Rule-based automotive product resource scheduling methods, devices, and electronic equipment

Abstract

This disclosure presents a method, apparatus, and electronic device for scheduling automotive product resources based on a rule engine. One specific implementation of the method includes: matching automotive product configuration information with configuration rules to obtain target automotive product configuration information; extracting configuration elements from the target automotive product configuration information and combining them for business purposes to obtain automotive product template configuration information; filling the automotive product template configuration information with configuration parameters to obtain automotive product information; performing value decoupling matching on the automotive product information and then determining product value to obtain automotive product value information; dynamically rendering and binding the product configuration interface to the page to obtain an automotive product configuration interface; and executing automotive resource scheduling operations. This implementation can achieve flexible configuration of automotive products, shorten the response time of automotive product configuration information, improve the personalized display and display quality of the configuration page, and reduce the waste of automotive resources.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to a method, apparatus, and electronic device for scheduling automotive product resources based on a rule engine. Background Technology

[0002] Page configuration display is a technology that dynamically adjusts the display of a page based on continuously updated rule information for automotive products. For automotive product-based page configuration display, the typical approach is to use fixed product configuration fields and hard-coding techniques to configure the page based on the acquired automotive product rule configuration information set, resulting in and displaying the automotive product configuration page, and then scheduling automotive resources based on the automotive product configuration page.

[0003] However, in practice, it has been found that when configuring and displaying pages based on automotive products using the above method, the following technical problems often arise: Due to the fixed and hard-coded technology of product configuration field information, when new or modified product configuration field information appears in the automotive product rule configuration information, a large amount of corresponding underlying code needs to be modified, resulting in a long development and modification cycle for page configuration display and difficulty in adapting to dynamically changing rule configuration information; In addition, the underlying encoding of the automotive product rule configuration information set is a set of underlying code for each configuration information, and the rule configuration and value determination are tightly coupled, with a large amount of identical underlying code, resulting in code redundancy and high code maintenance costs. When updating the automotive product rule configuration information, it is necessary to traverse and modify all underlying code, resulting in low system stability, difficulty in responding to changes in a timely manner, and consequently, inability to schedule automotive resources in a timely and accurate manner, leading to waste and stockpiling of automotive resources.

[0004] The information disclosed in this background section is only intended to enhance the understanding of the background of the present disclosure concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of this disclosure propose a method, apparatus, and electronic device for scheduling automotive product resources based on a rule engine, in order to solve one or more of the technical problems mentioned in the background section above.

[0007] In a first aspect, some embodiments of this disclosure provide a method for scheduling automotive product resources based on a rule engine, comprising: responding to the detection of a resource call request for an automotive product configuration page, using an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain target automotive product configuration information; extracting configuration elements from the target automotive product configuration information to obtain an automotive product atomic configuration information set; performing business combination on the automotive product atomic configuration information set to obtain automotive product template configuration information; filling configuration parameters into the automotive product template configuration information to obtain automotive product information; using a product value determination rule engine to perform value decoupling matching processing on the automotive product information to obtain target product value determination information; dynamically calling a pre-defined product value determination algorithm library to perform product value determination processing based on the target product value determination information to obtain automotive product value information; dynamically rendering and binding a page to the product configuration terminal based on the automotive product information and the automotive product value information to obtain an automotive product configuration page; and controlling a resource control terminal to perform resource scheduling operations based on the automotive product configuration page, wherein the resource scheduling operations include at least one of the following: transportation of automotive parts and transportation of complete vehicles.

[0008] Secondly, some embodiments of this disclosure provide a rule engine-based automotive product resource scheduling device, comprising: a configuration rule matching unit configured to, in response to detecting a resource call request for an automotive product configuration page, use an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain target automotive product configuration information; a configuration element extraction unit configured to extract configuration elements from the target automotive product configuration information to obtain an automotive product atomic configuration information set; a business combination unit configured to perform business combination on the automotive product atomic configuration information set to obtain automotive product template configuration information; and a configuration parameter filling unit configured to fill configuration parameters into the automotive product template configuration information. The system obtains automotive product information; a value decoupling and matching unit is configured to use a product value determination rule engine to perform value decoupling and matching processing on the aforementioned automotive product information to obtain target product value determination information; a combination unit is configured to dynamically call a pre-defined product value determination algorithm library to perform product value determination processing based on the aforementioned target product value determination information to obtain automotive product value information; and a control unit is configured to dynamically render and bind a page to a product configuration terminal based on the aforementioned automotive product information and the aforementioned automotive product value information to obtain an automotive product configuration page, and to control a resource control terminal to perform resource scheduling operations based on the aforementioned automotive product configuration page, wherein the aforementioned resource scheduling operations include at least one of the following: transportation of automotive parts and transportation of complete vehicles.

