18.2 million yuan; Project RD05 proportional constraint item: 6.8 + 3.2 = 10 million yuan; Total proportional constraint item amount for both projects Ad = 18.2 + 10 = 28.2 million yuan;

[0197] The amount of the balance item is:

[0198] Project RD04: Am = 54 - 6 - 18.2 = 298,000 yuan;

[0199] Project RD05: Am = 36 - 4.5 - 10 = 215,000 yuan;

[0200] S205. Allocate the difference balancing amount to each R&D item in the difference balancing item.

[0201] The difference balancing items are divided into primary adjustment R&D expense items and auxiliary adjustment R&D expense items;

[0202] The primary adjustment item for R&D expenses is personnel labor costs; the secondary adjustment item for R&D expenses is other related expenses.

[0203] Based on personnel information, the personnel labor costs for project RD04 are 81,000 yuan; the personnel labor costs for project RD05 are 85,000 yuan. The amount of the auxiliary adjustment R&D expense item is: difference for project RD04: 298,000 - 81,000 = 217,000 yuan; difference for project RD05: 215,000 - 85,000 = 130,000 yuan. The amount of the auxiliary adjustment R&D expense item is allocated to auxiliary items such as travel expenses, office expenses, and expert consultation fees to verify that the multi-dimensional constraints are met.

[0204] S3. Output the R&D expense aggregation results; the results are shown in Table 9.

[0205] Table 9 Results of R&D Expense Collection

[0206] Project Number Amortization of intangible assets (ten thousand yuan) Depreciation and design costs (ten thousand yuan) Personnel costs (ten thousand yuan) Other expenses (ten thousand yuan) Total (ten thousand yuan) RD04 6 18.2 8.1 21.7 54 RD05 4.5 10 8.5 13 36 total 10.5 28.2 16.6 34.7 90

[0207] S4, Output of the chain of evidence; the specific steps are the same as S4 in Example 1.

[0208] Example 3

[0209] This embodiment uses a biopharmaceutical company as an example to collect research and development expenses.

[0210] A method for allocating R&D expenses through a dual-path routing approach includes the following steps:

[0211] S1. Obtain basic data, operating data for February 2025, including project information, incurred project R&D expenses, personnel information, equipment data, and material data; project information is shown in Table 10; personnel information is shown in Table 11; equipment data is shown in Table 12; and design and testing costs are shown in Table 13.

[0212] Table 10 Project Information

[0213] Project Number Project Name Project cycle Actual R&D expenses in February 2025 (RMB 10,000) RD06 New drug development project A 2025.01-2025.12 85 RD07 Drug consistency evaluation 2025.01-2025.06 42

[0214] Table 11 Personnel Information

[0215] Personnel Number Name Project Attendance days in February Salary (ten thousand yuan) P201 Project Manager A RD06 20 3.5 P202 R&D Director B RD06 20 3.2 P203 Pharmacology researcher C RD06 18 2.8 P204 Clinical Researcher D RD06 20 2.6 P205 Quality Specialist E RD06 19 2.2 P206 Experimenter F RD06 20 1.8 P207 Analyst G RD07 20 2.4 P208 Statistician H RD07 18 2.0 P209 Data Clerk I RD07 19 1.6

[0216] Table 12 Equipment Data

[0217] Equipment Number Equipment Name Corresponding project February depreciation (in ten thousand yuan) EQ201 High Performance Liquid Chromatography RD06 4.5 EQ202 Gas chromatograph RD06 3.8 EQ203 Stability test chamber RD06 2.9 EQ204 Data Analysis Workstation RD07 2.2

[0218] Table 13 Costs of Designing Experiments

[0219] Fee type Corresponding project Amount (ten thousand yuan) Clinical trial fees RD06 5.5 Pharmacokinetic studies RD06 4.2 Bioequivalence test RD07 6.8

[0220] S2. If R&D expenses are included in the basic data, then the actual measurement and aggregation method is used to aggregate the R&D expenses; the specific steps are as follows:

[0221] S211. Confirm whether there are individual project R&D expenses in the R&D expenses, and whether there are R&D expense items and item fees;

[0222] S212. The R&D expenses include the R&D expenses for each project, but there are no R&D expense items and item fees. First, determine the R&D expense items to be collected, and then divide the determined R&D expense items into stable structure items, proportional constraint items and difference balance items according to the industry type.

