Modularization-based insurance background cost allocation method

Abstract

The invention discloses a modularization-based insurance mid-background cost allocation method, which comprises the following steps: S1, module division and definition: a mid-background allocation process is divided into a plurality of independent functional modules, each module corresponds to a clear sub-function and a resource demand, and a module attribute database is established to record resource consumption parameters; s2, a dynamic combination mechanism: constructing a module relationship network, and defining a data flow direction and a dependency relationship between modules through a data interface and a logic rule; s3, dynamic sequence adjustment: setting a configurable allocation strategy engine, and dynamically adjusting a module execution sequence according to business priority and resource consumption weight factors; and S4, rule adaptation and execution: an independent allocation rule subsystem is built in each module, and cost calculation is carried out according to a resource consumption proportion, a business volume and a custom formula rule. According to the invention, flexible adjustment, interrupted calculation and efficient execution of the cost allocation process can be realized, and the efficiency and accuracy of insurance cost management are improved.

Description

Technical Field

[0001] This invention relates to a method for allocating back-office expenses in insurance, and more particularly to a modular method for allocating back-office expenses in insurance. Background Technology

[0002] In the insurance industry, the allocation of middle and back-office expenses is a crucial aspect of cost management. Current methods for allocating middle and back-office expenses in the insurance sector typically employ static, linear process designs. This approach relies on predefined, fixed procedures; if data anomalies, rule conflicts, or business structure adjustments occur during the allocation process, the entire allocation process must be recalculated from scratch.

[0003] For example, when the cost allocation process is interrupted due to missing data from a business department, traditional methods can only terminate the current process, correct the data, and restart all calculation steps. Similarly, when an insurance company adds a new business line or adjusts business priorities, the existing cost allocation processes and rules cannot be quickly adapted, requiring manual redesign and deployment of a new allocation scheme. The root of these problems lies in the lack of process flexibility and dynamic adjustment capabilities in existing methods. They treat complex business processes as an indivisible whole, failing to efficiently handle local changes, resulting in low cost allocation efficiency and an inability to meet the needs of rapid iteration and refined management in the insurance business. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a modular insurance back-office expense sharing method, which can solve the problems of lack of flexibility in the existing insurance back-office expense sharing process, inability to quickly recover after the sharing is interrupted, and difficulty in adapting to dynamic changes in business.

[0005] To address the aforementioned technical problems, this invention provides a modular insurance back-office expense allocation method, comprising the following steps: S1. Module division and definition: The back-office expense allocation process is decomposed into several independent functional modules according to expense type or expense scenario. Each module corresponds to a clear sub-function and resource requirements, and a module attribute database is established to record resource consumption parameters; S2. Dynamic combination mechanism: A module relationship network is constructed, and the data flow and dependencies between modules are defined through data interfaces and logical rules; when the allocation process is interrupted due to data errors or rule conflicts, related modules are extracted from the module relationship network and dynamically combined according to the location of the interruption point, skipping the abnormal module and regenerating the allocation path; S3. Dynamic sequence adjustment: A configurable allocation strategy engine is set up to dynamically adjust the module execution order according to business priority and resource consumption weight factors; S4. Rule adaptation and execution: Each module has a built-in independent allocation rule subsystem that supports expense calculation based on resource consumption ratio, business volume, and custom formula rules; according to the dynamically combined module sequence, the allocation rules of each module are executed sequentially, and the calculation results are summarized level by level to finally complete the accurate allocation of back-office expenses.

[0006] Furthermore, step S1 breaks down the middle and back-end cost-sharing process into the following independent functional modules:

[0007] Payroll module: Connects to the human resources and payroll system to obtain salary, performance, benefits, and social security data. Based on the human resources clearing situation, it aggregates costs to the corresponding vehicle, non-vehicle, and agricultural insurance categories or detailed insurance types, as well as individual customers, group customers, and agricultural center customers. Finally, it processes the data to the general ledger and the deferred data resource pool.

[0008] Group Customer Performance Allocation Module: The module coordinates the human resources and payroll system and the performance system to obtain non-vehicle group customer performance data. It matches the amounts on both sides by the organization, department, and amount fields to see if they are consistent. If they are consistent, the detailed dimension data of the performance system is used for allocation. If they are inconsistent, the payroll system data is used. Then, the sales management and budget system is used to obtain performance tracking data and allocate the performance to the detailed insurance policies. The detailed allocation results are processed and sent to the deferred data resource pool.