[0009] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0010] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0011] The above embodiments of this disclosure have the following beneficial effects: The rule engine-based automotive product resource scheduling method of some embodiments of this disclosure can realize flexible configuration of automotive products through the collaborative processing of dual rule engines and the three-layer architecture of atom-template-product, realize the decoupling of configuration information and value determination, shorten the response time of automotive product configuration information, reduce the workload of configuration information maintenance and adjustment, improve the personalized display and display quality of configuration page, and reduce the waste of automotive resources. Specifically, the reasons for the long development and modification cycle of related page configuration displays, the difficulty in adapting to dynamically changing rule configuration information, the low system stability, and the inability to respond to changes in a timely manner are as follows: Due to the fixed and hard-coded technology of product configuration field information, when new or modified product configuration field information appears in the automotive product rule configuration information, a large amount of corresponding underlying code needs to be modified, resulting in a long development and modification cycle for page configuration displays and difficulty in adapting to dynamically changing rule configuration information. In addition, the underlying encoding of the automotive product rule configuration information set is a set of underlying code for each configuration information, and the rule configuration and value determination are tightly coupled, with a large amount of identical underlying code, resulting in code redundancy and high code maintenance costs. When updating the automotive product rule configuration information, all underlying code needs to be traversed and modified, resulting in low system stability and difficulty in responding to changes in a timely manner. Consequently, automotive resources cannot be scheduled in a timely and accurate manner, resulting in waste and stockpiling of automotive resources. Based on this, some embodiments of the rule engine-based automotive product resource scheduling method of this disclosure can first respond to the detection of a resource call request for the automotive product configuration page, and use the automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the above resource call request to obtain the target automotive product configuration information. Here, the accuracy of configuration rule matching can be achieved by utilizing an automotive product configuration rule engine. Secondly, configuration elements are extracted from the aforementioned target automotive product configuration information to obtain an atomic configuration information set. Here, the atomic configuration information represents the smallest configuration element unit, supporting independent creation, modification, and version management. This breaks the rigid limitations of hard-coding technology, providing a flexible foundational component for configuring automotive product information and improving the flexibility of configuration and modification. Thirdly, the aforementioned atomic configuration information set is combined using business logic to obtain automotive product template configuration information. Here, flexible configuration of the atomic configuration information set enables the reuse and standardization of atomic configuration information, reducing repetitive configuration work and improving configuration efficiency. Next, configuration parameters are populated into the aforementioned automotive product template configuration information to obtain automotive product information. Here, configuring automotive product information does not require redefining configuration elements; only configuration parameters need to be filled in. This improves the efficiency and personalization of configuring automotive product information, avoids extensive modifications to the underlying code, and achieves flexible configuration through a layered decoupled configuration architecture of atomic-template-product.Subsequently, using a product value determination rule engine, the aforementioned automotive product information is subjected to value decoupling and matching processing to obtain the target product value determination information. Here, the decoupled and independent configuration of product value determination and product configuration information is achieved, resolving configuration redundancy issues. If a problem arises with either the configuration information or the product value determination, only the corresponding part needs to be modified, without altering the entire code, improving modification efficiency and reducing workload. Then, based on the target product value determination information, a pre-defined product value determination algorithm library is dynamically invoked to perform product value determination processing, obtaining the automotive product value information. Here, flexible configuration calculation and independent iteration of automotive product value information can be achieved. Finally, based on the aforementioned automotive product information and automotive product value information, the product configuration end is dynamically rendered and bound to a page, resulting in an automotive product configuration page. Based on the automotive product configuration page, the resource control terminal executes resource scheduling operations, which include at least one of the following: transportation of automotive parts and transportation of the complete vehicle. Here, personalized and adaptive configuration of the automotive product configuration page can be achieved. Through a precise automotive product configuration page, accurate scheduling of automotive resources can be realized, reducing waste of automotive resources. Therefore, this rule engine-based automotive product resource scheduling method, through the collaborative processing of dual rule engines and a three-layer architecture of atoms-templates-products, can achieve flexible configuration of automotive products, decouple configuration information from value determination, shorten the response time of automotive product configuration information, reduce the workload of configuration information maintenance and adjustment, improve the personalized display and display quality of configuration pages, and reduce the waste of automotive resources. Attached Figure Description