[0223] Based on the basic data, the industry is identified as the manufacturing sector. The stable structural item is identified as personnel labor costs; the proportional constraint items are depreciation and long-term amortization expenses, design and testing costs, etc.; the difference balancing items are direct input costs and other related costs; and the R&D expenses are divided into multiple categories based on the subcategories in the R&D expense account. The multiple categories include at least statistical, tax, and scientific and technological categories.

[0224] S204. Based on the data in each project information and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item for each project's R&D expense item;

[0225] The amount of the difference balancing item is Am = Ai - Ah - Ad; Specifically:

[0226] Labor costs for project RD06: 161,000 yuan; Labor costs for project RD07: 60,000 yuan;

[0227] The total amount of stable structure items for the two projects is Ah = 16.1 + 6 = 22.1 million yuan;

[0228] Based on the equipment data and design experiment cost records, confirm the cost amount for the proportional constraints; Proportional constraints for project RD06: 4.5 + 3.8 + 2.9 + 5.5 + 4.2 = 209,000 yuan; Proportional constraints for project RD07: 2.2 + 6.8 = 90,000 yuan; Total proportional constraints for both projects: Ad = 209,000 + 90,000 = 299,000 yuan.

[0229] The amount of the balance item is:

[0230] Project RD06: Am = 85 - 16.1 - 20.9 = 480,000 yuan

[0231] Project RD07: Am = 42 - 6 - 9 = 270,000 yuan;

[0232] S205. Allocate the difference balancing amount to each R&D item in the difference balancing item.

[0233] The difference balancing items are divided into primary adjustment R&D expense items and auxiliary adjustment R&D expense items;

[0234] The primary adjustment item for R&D expenses is direct input costs; the secondary adjustment item for R&D expenses is other related expenses; direct input costs are shown in Table 14.

[0235] Table 14 Direct Input Costs

[0236] Material Number Material Name Corresponding project quantity Unit price (ten thousand yuan) Amount (ten thousand yuan) M201 Chemical raw materials RD06 100kg 0.2 20 M202 auxiliary materials RD06 500kg 0.05 25 M203 Laboratory consumables RD07 - - 15

[0237] Direct input costs for project RD06: 450,000 yuan; Direct input costs for project RD07: 150,000 yuan; Auxiliary adjustment items (other related expenses): Difference for project RD06: 480,000 - 450,000 = 30,000 yuan; Difference for project RD07: 270,000 - 150,000 = 120,000 yuan; Allocate the difference to auxiliary items such as water and electricity fees, rental fees, and expert consultation fees to verify that the multi-dimensional constraints are met; Total error E = |127 - 127| = 0, which meets the convergence condition.

[0238] The results are shown in Table 15.

[0239] Table 15 Results of R&D Expense Collection

[0240] Project Number Personnel costs (ten thousand yuan) Depreciation and testing costs (ten thousand yuan) Direct investment costs (ten thousand yuan) Other expenses (ten thousand yuan) Total (ten thousand yuan) RD06 16.1 20.9 45 3 85 RD07 6 9 15 12 42 total 22.1 29.9 60 15 127

[0241] S4, Output of the chain of evidence; the specific steps are the same as S4 in Example 1.

[0242] Example 4

[0243] This embodiment uses a technology service company as an example to collect R&D expenses.

[0244] A method for allocating R&D expenses through a dual-path routing approach includes the following steps:

[0245] S1. Obtain basic data, operating data for March 2025, including project information, data on incurred R&D expenses, personnel information, equipment data, and material data; the total R&D expenses are RMB 1 million; project information is shown in Table 16; personnel information is shown in Table 17; equipment data is shown in Table 18; and data on incurred R&D expenses is shown in Table 19.