[0009] The overall sales module connects to the expense control system to obtain the actual expense reimbursement data for vehicle, non-vehicle, and agricultural insurance in the current month, associates the detailed budget data of the expense reimbursement data, and then uses the premium income to allocate to the policy details data. The allocation results are then processed and sent to the deferred data resource pool.

[0010] Insurance type allocation module: It connects to the general ledger, expense control and deferred resource pool data, and allocates the insurance to the detailed insurance types according to the customized allocation rules and allocation level;

[0011] Channel Allocation Module: Connects to the insurance type allocation module to allocate the allocation results to detailed channels and customer groups according to the customized allocation rules;

[0012] Online Sales Allocation Module: Connects to the allocation results of the channel allocation module to perform secondary allocation of auto insurance online sales costs;

[0013] Level 3 and 4 cost allocation module: Connects with the cost allocation module of channel allocation module and network sales allocation module to allocate the middle and back office expenses of headquarters and branch offices to the front office departments;

[0014] Off-balance sheet assessment allocation module: Data that cannot be obtained from the system is processed into the deferred data resource pool through manual entry and allocated to the detailed dimensions according to custom allocation rules.

[0015] Furthermore, when an anomaly is found in the data of a certain case during the claims apportionment calculation, causing an interruption, step S2 directly calls the completed underwriting module data, combines the financial settlement module and the corrected claims calculation module, and resumes the apportionment process from the point of interruption.

[0016] Furthermore, step S3 sets multiple alternative rules for the same cost allocation target, and determines the priority and effective conditions of the rules through arbitration by weighted average method; step S3 uses the cost allocation strategy engine to prioritize the cost allocation modules related to new products and complete the cost allocation first; or during the peak settlement period at the end of the month, it automatically adjusts the execution order of modules to optimize calculation efficiency.

[0017] Furthermore, the detailed budget data used in the associated reimbursement forms in the overall sales module includes the scope of insurance types and the level of application.

[0018] Furthermore, step S3 includes: generating a unique allocation serial number for each original cost after each allocation, and recording the complete allocation path to form an allocation tree, thereby realizing full-link allocation traceability and penetrating query.

[0019] Furthermore, the allocation tracing and penetration query process in step S3 is as follows: Drill down: From a certain total management expense at headquarters, directly drill down to find out which channel and which type of insurance ultimately borne how much amount; Summarize upward: From the cost of a certain grassroots salesperson, trace back to which headquarters expenses it originated from and how many times it was allocated.

[0020] Furthermore, the module attribute database records all costs to be allocated and sets metadata tags to form...

[0021] A deferred resource pool is established, and a driver data pool is pre-built based on the expense reimbursement model, procurement expense model, and IT expense model to centrally manage all the allocation benchmark data that will be used. When step S4 is executed, the allocation is performed in a hierarchical recursive manner, and the output results of the upper-level module are used as the input resources of the lower-level module. At the same time, the lower-level module calls the driver data pool to perform secondary allocation using more granular driver data, so that the expenses are ultimately traced back to the last business unit.

[0022] Furthermore, the metadata tags include amount, expense type, department where the expense occurred, original voucher number, and cost attribute; the allocation benchmark data includes premiums for each type of insurance, number of policies for each channel, number of employees and working hours for each department.

[0023] Furthermore, the expense reimbursement model connects to the actual sales data of expense control and expense reimbursement forms, classifies data into vehicle, non-vehicle, and agricultural categories, and allocates the data to detailed insurance policies from different dimensions; the procurement expense model connects to the detailed ledger data of the procurement system, replaces the general ledger data that is not divided into detailed categories, and performs expense restoration before allocation; the IT expense model selects different allocation factors for IT-related expense data types, such as system maintenance fees and hardware procurement costs.

[0024] Compared with the prior art, the present invention has the following beneficial effects: The modular insurance back-office cost allocation method provided by the present invention, through modular design and dynamic combination technology, realizes flexible adjustment, interruption and continuation of calculation and efficient execution of the cost allocation process, thereby improving the efficiency and accuracy of insurance cost management. Attached Figure Description

[0025] Figure 1 This invention is based on a modular flowchart of insurance back-office cost sharing.