[0012] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0013] Figure 1 This is a flowchart of some embodiments of the rule engine-based automotive product resource scheduling method according to this disclosure;

[0014] Figure 2 This is a schematic diagram of the structure of some embodiments of the rule engine-based automotive product resource scheduling device according to the present disclosure;

[0015] Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0016] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0017] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0018] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0019] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0020] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0021] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] Figure 1 A flow 100 of some embodiments of the rule-engine-based vehicle product resource scheduling method according to this disclosure is shown. The rule-engine-based vehicle product resource scheduling method includes the following steps:

[0023] Step 101: In response to the detection of a resource call request for the vehicle product configuration page, the vehicle product configuration rule engine is used to match the configuration rules of the vehicle product configuration information corresponding to the resource call request to obtain the target vehicle product configuration information.

[0024] In some embodiments, the execution entity (e.g., an electronic device) of the above-described rule engine-based automotive product resource scheduling method can, in response to detecting a resource call request for an automotive product configuration page, utilize an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain the target automotive product configuration information. The automotive product configuration page can be a page used to display configuration information related to automotive products (auto finance products). The related configuration information can be application information in a car rental system. The application information can be application materials submitted by a user (e.g., a consumer or a dealer) regarding a leased automotive product. The application materials can include, but are not limited to, at least one of the following: user identity information, automotive product model, configuration and value, rental information (e.g., lease term, down payment ratio, monthly payment amount), and rental channel information (e.g., rental through a car dealer, online platform, or authorized car store). The resource call request can be a scheduling request for resources related to the automotive industry (e.g., automotive parts, vehicle transportation resources). The automotive product configuration information can be application information related to automotive products. The aforementioned automotive products can be payment methods and purchase discounts tailored to users with insufficient funds when purchasing or leasing a car. The aforementioned target automotive product configuration information can be the configuration rule information that best matches the aforementioned automotive product configuration information from the configuration rule information included in the aforementioned automotive product configuration rule engine. The aforementioned automotive product configuration rule engine can be a rule engine used to determine and output the automotive product that best matches the input automotive product configuration information. The aforementioned automotive product configuration rule engine can perform configuration rule matching through the following steps: using the RETE algorithm, performing pattern matching between the stored set of configuration rule information related to the automotive product configuration information and the aforementioned automotive product configuration information to obtain a set of configuration rule information that matches the automotive product configuration information. Secondly, in response to the determination that the configuration rule information set includes multiple configuration rule information, selecting the configuration rule information with the highest priority from the aforementioned configuration rule information set as the target configuration rule information according to the rule execution priority.

[0025] In some optional implementations of certain embodiments, the above-mentioned use of an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain the target automotive product configuration information may include the following steps:

[0026] The first step is to obtain a multi-source product association dataset associated with the initial automotive product configuration information set corresponding to the aforementioned resource call request. This multi-source product association data includes: configuration object information, configuration device information, configuration timing behavior information, and object association information. The multi-source product association data in this dataset can be multimodal data related to the initial automotive product configuration information. The initial automotive product configuration information set can be multiple automotive product configuration information received by the product configuration terminal. The configuration object information can be information representing the identity of the configuration object sending the automotive product configuration information. For example, the configuration object information can include, but is not limited to, at least one of the following: personal / enterprise application forms, transfer information (transaction information), credit data, and ownership value attribute values ​​(asset and liability information). The configuration device information can be information about the device used by the configuration object to send the automotive product configuration information. The configuration device information can include, but is not limited to, at least one of the following: terminal device fingerprint information, IP (Internet Protocol) information, and behavioral trajectory information of the operating device. The configuration timing behavior information can be the behavioral information of the configuration object. The aforementioned configuration time-series behavioral information may include, but is not limited to, at least one of the following: recent application frequency, cross-platform application record information, value transfer time-series information (fund transfer time-series information), and consumption / business behavior time-series information. The aforementioned object association information may be information about objects that are associated with the configured object. The aforementioned object association information may include, but is not limited to, at least one of the following: the configured object's social association information, device association and binding information, vehicle product context information, and federal risk-sharing information.

