Modularized middle platform service system of enterprise-level B2B / B2C digital mall

Abstract

The application discloses a modular middle platform service system of an enterprise-level B2B / B2C digital mall, relates to the technical field of electronic commerce, and is characterized in that in the channel management module of the system, a middle variable calculation unit, a composite decision unit and an instruction output unit are formed. The composite decision unit calculates strategy instruction parameters through a composite decision model containing an impact effect function for a sudden response and a health degree adjustment function for saturation and inhibition adjustment. The application can generate strategy instructions matched with real commercial dynamics in a nonlinear manner, and improves the response capability of the system in a high-risk scenario and the overall efficiency of resource allocation.

Description

Technical Field

[0001] This invention relates to the field of e-commerce technology, and in particular to a modular middleware service system for an enterprise-level B2B2C digital marketplace. Background Technology

[0002] Modular middleware service systems for enterprise-level B2B2C digital e-commerce are currently the mainstream technical architecture. These typically include a channel management module, used to generate strategy instructions based on marketing activities and channel performance. Existing channel management modules generally employ linear response models or simple rule engines to generate these instructions. This linear model has a fundamental technical flaw when facing the complex non-linear dynamics of the real business world: the generated strategy instructions cannot match the non-linear changes in the business scenario. This flaw results in the system's inability to provide exponentially increasing resource support when facing intense market shocks. Furthermore, in daily channel management, it lacks the fine-grained ability to saturate and suppress channel performance, leading to resource misallocation and overall operational inefficiency. Summary of the Invention

[0003] This invention provides a modular middleware service system for enterprise-level B2B2C digital e-commerce, aiming to solve the technical problem of inaccurate strategy instruction generation caused by the use of a linear response model in the channel management module in the prior art.

[0004] In view of the above problems, the present invention provides a modular middleware service system for an enterprise-level B2B2C digital e-commerce platform. The system includes a marketing center module, an order center module, an inventory center module, and a channel management module. The system also includes a processor and a memory. The memory stores a computer program. When the processor runs the computer program, the channel management module is configured as follows: The intermediate variable calculation unit receives business event data from the marketing center module and channel historical performance data from the order center module, and calculates market impact intensity index and channel health index. The composite decision-making unit calculates strategy instruction parameters based on the benchmark value, the market impact intensity index, and the channel health index through a composite decision-making model. The instruction output unit encapsulates the strategy instruction parameters into a structured strategy instruction and outputs it to the inventory center module.

[0005] The technical solution provided in this application has at least the following technical effects: by introducing a decision model based on nonlinear dynamic response into its channel management module, the linear response mismatch problem of the existing technology is fundamentally solved, enabling the system to generate strategy instructions that match real business dynamics, thereby improving the system's overall responsiveness in dealing with extreme market events and the efficiency of resource allocation when making refined channel adjustments. Attached Figure Description

[0006] Figure 1 This is a schematic diagram of the structure of a modular middleware service system for an enterprise-level B2B2C digital mall, provided as an embodiment of the present invention. Detailed Implementation

[0007] The above technical solutions will now be described in detail with reference to the accompanying drawings and specific embodiments to provide a better understanding of them. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments used only to explain 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. Furthermore, it should be noted that, for ease of description, only the parts related to the present invention are shown in the drawings, not all of them.

[0008] Please see Figure 1 A modular middleware service system for an enterprise-level B2B2C digital e-commerce platform, comprising a marketing center module, an order center module, an inventory center module, and a channel management module. The system also includes a processor and a memory. The memory stores computer programs. When the processor runs the computer programs, the channel management module is configured as follows: The intermediate variable calculation unit receives business event data from the marketing center module and channel historical performance data from the order center module, and calculates market impact intensity index and channel health index. The composite decision-making unit calculates strategy instruction parameters based on the benchmark value, the market impact intensity index, and the channel health index through a composite decision-making model. The instruction output unit encapsulates the strategy instruction parameters into a structured strategy instruction and outputs it to the inventory center module.

[0009] In a specific embodiment of this invention, the system is launched within a modular middleware service system of an enterprise-level B2B2C digital marketplace. When the processor in the system runs the computer program stored in memory, an intermediate variable calculation unit is constructed within the channel management module. This intermediate variable calculation unit is responsible for transforming raw data from different business modules into quantitative intermediate variables required for subsequent decision-making.

[0010] The process begins with data reception. The intermediate variable calculation unit receives data from the Marketing Center and Order Center modules via predefined application programming interfaces (APIs) or internal message queues. The business event data received from the Marketing Center module is a JSON-formatted data message, for example, {"discount":0.7,"scaleIndicator":5}, whose data structure includes discount depth and activity scale metrics. The channel historical performance data received from the Order Center module includes voucher redemption rate and compensation occurrence frequency.