[0026] Figure 2 This is a schematic diagram of the modular insurance back-office cost-sharing system architecture of the present invention;

[0027] Figure 3 This is a diagram showing the multiple independent functional modules broken down from the background allocation process of this invention. Detailed Implementation

[0028] The present invention will now be further described with reference to the accompanying drawings and embodiments.

[0029] Figure 1 This invention is based on a modular flowchart of insurance back-office cost sharing. Figure 2 This is a schematic diagram of the modular insurance back-office cost sharing system architecture of the present invention.

[0030] Please see Figure 1 and Figure 2The modular insurance back-office cost sharing method provided by this invention includes the following steps:

[0031] S1. Module Division and Definition: The back-end cost allocation process is broken down into independent functional modules. Each module corresponds to a clear sub-function and resource requirements, and a module attribute database is established to record resource consumption parameters. Module scope is divided according to cost type or cost scenario, including: expense reimbursement model, procurement cost model, IT cost model, etc.

[0032] Reimbursement model: Connects the actual sales data from expense control to reimbursement forms, and divides the data into vehicle, non-vehicle, and agricultural categories to allocate to detailed insurance policies according to different dimensions;

[0033] Procurement cost model: Connects to the detailed ledger data of the procurement system, replaces the general ledger data that does not distinguish between detailed data, performs cost restoration before allocation, and improves data accuracy;

[0034] IT cost model: Different allocation factors are selected for IT-related cost types, such as system maintenance fees and hardware procurement costs, to improve the rationality of cost allocation.

[0035] S2. Dynamic Combination Mechanism: A module relationship network is constructed, defining data flow and dependencies between modules through data interfaces and logical rules. When the allocation process is interrupted due to data errors, rule conflicts, or other reasons, the system can dynamically combine related modules from the module relationship network based on the location of the interruption point, skipping the abnormal module and regenerating the allocation path. For example, if an abnormal case data is found to cause an interruption in the claims allocation calculation, the system can directly call the completed underwriting module data (core business allocation driver data), combine the financial settlement module (referring to the completed allocation module data) and the corrected claims calculation module (data after the interruption point is repaired), and resume the allocation process from the interruption point, instead of re-executing all steps, reducing system execution time (allocation consists of multiple cost modules and factor modules; the above is an example).

[0036] S3. Dynamic Sequence Adjustment: A configurable allocation strategy engine allows for dynamic adjustment of module execution order based on factors such as business priority and resource consumption weight. For example, when an insurance company launches a new product, the allocation strategy engine can prioritize the allocation modules related to the new product to complete cost allocation first; or during peak settlement periods at the end of the month, the module execution order can be automatically adjusted to optimize calculation efficiency.

[0037] S4. Rule Adaptation and Execution: Each module has a built-in independent cost allocation rule subsystem, supporting cost calculation based on various rules such as resource consumption ratio, business volume, and custom formulas. The system executes the cost allocation rules of each module sequentially according to the dynamically combined module sequence, and summarizes the calculation results level by level to finally achieve accurate cost allocation for the middle and back-end systems.

[0038] This invention utilizes a modular allocation method to break down a single allocation model into multiple sets of model rules, reducing a task that originally required 24 hours to 10 hours. It modularizes over 200 rules into 8 modules, further reducing system runtime and improving data output efficiency. During the operation of the modular model, it effectively addresses the following issues with significant results:

[0039] 1. If abnormal data allocation is detected midway, the abnormal data can be used to locate the running model and rules. The issue can be resolved by pausing or changing the data range without affecting the operation of other models.

[0040] 2. When adjusting or adding rules, the scope of impact and adjustment plan can be quickly located, mainly for application to the rationality of the allocation results and business expansion.

[0041] 3. When questioning or modifying certain allocation results, unlike a one-size-fits-all allocation model, a modular allocation method only requires locating the module to be reprocessed and re-executed, reducing time costs and improving the flexibility of the allocation process.

[0042] Please continue reading Figure 3 Preferably, the present invention breaks down the middle and back-office cost allocation process into the following eight cost allocation modules: payroll module, group customer performance cost allocation module, overall sales allocation module, insurance type cost allocation module, channel cost allocation module, third and fourth level cost allocation module, online and offline sales cost allocation module, and off-balance sheet performance evaluation cost allocation module. Each module is divided according to cost type and function, as described in detail below:

[0043] 1. Payroll Module: Connects to the human resources and payroll system to obtain payroll data such as salary, performance, benefits, and social security. Based on the human resources clearing situation, the costs are aggregated to the corresponding vehicle, non-vehicle, and agricultural insurance categories or detailed insurance types, as well as the customer nature of individual customers, group customers, and agricultural centers. Finally, the data is processed to the general ledger and the deferred data resource pool.