[0027] The second step involves constructing a behavioral graph from the aforementioned multi-source product association dataset to obtain a heterogeneous graph of object behavioral structures. This heterogeneous graph can be a graph representing the relationships and attributes of configuration objects, with configuration objects as core nodes and configuration object information, configuration device information, configuration time-series behavioral information, and object association information as related nodes. In practice, the executing entity can first perform data cleaning and privacy desensitization on the aforementioned multi-source product association dataset to obtain a de-identified multi-source product association dataset. This privacy desensitization can be performed using a differential privacy desensitization method. Then, a data fusion algorithm based on a weighted average method is used to perform multi-source fusion on the de-identified multi-source product association dataset to obtain a fused multi-source product association dataset. Finally, a knowledge graph is constructed from the fused multi-source product association dataset to obtain the heterogeneous graph of object behavioral structures.

[0028] The third step involves extracting cross-modal features from the aforementioned heterogeneous graph of object behavior structure, resulting in a cross-modal graph behavior feature vector set. The cross-modal graph behavior feature vectors in this set can represent the identity information, temporal behavior information, association information, and federated feature information of the configured objects. This cross-modal feature extraction can be performed using a graph convolutional neural network model.

[0029] The fourth step involves performing risk causality tracing identification on the heterogeneous behavioral structure graph of the aforementioned object based on the cross-modal graph behavioral feature vector set, thereby obtaining risk causality tracing identification information. This risk causality tracing identification information can represent the structural causal chain between the behavioral information and risk information of the configured object. This risk causality tracing identification can trace the behavioral chain back before the configured object exhibits risky behavior, identifying key nodes that cause the risk and allowing for early intervention in risk events. As an example, the executing entity can utilize the PSM (Propensity Score Matching) algorithm to perform risk causality tracing identification on the heterogeneous behavioral structure graph of the aforementioned object based on the cross-modal graph behavioral feature vector set, thereby obtaining risk causality tracing identification information.

[0030] The fifth step involves performing multi-dimensional risk identification on the heterogeneous graph of the object's behavioral structure based on the aforementioned cross-modal graph behavioral feature vector set, thereby obtaining a risk information set for the configured object. This risk information set can include information on credit risk, fraud risk, and time-series behavioral risk for a single entity, as well as group fraud risk for a group. For example, the executing entity can first input the aforementioned cross-modal graph behavioral feature vector set into the risk identification big model to obtain credit risk identification information and fraud risk identification information for a single entity. This risk identification big model can be based on a general big model (e.g., GLM-4 (General Language Model 4, the fourth generation of the Zhipu series of big language models), LLaMA 3 (Large Language Model Meta AI 3)) as a risk control domain model, undergoing deep fine-tuning and adversarial training, and incorporating a risk control knowledge base, historical risk cases, and regulatory rules to achieve the big model's reasoning ability regarding risk control. Then, the aforementioned heterogeneous graph of object behavior structure is input into heterogeneous graph neural networks (HGNNs) to obtain group fraud risk identification information for the group. Next, the aforementioned cross-modal graph behavior feature vector set is input into a time-series risk identification model to obtain time-series behavior risk identification information. This time-series risk identification model can be a model that sequentially inputs the aforementioned cross-modal graph behavior feature vector set into a long short-term memory neural network model, then inputs the resulting model into a self-attention mechanism to output time-series behavior risk identification information, thereby mining anomalous information in time-series behavior. Finally, the aforementioned credit risk identification information, fraud risk identification information, group fraud risk identification information, and time-series behavior risk identification information are weighted and aggregated to obtain a set of configuration object risk information.

[0031] Step 6: Based on the aforementioned risk causal tracing identification information and the aforementioned configuration object risk information set, perform risk interception control on the aforementioned initial automotive product configuration information set to obtain the automotive product configuration information. In practice, the aforementioned executing entity can first determine the initial automotive product configuration information set corresponding to the aforementioned risk causal tracing identification information and the aforementioned configuration object risk information set as the risk configuration information set. Then, control the product configuration terminal to intercept and control the aforementioned risk configuration information set, and receive the initial automotive product configuration information set after removing the risk configuration information set to obtain the automotive product configuration information.