[0011] Upon receiving business event data, the intermediate variable calculation unit initiates the quantification process for the market impact intensity indicator (I). Quantification is accomplished using a weighted combination formula: The DiscountDepth in this formula is calculated from the discount data in the business event data. For example, when the discount is 0.7, The `log()` function is used to model the marginal effect of `DiscountDepth`. `ScaleIndicator` directly uses the activity scale indicator from business event data. Weighting factors. and It is a floating-point number pre-configured in the system.

[0012] After receiving historical channel performance data, the intermediate variable calculation unit initiates the quantification process for the channel health index (H). Quantification is accomplished using an asymmetric weighted combination formula: The VoucherUsageRate and CompensationFrequency in this formula directly use historical channel performance data. The weighting factors α and β are pre-configured floating-point numbers in the system, with β being greater than α, to reflect in the model that negative behavior damages channel trust more than positive behavior brings gains.

[0013] The outputs of this stage of the process are two numerical indicators generated by the intermediate variable calculation unit: the market impact intensity indicator (I) and the channel health indicator (H). These two numerical indicators are then passed to the composite decision-making unit to initiate the core decision-making process of the next stage.

[0014] In the previous stage of the process, the intermediate variable calculation unit generated two numerical indicators: the market impact intensity index (I) and the channel health index (H). These two numerical indicators were then passed to the composite decision-making unit. The composite decision-making unit is the core link in the strategy instruction generation process; all its work revolves around applying a composite decision-making model to transform the input intermediate variables into the final strategy instruction parameters.

[0015] Upon receiving the market impact intensity index (I) and channel health index (H), the composite decision-making unit immediately invokes an internally fixed composite decision-making model. This model is designed to include two independent functional components: the first is an impact effect function for abrupt responses, and the second is a health adjustment function for saturation and inhibition regulation.

[0016] Next, the composite decision-making unit first inputs the market shock intensity index (I) into the shock effect function for calculation. This shock effect function is a piecewise exponential function with a threshold. The function works as follows: a floating-point number is pre-configured in the system as a threshold, for example, Threshold=3. When the input market shock intensity index (I) is less than or equal to this threshold, the function uses a linear formula. Perform calculations, where This is a preset linear coefficient. When the input market shock intensity indicator (I) exceeds this threshold, the function switches to an exponential formula. Perform calculations, where It is a preset exponential coefficient, and exp() is the natural exponential function.

[0017] To illustrate this calculation process, let's assume the system is configured as follows: , Threshold=3. If the market shock intensity index I calculated in the previous stage is 2.5, since 2.5<=3, the function output result is: If the market shock intensity index I calculated in the previous stage is 4.416, since 4.416 > 3, the function switches to exponential mode, and the output result is... .

[0018] After calculating the impact effect, the composite decision-making unit continues to input another intermediate variable, namely the channel health index (H), into the second functional component of the composite decision-making model, the health adjustment function. This health adjustment function is an S-shaped curve function, which maps the linear channel health index to a non-linear adjustment factor with saturation and inhibition ranges.

[0019] In this embodiment, the health adjustment function is given by the hyperbolic tangent function (tanh), and its specific formula is as follows: In this formula, H is the input channel health metric. γ and These are parameters pre-configured in the system, where γ controls the steepness of the curve. This is the offset of the curve's center point. The output range of the hyperbolic tangent function is between -1 and 1. The transformation ultimately smoothly constrains the output adjustment factor between 0 and 1.

[0020] To illustrate this calculation process, let's assume the system is configured with γ=2.5. =0. If the channel health index H calculated in the previous stage is 0.83, then the function output result is The value of tanh(2.075) is approximately 0.969, therefore the final adjustment factor value is approximately... When the input channel health index H value is very high or very low, the output of this function will approach 1 or 0 respectively, thus creating a saturation and inhibition effect.

[0021] After calculating the impact effect and health adjustment results separately, the composite decision-making unit proceeds to the final step: generating the strategy instruction parameters. This process uses a multiplicative combination formula to combine an externally input baseline value (BaseSales) with the results calculated in the previous two steps. BaseSales is a value representing historical baseline sales. The final strategy instruction parameters ( ) is calculated using the following formula: .

[0022] Combining all the previous numerical examples: assuming BaseSales = 1000 and the market shock intensity index I = 4.416, the following calculations were performed. The channel health index H was calculated when it was 0.83. Substituting these values ​​into the final formula, the strategy command parameters are calculated. The calculated value This refers to the strategy instruction parameters ultimately generated by the composite decision-making unit and used by downstream modules for execution.

[0023] After the composite decision-making unit completes the calculation of the strategy instruction parameters, the output of this stage of the process is a purely numerical parameter, such as... This numerical parameter is then passed to the instruction output unit. The instruction output unit is the final stage in the strategy instruction generation process, and its entire work revolves around transforming the calculated numerical parameter into a structured instruction that can be executed by downstream modules.