[0044] 2. Group Customer Performance Allocation Module: The module integrates the human resources and payroll system and the performance system to obtain non-vehicle group customer performance data. It matches the amounts on both sides by the organization, department, and amount fields to see if they are consistent. If they are consistent, the detailed dimension data of the performance system is used for allocation. If they are inconsistent, the payroll system data is used. Then, the sales management and budget system is used to obtain performance tracking data, and the performance is allocated to the detailed insurance policies. The detailed allocation results are processed and sent to the deferred data resource pool.

[0045] 3. Overall Sales Module: Connects to the expense control system to obtain the actual monthly reimbursement data for vehicle, non-vehicle, and agricultural insurance, and associates the detailed budget data used by the reimbursement forms, including the scope of insurance types and the level of use (self, branch, company). The allocation rules are different depending on the business attributes of the major insurance categories. The premium income is then allocated to the policy details data, and the allocation results are processed and sent to the deferred data resource pool.

[0046] 4. Insurance Allocation Module: Connects to the general ledger, expense control, and deferred resource pool data to allocate expenses to detailed insurance types according to customized allocation rules and allocation levels;

[0047] 5. Channel Allocation Module: Connects to the insurance product module's allocation results and allocates the allocation to detailed channels, customer groups, and other fields according to custom allocation rules;

[0048] 6. Online and offline sales cost allocation module: This module connects with channel allocation results to perform secondary allocation of auto insurance online and offline sales costs;

[0049] 7. Level 3 and Level 4 Allocation Module: Connects with the channel allocation module and the online sales allocation module to allocate the middle and back office expenses of the headquarters and branch offices to the front-end operations departments;

[0050] 8. Off-the-table assessment allocation module: Data that cannot be obtained from the system is processed into the data resource pool for allocation by manual entry and then allocated to the detailed dimensions according to the custom allocation rules.

[0051] This invention constructs an intelligent, accurate, and traceable enterprise-level cost allocation platform. Its core lies in employing a "rule-driven, data-layered, and progressively refined" intelligent allocation architecture. By introducing key technologies such as a dynamic allocation rule engine, a multi-dimensional data traceability mechanism, and a cost driver pool, it addresses the pain points of traditional allocation methods, including strong subjectivity, low transparency, poor flexibility, and crude accounting. The specific implementation is as follows:

[0052] 1. Peripheral Systems: Identify the peripheral systems to be connected and the data connection methods, frequencies, and data formats;

[0053] 2. Module division: Modules can be divided according to data acquisition methods, data types, allocation rules, etc.

[0054] 3. Dimension Confirmation: Determine the allocation dimensions for each module, including the source field dimension to be allocated, the allocation dimension, and other attribute dimensions, etc.

[0055] 4. Module operation: Divide the module functions and scope according to the business scenario. It is necessary to determine the interrelationship between modules. If they run independently, they will not affect each other. If there is a sequence, they need to be classified into parent modules.

[0056] 5. Execution time: After determining the functions and scope of different modules, it is necessary to determine the execution time requirements of the modules and place different modules in different scheduled tasks according to the time requirements;

[0057] 6. Modular allocation model operation and use.

[0058] Traditional cost-sharing schemes often use a fixed percentage (such as a percentage of premium income) for one-time cost-sharing, which cannot adapt to complex business scenarios and the requirements of refined management. As a preferred implementation, this invention designs and implements a configurable, computable, and nestable dynamic intelligent cost-sharing rule engine. The specific implementation is as follows:

[0059] Rule-as-Code Visual Configuration: Provides a graphical user interface (GUI) for business personnel (non-technical personnel) to define allocation rules. Rule elements (such as allocation drivers: premium income, number of policies, labor costs, working hours ratio; allocation hierarchy: company -> department -> product line -> channel) can be dragged and combined like building blocks, and the engine compiles them into executable calculation logic.