[0032] The seventh step involves using an automotive product configuration rule engine to match the above automotive product configuration information with configuration rules, thereby obtaining the target automotive product configuration information.

[0033] Optionally, the above-mentioned use of an automotive product configuration rule engine to match the automotive product configuration information corresponding to the resource call request with configuration rules to obtain the target automotive product configuration information may include the following steps:

[0034] The first step is to perform multi-dimensional semantic extraction on the aforementioned automotive product configuration information to obtain a configuration multi-dimensional semantic information set. The configuration multi-dimensional semantic information in this set can be the semantic information corresponding to the matching conditions that need to be matched within the aforementioned automotive product configuration information. This configuration multi-dimensional semantic information set may include, but is not limited to, at least one of the following: automotive product model, user level information, component import channel information, matching priority information, rule number, and rule matching weight.

[0035] The second step involves performing condition bitmasking and feature encoding on the aforementioned multidimensional semantic information set and the configuration rule information set included in the automotive product configuration rule engine to obtain configuration bit operation feature values ​​and a configuration semantic feature vector set. The configuration bit operation feature values ​​can be represented as feature values ​​of the multidimensional semantic information set in the form of a bitmap. These feature values ​​can be those where the value is 1 at the corresponding bitmap position and 0 at the remaining positions. The number of bits in the configuration bit operation feature values ​​can be the number of semantic information corresponding to all conditions included in the automotive product configuration rule engine. The configuration semantic feature vectors in the configuration semantic feature vector set can represent the semantic information of the multidimensional semantic information. In practice, the executing entity can first perform atomic decomposition on the configuration rule information set included in the automotive product configuration rule engine to obtain a set of rule atomic condition information. The rule atomic condition information in this set can be the smallest indivisible logical unit that can be decomposed from the condition information in the preset configuration rule information. The rule atomic condition information can be "city is Shanghai" or "age is greater than 18 years old". Secondly, bit-mapped processing is performed on the aforementioned set of rule atomic condition information to obtain a rule bitmap. This rule bitmap can be a bitmap that assigns a unique index number to the set of rule atomic condition information, starting from 0, and defines the position of a bit corresponding to that index number. Then, bitmap initialization and feature encoding processing are performed on the aforementioned multidimensional semantic information set to obtain an initial configuration bitmap and a configuration semantic feature vector set. This feature encoding processing can be based on a feature dictionary. The initial configuration bitmap can be a binary number where all bits are 0. Finally, the initial configuration bitmap and the rule bitmap are matched to obtain configuration bitwise operation feature values. This matching can be a process where if there is information in the multidimensional semantic information set that matches the set of rule atomic condition information, the corresponding position is set to 1.

[0036] The third step involves performing decision tree matching between the aforementioned configuration semantic feature vector set and the corresponding automotive product configuration decision tree of the aforementioned configuration rule information set, resulting in a matching decision node set. The matching decision nodes in this set can be leaf nodes that match the aforementioned configuration semantic feature vector set, i.e., preset configuration rule information. The aforementioned automotive product configuration decision tree can be a decision tree where the root and intermediate nodes are selected from the preset configuration rule information set, which has the highest discriminative power and most frequently appears in the conditions, and the structured configuration rule information is added to the leaf nodes. This automotive product configuration decision tree is obtained through the following steps:

[0037] Sub-step 1 involves performing rule structure parsing on the aforementioned configuration rule information set to obtain a structured configuration rule information set. The structured configuration rule information in this set can be information that represents configuration rule information in a structured form. For example, the structured configuration rule information may include: the product type of the automotive product, user level information, user basic information, rule number, rule priority, rule weight, rule condition list (condition information of the configuration rule information), and structured information of the rule information configuration results (product parameters, rates, terms, etc. of the financial product returned after matching).