[0246] Table 16 Project Information

[0247] Project Number Project Name Project cycle Project Status RD08 Smart City Solutions 2025.01-2025.12 In progress RD09 IoT platform development 2025.01-2025.09 In progress RD10 Digital transformation consulting 2025.03-2025.08 New Start

[0248] Table 17 Personnel Information

[0249] Project Number Number of R&D personnel Average monthly salary per person (in ten thousand yuan) Personnel costs (ten thousand yuan) RD08 6 2.2 13.2 RD09 4 2.0 8.0 RD10 3 1.9 5.7

[0250] Table 18 Equipment Data

[0251] Equipment Number Equipment Name Monthly depreciation (in ten thousand yuan) Project Ownership EQ301 cloud server cluster 5 RD08 accounts for 60%, and RD09 accounts for 40%. EQ302 IoT gateway devices 3.5 RD09 accounts for 100%

[0252] Table 19 Data on R&D Expenses Incurred

[0253] Subject Name Cumulative amount in February (ten thousand yuan) Intangible asset amortization 12 Depreciation expense and long-term amortization 8.5 Personnel costs 35 Other related expenses 15

[0254] S2. If R&D expenses are included in the basic data, then the actual measurement and aggregation method is used to aggregate the R&D expenses; the specific steps are as follows:

[0255] S211. Confirm whether there are R&D expenses for each project in the R&D expenses, and whether there are R&D expense items and item fees;

[0256] S212. There is no R&D expense for each project in the R&D expenses. There are R&D expense items and item costs. According to the industry type, the R&D expense items are divided into stable structural items, proportional constraint items and difference balancing items.

[0257] Based on the basic data, it was determined to be a non-manufacturing sector. The stable structural item was identified as amortization of intangible assets; the proportional constraint items were depreciation expenses and long-term amortization expenses; the difference balancing items were personnel labor costs and other related expenses; and the R&D expenses were divided into multiple categories based on the subcategories in the R&D expense account. The multiple categories include at least statistical, tax, and scientific and technological categories.

[0258] S203. Determine the proportion of R&D expenses for each project to the total R&D expenses, and obtain the R&D expenses for each project; RD08 accounts for 45%, with a specific amount of 450,000 yuan; RD09 accounts for 35%, with a specific amount of 350,000 yuan; RD10 accounts for 20%, with a specific amount of 200,000 yuan;

[0259] S204. Based on the data in each project information and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item for each project's R&D expense item;

[0260] The amount of the difference balancing item is Am = Ai - Ah - Ad; Specifically:

[0261] Based on intangible asset amortization data, the amortization is allocated according to the proportion of project working hours.

[0262] Amortization of intangible assets for project RD08: 80,000 yuan; amortization of intangible assets for project RD09: 40,000 yuan; amortization of intangible assets for project RD10: 0 yuan; the total amount of stable structural items for the three projects, Ah = 120,000 yuan.

[0263] Based on the basic data, confirm the amount of the proportional constraint items; Depreciation expense for RD08 project: 5 × 60% = 30,000 yuan, Depreciation expense for RD09 project: 5 × 40% + 35,000 = 55,000 yuan, Depreciation expense for RD10 project: 0 yuan; The total amount of the proportional constraint items for the three projects is Ad = 85,000 yuan.

[0264] The amount of the balance item is:

[0265] Project RD08: Am = 45 - 8 - 3 = 340,000 yuan

[0266] Project RD09: Am = 35 - 4 - 5.5 = 255,000 yuan

[0267] Project RD10: Am = 20 - 0 - 0 = 200,000 yuan;

[0268] S205. Allocate the difference balancing amount to each R&D item in the difference balancing item.

[0269] The difference balancing items are divided into primary adjustment R&D expense items and auxiliary adjustment R&D expense items;

[0270] The primary adjustment item for R&D expenses is personnel costs; the secondary adjustment item is other related expenses; the secondary adjustment items (other related expenses) are: RD08 project difference: 34 - 13.2 = 20.8 million yuan; RD09 project difference: 25.5 - 8.0 = 17.5 million yuan; RD10 project difference: 20 - 5.7 = 14.3 million yuan; the difference is allocated to secondary items such as water and electricity fees, rental fees, and expert consultation fees to verify that the multi-dimensional constraints are met; the total error E = |100 - 100| = 0, which meets the convergence condition.