[0060] Multi-rule strategy and arbitration mechanism: Supports setting multiple alternative rules for the same allocation target (e.g., market expenses can be allocated according to "premium percentage" or "strategic weight coefficient"). The system supports setting the priority and effective conditions of rules, or arbitrating through weighted average and other methods, greatly enhancing the flexibility to meet different management needs.

[0061] Real-time calculation and simulation prediction: The engine supports a "simulated allocation" function based on historical data. Financial personnel can predict the results under different rules before allocation is executed, providing data support for management decisions and thus selecting the optimal allocation scheme. During allocation execution, the engine connects to the source system to pull and calculate data in real time.

[0062] Furthermore, considering that traditional allocation is often linear and one-dimensional, this invention further designs a hierarchical, recursive allocation model based on a cost driver pool, and abstracts and pools the cost drivers for each allocation. The specific implementation is as follows:

[0063] Deferred resource pool and driver tagging: When all deferred expenses enter the system, they not only include amount information, but are also tagged with rich metadata (such as expense type, department incurred, original voucher number, cost attribute (fixed / variable)), forming a clear deferred resource pool.

[0064] Motivation Data Pool: A motivation data pool is pre-built from the general ledger, business systems (CRM, underwriting system), human resources system, etc., based on expense reimbursement model, procurement expense model, and IT expense model. This motivation data pool centrally manages all possible allocation benchmark data (such as premiums for each type of insurance, number of policies for each channel, number of employees and working hours records for each department); and ensures that the motivation data is audited and consistent in definition, laying the foundation for accurate allocation.

[0065] Layered recursive allocation: The allocation process is not a simple sequential execution, but a process of data refinement at each layer and value chain tracing at each level. The output of the upper-level module (such as the cost allocated by insurance type) serves as the input resource for the lower-level module. At the same time, the lower-level module can call more granular driver data from the driver data pool for secondary allocation, ensuring that the cost can ultimately be traced back to the most basic business unit (such as a sales team or a specific product).

[0066] This invention solves the "black box" problem of cost allocation through end-to-end cost allocation tracing and penetrating query; the specific implementation is as follows:

[0067] Allocation Voucher Chain: For each original expense, a unique allocation serial number is generated after each allocation, and the complete allocation path (rule version, driving data, allocation amount, target object) is recorded. This ultimately forms a clear allocation tree.

[0068] Two-way traceability query: Users can query the report as follows:

[0069] Drill-down: Starting from a general management expense at headquarters, directly trace back to which channel and which type of insurance ultimately covered how much of the amount.

[0070] Roll-Up: Tracing back the costs and expenses of a frontline salesperson to their origins in specific headquarters expenses and how many times they were allocated.

[0071] This provides unprecedented transparency and credibility for cost control, profit analysis, and performance evaluation, achieving true "precise accounting".

[0072] The data in this invention is accessed from the source system and the intelligent data entry platform. After preprocessing by the deferred resource pool, it is driven by a dynamic rule engine and allocated in a hierarchical and recursive manner through insurance types, channels, online sales, and third- and fourth-level modules. Each allocation calls the driving force data pool and generates a traceable allocation voucher chain, ultimately accurately calculating the cost down to the finest granular cost object.

[0073] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be defined by the claims.

Claims

1. A modular based back office cost allocation method in insurance characterized by, Comprise the following steps: S1, module division and definition: the back office allocation process is disassembled into several independent functional modules according to the cost type or cost scene, each module corresponds to clear sub function and resource demand, and a module attribute database is established to record resource consumption parameters; S2, dynamic combination mechanism: build a module relationship network, define the data flow and dependency relationship between modules through data interface and logical rules; when the allocation process is interrupted due to data error or rule conflict, according to the position of the interruption point, extract the associated modules from the module relationship network for dynamic combination, skip the abnormal module to generate the allocation path again; S3, dynamic sequence adjustment: set a configurable allocation strategy engine to dynamically adjust the module execution order according to business priority, resource consumption weight factor; S4, rule adaptation and execution: each module has an independent allocation rule subsystem, which supports cost calculation according to resource consumption ratio, business volume and custom formula rules; according to the module sequence after dynamic combination, execute the allocation rules of each module in turn, and gradually aggregate the calculation results to finally complete the accurate allocation of back office cost.