[0038] Sub-step 2: Generate an initial configuration decision tree based on the aforementioned structured information set of configuration rules. This initial configuration decision tree can use nodes to represent structured information about configuration rules, and directed connections to represent the inclusion relationship, preconditions, and post-execution relationships between two related structured information sets of configuration rules. For example, the root node of the initial configuration decision tree can be the most frequently used and highly distinguishable structured information set of configuration rules. For example, the root node could be the product type of the car product, user level information, or information on how the car product is acquired. The intermediate nodes of the initial configuration decision tree can be nodes used to represent intermediate judgment conditions in the configuration rule information filtering process. For example, the intermediate nodes could be user age or user income level. The leaf nodes of the initial configuration decision tree can be nodes representing the results of rule matching. For example, the leaf nodes can include, but are not limited to, at least one of the following: rule number, rule priority, rule condition list, and configuration result information. As an example, the executing entity can first use the ID3 (Iterative Dichotomiser 3) algorithm to determine the information gain value set of the aforementioned structured information set of configuration rules. Then, the configuration rule structured information with the largest value is selected from the set of information gain values ​​as the splitting attribute of the current node, and the decision tree is recursively constructed to obtain the initial configuration decision tree.

[0039] Sub-step 3 involves performing bit mapping on the initial configuration decision tree to obtain an automotive product configuration decision tree. This automotive product configuration decision tree includes a set of rule bit masks. The rule bit masks in this set can be masks representing each node in the initial configuration decision tree in the form of a bitmap. When the bitmap exceeds 64 bits, it is expanded using block masks. The bit mapping can be a bit mapping where bit positions are assigned to the rule information corresponding to each node, and the values ​​of the corresponding satisfying bit positions are set to 1.

[0040] The fourth step involves performing bitwise operation matching on the aforementioned configuration bitwise operation feature values ​​and the rule bitmask set corresponding to the aforementioned matching decision node set to obtain a matching product configuration rule information set. The matching product configuration rule information in this set can be preset configuration rule information that satisfies the aforementioned configuration bitwise operation feature values. The bitwise operation matching can be performed using a bitwise AND operation.

[0041] The fifth step involves sorting the aforementioned set of matching product configuration rules to obtain the target vehicle product configuration information. This target vehicle product configuration information can be the most matching rule information within the set of matching product configuration rules. In practice, the executing entity can first determine the priority and weighted sum of the matching product configuration rule information set to obtain a rule matching value set. Then, the matching product configuration rule information corresponding to the largest value in the rule matching value set is determined as the target vehicle product configuration information. Thus, by using conditional bitmasking, the vehicle product configuration information is only executed once in the feature encoding stage. In subsequent matching processes, conditional judgments are no longer needed; instead, bitwise operations are performed directly on the configuration bitwise feature values. This transforms complex rule matching into bitwise operations, improving matching efficiency. Furthermore, decision tree matching based on decision trees avoids full rule traversal matching, further improving matching efficiency and accuracy.

[0042] In addressing the aforementioned technical challenges in the application scenario—dynamic scheduling of vehicle resources after accurate matching of heterogeneous and complex automotive product configuration information in an industrial setting—the following technical problems often arise: Because the automotive product configuration rule engine in industrial settings contains a large amount of configuration rule information, and the configuration information to be matched is personalized and has diverse data formats, configuration rule matching requires a full match with every rule in the engine. This results in low matching efficiency and accuracy, low matching response performance, and consequently, low accuracy in scheduling vehicle resources based on the target automotive product configuration information, leading to wasted vehicle resources and prolonged scheduling time. Considering the following requirements for this application scenario: adaptability to high-complexity configuration rules and large data volumes, adaptability to high-precision rule matching, adaptability to collaborative matching based on multiple networks, adaptability to fuzzy dynamic matching, and adaptation to triplet structured processing, we have decided to adopt the following solution:

[0043] In some optional implementations of certain embodiments, the above-mentioned use of an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain target automotive product configuration information, and controlling the resource control terminal to perform resource scheduling operations based on the target automotive product configuration information, may include the following steps:

[0044] The first step involves performing ternary structuring on the aforementioned automotive product configuration information and the configuration rule information set included in the automotive product configuration rule engine, resulting in a product configuration ternary set and a rule configuration ternary set. The product configuration ternary sets in the product configuration ternary set can represent automotive product configuration information in the form of a configuration object, a relation, and a configuration attribute value. For example, the product configuration ternary set could be {configuration object, qualification level, A-level}. The rule configuration ternary sets in the rule configuration ternary set can represent configuration rule information in the form of configuration rule antecedent and configuration rule consequent. For example, the configuration rule information could be "If the configuration object's qualification is A-level and the credit limit is between 500,000 and 1,000,000, then configure the standard fee rate and 12-month repayment method," then the rule configuration ternary set could be [antecedent {(qualification level: A-level), (credit limit range: 500,000-1,000,000)}, consequent {(fee rate type: standard), (repayment period: 12 months)}].