[0024] The instruction output unit receives the policy instruction parameters. Next, the structured encapsulation process for the strategy instruction is initiated. The purpose of this process is to package a context-free numerical value into a machine-readable data structure containing contextual information such as the action to be executed and the target object. In this embodiment, encapsulation is accomplished by constructing a JSON object. The instruction output unit then outputs the received numerical parameters... As the value of the quantity field, other key fields such as action, channelId, and productId are obtained from the upstream process or supplemented according to the system configuration.

[0025] To illustrate this encapsulation process, the instruction output unit will... Combined with context information, a JSON data message in the following format is generated: { "action":"castVoucher", "channelId":"CID-001", "productId":"PID-XYZ", "quantity":2983 } This JSON data message is the final, complete structured strategy instruction.

[0026] After encapsulating the structured strategy instructions, the instruction output unit performs the final step: outputting the encapsulated instructions to downstream modules. This output is accomplished through the system's internal microservice communication mechanism. In this embodiment, the instruction output unit uses the generated JSON data packet as the message body and publishes it through a predefined message queue topic, such as `voucher-casting-requests`. The system's inventory center module, as the consumer of this message queue topic, listens for and receives this message.

[0027] Upon receiving this structured strategy instruction, the inventory center module parses the JSON data message to obtain information such as the defined execution action, target product, and quantity, and then performs subsequent resource allocation operations accordingly. This completes the entire workflow from raw data input to final instruction execution.

[0028] It should be noted that several parameters involved in the aforementioned calculation process, including weighting factors, function shape parameters, and baseline values, are stored in a separate configuration module or database table during system runtime for use by the calculation unit. The determination of these values ​​involves two stages: initial setting and periodic optimization. In the initial setting stage, a set of initial baseline values ​​is obtained by performing statistical regression analysis on historical business data. In the periodic optimization stage, the system can further configure a machine learning feedback loop. This loop compares the model's predictions (e.g., estimated voucher distribution volume) with actual business results (e.g., final sales revenue and compensation occurrence rate) and uses optimization algorithms (e.g., gradient descent or particle swarm optimization) to iteratively adjust the aforementioned weighting factors and function shape parameters, so that the final business results approach a preset business objective. The baseline values ​​are obtained by periodically querying and aggregating historical transaction data from the order center.

[0029] In another alternative implementation, the health adjustment function can also be implemented using the Sigmoid function, which can also achieve the adjustment effect of an S-shaped curve.

[0030] The embodiments described above are only some embodiments of the present invention, and not all embodiments. It should be noted that the present invention is not limited to the embodiments described above. Any modifications, equivalent substitutions, variations, or improvements made by those skilled in the art based on the above embodiments without creative effort, within the scope of the concept and principles of the present invention, should fall within the protection scope claimed by the claims of the present invention.

Claims

1. A modular middleware service system for an enterprise-level B2B2C digital e-commerce platform, the system comprising a marketing center module, an order center module, an inventory center module, and a channel management module, characterized in that, The system also includes a processor and a memory, the memory storing a computer program, and when the processor runs the computer program, the channel management module is configured as follows: The intermediate variable calculation unit receives business event data from the marketing center module and channel historical performance data from the order center module, and calculates market impact intensity index and channel health index. The composite decision-making unit calculates strategy instruction parameters based on the benchmark value, the market impact intensity index, and the channel health index through a composite decision-making model. The instruction output unit encapsulates the strategy instruction parameters into a structured strategy instruction and outputs it to the inventory center module. The composite decision-making model includes: Impact effect function for sudden response; A health regulation function used for saturation and inhibition regulation; The process by which the composite decision-making unit calculates the strategy instruction parameters includes: multiplying and combining the benchmark value, the result of the impact effect function, and the result of the health adjustment function; The composite decision-making unit calculates the strategy instruction parameters using the following formula. : Where BaseSales is the benchmark value, I is the market impact intensity index, and H is the channel health index. It is the impact effect function, It is the health adjustment function; The impact effect function It is a piecewise exponential function with a threshold, and the piecewise exponential function switches to exponential growth mode when the market shock intensity index is higher than the threshold; The health regulation function is an S-shaped curve function, which is either a hyperbolic tangent function or a sigmoid function, to achieve the saturation and inhibition effects of channel regulation.

2. The modular middleware service system for an enterprise-level B2B2C digital marketplace according to claim 1, characterized in that, The process by which the intermediate variable calculation unit calculates the market impact intensity index includes: weighting the logarithm of the discount depth and the activity scale index in the business event data.

3. A modular middleware service system for an enterprise-level B2B2C digital marketplace according to claim 1 or 2, characterized in that, The process by which the intermediate variable calculation unit calculates the channel health index includes: performing an asymmetric weighted combination of the voucher redemption rate and compensation occurrence frequency in the channel's historical performance data.

4. The modular middleware service system for an enterprise-level B2B2C digital marketplace according to claim 1, characterized in that, The health status adjustment function is given by the hyperbolic tangent function: Where H is the input channel health metric, and These are parameters pre-configured in the system, among which Control the steepness of the curve, It is the offset of the center point of the curve.

Images