[0271] The results are shown in Table 20.

[0272] Table 20 Results of R&D Expense Collection

[0273] Project Number Amortization of intangible assets (ten thousand yuan) Depreciation expense (ten thousand yuan) Personnel costs (ten thousand yuan) Other expenses (ten thousand yuan) Total (ten thousand yuan) RD08 8 3 13.2 20.8 45 RD09 4 5.5 8.0 17.5 35 RD10 0 0 5.7 14.3 20 total 12 8.5 26.9 52.6 100

[0274] S4, Output of the chain of evidence; the specific steps are the same as S4 in Example 1.

Claims

1. A dual-path diversion R&D credit collection method, characterized in that, Includes the following steps: S1. Obtain basic data, which includes at least project information, sales revenue, and personnel information; S2. Determine the method for collecting R&D expenses based on whether there are R&D expenses in the basic data; If there are no R&D expenses, configure the R&D expense ratio requirement in the S1 basic data and use the budget collection method to collect R&D expenses; If there are R&D expenses, the actual measurement and aggregation method shall be used to aggregate the R&D expenses; S3, Output the results of R&D expense aggregation.

2. The dual-path diversion R&D fee collection method according to claim 1, wherein, The basic data in S1 also includes at least equipment data and material data.

3. The dual-path diversion R&D collection method according to claim 2, wherein, The project information shall include at least the project number, project name, and project start and end dates; The personnel information shall include at least name, project affiliation, and total number of days worked; The equipment data shall include at least the equipment name, equipment number, equipment quantity, monthly depreciation, and total usage time of the equipment; The material data shall include at least the material name, material number, material quantity, material unit price, and material cost; The sales revenue mentioned refers to monthly, quarterly, or annual revenue.

4. The dual-path diversion R&D collection method of claim 1, wherein, In S2, the specific steps of the budget collection method are as follows: S201. Calculate the total R&D expenses based on the sales revenue and R&D ratio requirements; S202. Determine the R&D expense items to be collected, and divide the R&D expense items into stable structural items, proportional constraint items, and difference balancing items; S203. Determine the proportion of the R&D expenses of the i-th project to the total R&D expenses, and obtain the R&D expenses of the i-th project as Ai; where i ≥ 1; S204. Based on the data in the information of the i-th project and the determined R&D expense items, determine the amount Ah of the stable structure item and the amount Ad of the proportional constraint item of the R&D expense item in the i-th project; Balance item amount Am = Ai - Ah - Ad; S205. Allocate the difference balancing amount to each R&D item in the difference balancing item.

5. The dual-path diversion R&D collection method according to claim 4, wherein, In step S202, after determining the R&D expense items to be collected, the method further includes: dividing the R&D expenses into multiple categories based on the subcategories in the R&D expense items; the multiple categories include at least statistical categories, tax categories, and scientific and technological categories. The R&D expense items include one or more of the following: personnel costs, direct input costs, depreciation and long-term amortization costs, intangible asset amortization costs, design and testing costs, other related costs, outsourced R&D costs, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures.

6. The dual-path diversion R&D collection method according to claim 4, wherein, In step S202, when determining the R&D expense items into stable structure items, proportional constraint items, and difference balancing items, the industry type is first determined based on the basic data, and then the items are divided according to the industry type. Industry types are divided into manufacturing and non-manufacturing; If the basic data analysis is for the manufacturing industry, the stable structural items are one or more of the following: personnel labor costs and intangible asset amortization costs; the proportional constraint items are one or more of the following: depreciation costs and long-term amortization costs, design and testing costs, outsourced research and development costs, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; and the difference balancing items are one or more of the following: direct input costs and other related costs. If the basic data analysis is non-manufacturing, the stable structural item is one or more of the following: intangible asset amortization expenses and direct input expenses; the proportional constraint item is one or more of the following: depreciation expenses and long-term amortization expenses, design and testing expenses, commissioned external research and development expenses, land and building expenditures, instrument and equipment expenditures, patent and proprietary technology expenditures, and capitalized computer software expenditures; the difference balancing item is one or more of the following: personnel labor costs or other related expenses.