2. The modular based underwriting back office expense allocation method of claim 1, wherein, The step S1 disassembles the back office allocation process into the following multiple independent functional modules: Salary module: interface with human resource salary system to obtain salary, performance, welfare, social security subject data, according to human clearing and settlement, collect the cost to corresponding car, non car, agricultural insurance category or detailed insurance, and individual customer, group customer, three agricultural center customer, finally process the data to general ledger and unallocated data resource pool; Group customer performance allocation module: integrate human resource salary system and performance system to obtain non car group customer performance data, match the amount on both sides through agency, department and amount fields, if consistent, use detailed dimension data of performance system for allocation, if inconsistent, use salary system data, then obtain performance follow-up data through sales management budget system, and allocate performance to detailed policy, and process the detailed allocation result to unallocated data resource pool; Coordinated real sales module: interface with fee control system to obtain car, non car and agricultural insurance actual reimbursement single data, associate reimbursement single with budget detailed data, and then allocate to policy detailed data using premium income, and process the allocation result to unallocated data resource pool; Insurance type allocation module: interface with general ledger, fee control and unallocated resource pool data, allocate to detailed insurance according to self-defined allocation rules and allocation level; Channel allocation module: interface with insurance type allocation module allocation result, allocate to detailed channel and customer group field according to self-defined allocation rules; Power grid sales allocation module: interface with channel allocation module allocation result, perform secondary allocation on car insurance power grid sales cost; Three and four level allocation module: interface with channel allocation module and power grid sales allocation module allocation result, allocate back office cost of headquarters and branch to front operation department; Off balance sheet assessment allocation module: data that cannot be obtained through system is processed to unallocated data resource pool through manual reporting, and is allocated to detailed dimension according to self-defined allocation rules.

3. The modular based underwriting back office expense allocation method of claim 1, wherein, When the data of a case is found abnormal in the loss allocation calculation and the calculation is interrupted, the step S2 directly calls the completed underwriting module data, combines the financial settlement module and the corrected loss calculation module, and recovers the allocation process from the interruption point.

4. The modular based underwriting back office expense allocation method of claim 1, wherein, The step S3 sets multiple alternative rules for the same allocation target, arbitrates the priority and effective conditions of the rules by weighted average, promotes the priority of the allocation module related to the new product by the allocation strategy engine, and preferentially completes the cost allocation; or automatically adjusts the execution order of the modules to optimize the calculation efficiency during the settlement peak at the end of the month.

5. The modular based underwriting back office expense allocation method of claim 2, wherein, The budget detailed data used by the associated reimbursement single in the overall marketing module includes the scope of the insurance type and the use level.

6. The modular based underwriting back office expense allocation method of claim 1, wherein, The step S3 includes: generating a unique allocation serial number for each original cost after each allocation, recording the complete allocation path to form an allocation tree, and realizing full-link allocation traceability and penetration query.

7. The modular based underwriting back office expense allocation method of claim 6, wherein, The allocation traceability and penetration query process in the step S3 is as follows: Downward drilling: directly penetrating and querying from a certain total management cost of the headquarters to find out which channel and which insurance type finally bear how much amount; Upward summary: reverse tracing the source of a cost and expense of a certain basic sales personnel to several headquarters costs and the number of allocations.

8. The modular based underwriting back office expense allocation method of claim 1, wherein, The module attribute database records all the costs to be allocated and sets metadata tags to form a pool of costs to be allocated, and pre-constructs a cause data pool according to the reimbursement cost model, the procurement cost model and the IT cost model to centrally manage all allocation reference data that will be used; the step S4 is executed by using a hierarchical recursive method for allocation, and the output result of the upper module is used as the input resource of the lower module; at the same time, the lower module calls the more detailed cause data in the cause data pool for secondary allocation, so that the cost is finally traced to the most terminal business unit.

9. The modular based underwriting back office expense allocation method of claim 8, wherein, The metadata tags include amount, cost type, department of occurrence, original voucher number and cost attribute; and the allocation reference data includes the premium of each insurance type, the number of policies of each channel, the number of employees of each department and the work hour record.

10. The modular based underwriting back office expense allocation method of claim 8, wherein, The reimbursement cost model interfaces the expense control to the real marketing data of the reimbursement single, divides the car, non-car and agricultural data, and allocates to the detailed insurance policy from different dimensions; the procurement cost model interfaces the detailed account data of the procurement system, replaces the total account data without details, and restores the cost before allocation; and the IT cost model selects different allocation factors for the IT related cost data types, such as system maintenance fee and hardware procurement cost.

Images