7. The method for collecting R&D expenses through dual-path diversion according to claim 4, characterized in that, The specific method for allocating the balance sheet amount to each R&D item within the balance sheet in step 205 is as follows: The difference balancing items are divided into primary adjustment R&D expense items and auxiliary adjustment R&D expense items; First, the amount of the difference balancing item is allocated using the main adjustment R&D expense account; when the main adjustment R&D expense account does not meet the preset constraints, the remaining difference is transferred to the auxiliary adjustment R&D expense account for further allocation. After each round of allocation is completed, multi-dimensional constraint verification is performed on the allocation results; If the verification fails, repeat the transfer and allocation process until the error in the balance item amount meets the preset threshold or reaches the preset stop condition. The multi-dimensional constraint verification includes at least one of the following: total amount error, project ratio deviation, inventory constraint, integer constraint, continuous precision or caliber ratio limit.

8. The method for collecting R&D expenses through dual-path diversion according to claim 1, characterized in that, In S2, the specific steps of the measured aggregation method are as follows: S211. Confirm whether there are individual project R&D expenses in the R&D expenses, and whether there are R&D expense items and item fees; S212. If there is no R&D expense for the i-th project in the R&D expenses, and there is no R&D expense item and item fee, the R&D expenses shall be collected using steps S202-S205 in the budget collection method described in any of claims 4-7. S213. If there is R&D expense for the i-th project in the R&D expenses, but there is no R&D expense account and account expense; first determine the R&D expense account to be collected, and determine the stable structure item, proportional constraint item and difference balancing item in the R&D account, and use steps S204-S205 in the budget collection method of any one of claims 4-7 to allocate the R&D expenses. S214. If there is no R&D expense for the i-th project in the R&D expenses, there are R&D expense items and item costs; determine the stable structure items, proportional constraint items and difference balancing items in the R&D items; use steps S203-S205 in the budget collection method described in any of claims 4-7 to collect R&D expenses.

9. The method for collecting R&D expenses through dual-path diversion according to claim 1, characterized in that, It also includes step S4, outputting the chain of evidence; specifically, it includes the following steps: S41. Extract the association identifier by combining the R&D expense collection results and basic data; The associated identifiers include at least the project identifier, business entity identifier, resource identifier, and subject identifier; S42. Mapping between project numbering and R&D subject structure; In the R&D expense aggregation results, the item number projects are mapped to the corresponding project numbers and item numbers in the basic data, thus establishing a mapping relationship. S43, Generation of the chain of evidence; Based on the mapping relationship, the business data corresponding to each type of expense are associated and processed to generate corresponding supporting documents; including at least: personnel data, equipment data, and material data. The personnel data generation steps are as follows: based on the personnel number, project number, and time period, the attendance data and salary data are associated to generate the basis for labor cost allocation. The basis for labor cost allocation includes a working hour percentage allocation table or a labor cost allocation table. The equipment data generation steps are as follows: based on the equipment number, project number, and time period, the equipment operation record is associated with the depreciation calculation data to generate the basis for depreciation expense allocation. The basis for depreciation expense allocation includes a depreciation calculation table or a depreciation allocation table. The material data generation steps are as follows: based on the material number, project number, and time period, the material requisition form and the material return form are associated to generate a material collection and allocation list, which includes a material allocation detail table. S44, Data consistency verification; Consistency checks are performed on personnel data, equipment data, and material data respectively; the checks include consistency checks on attendance data and wage data, and consistency checks on work hour allocation and labor costs; Consistency verification between equipment operation records and depreciation data, and consistency verification between depreciation allocation and equipment costs; Consistency verification between inventory data and inbound / outbound documents, and consistency verification between material requisition documents and material collection results; S45, Output the chain of evidence; The verified supporting documents are packaged to generate an evidence chain file package; the evidence chain file package includes at least: Supporting documents for personnel expenses; Supporting documents for equipment costs; Supporting documents for material costs; List of evidence chains; A report linking the collected results with the original business data.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method according to any one of claims 1-9.

Images