[0045] The second step is to generate a conditional sharing degree model based on the aforementioned rule configuration triplet set. This conditional sharing degree model can be a mathematical model used to quantify the sharing degree of each rule configuration triplet located in the antecedent within the configuration rule information set, and to guide the sorting of the rule configuration triplet set. In practice, the executing entity can calculate the antecedent triplet sharing degree by calculating the ratio of the number of times each rule configuration located in the antecedent within the aforementioned rule configuration triplet set appears in the aforementioned configuration rule information set to the number of configuration rule information items included in the configuration rule information set. Then, based on the aforementioned antecedent triplet sharing degree, the aforementioned rule configuration triplet set is sorted to obtain the sorted antecedent rule configuration triplet set, which serves as the conditional sharing degree model.

[0046] The third step involves constructing an initial single-input matching network based on the aforementioned conditional sharing model. This initial single-input matching network can be an Alpha network where the rule configuration triplet with the highest antecedent triplet sharing degree serves as the top-level node, and the triplet with the next highest sharing degree serves as the lower-level node. This hierarchical storage of the antecedent rule configuration triplets reduces redundant matching operations. Each node in this initial single-input matching network stores its corresponding rule configuration triplet and initializes Alpha memory to cache successfully matched antecedent triplet instances. In practice, the execution entity utilizes the Alpha network construction method in the Rete algorithm to construct the initial single-input matching network based on the aforementioned conditional sharing model.

[0047] The fourth step is to generate a matching fitness function for the initial single-input matching network described above. This matching fitness function can be a multi-objective fitness function that balances rule matching efficiency (rule matching time) and rule matching accuracy (the percentage of effective matching rules).

[0048] Fifth, based on the aforementioned matching fitness function, the initial single-input matching network is genetically updated to obtain the updated single-input matching network. In practice, the execution entity can utilize a genetic algorithm to genetically update the initial single-input matching network based on the aforementioned matching fitness function to obtain the updated single-input matching network. The initialization parameters of the genetic algorithm may include: an encoding length of 20, a population size of 130, a crossover probability of 75%, a mutation probability of 6%, using the sorted antecedent rule configuration triplet set as the initial population individuals, and the population evaluation function being the matching fitness function. The updated single-input matching network effectively solves the problem of the initial single-input matching network easily getting trapped in local optima.

[0049] Step 6: Based on the updated single-input matching network and the configuration rule information set, generate a multi-input matching network. This multi-input matching network can be a Beta network used to process and connect multiple facts output by the updated single-input matching network to determine whether the conditions of the complete configuration rule information are met. In practice, the executing entity can utilize the Beta network construction method in the Rete algorithm to generate the multi-input matching network based on the updated single-input matching network and the configuration rule information set.

[0050] Step 7: Construct node indexes for the updated single-input matching network and the multi-input matching network to obtain a node index information set, and generate a configuration matching cost model based on the node index information set. The node index information in the node index information set can be a structured index relationship established on Alpha and Beta nodes, forming a fast location link between nodes by binding keyword identifiers to nodes, i.e., a fast retrieval directory for nodes. The keyword identifier can be an identifier that combines the configuration object and the association relationship of the rule configuration triple. The link can be formed by connecting similar nodes according to the keyword identifier, creating a mapping relationship between the index keyword identifier and the node set, while recording the parent-child association relationship between nodes to form a complete index link. The configuration matching cost model can be an evaluation function used to quantify the resource consumption and time cost of performing matching operations through nodes to determine the node matching cost. In practice, the executing entity can first determine the sum of the matching costs of the left and right nodes and the node matching consumption costs of each node in the updated single-input matching network and the multi-input matching network using the node index information set, as the first configuration matching cost function. The node matching cost represents the time and resource consumption costs of node matching. The matching costs for the left and right nodes can be calculated recursively. For example, if the left or right node is a leaf node, the matching cost is a preset base cost representing the node's own overhead. If the left or right node is a non-leaf node, the matching cost is the sum of the matching costs of all its child nodes. Next, the sum of the product of the left node's Alpha memory consumption cost and the right node's Beta memory consumption cost, and the product of the right node's Alpha memory consumption cost and the left node's Beta memory consumption cost, is determined as the node matching cost. The Alpha and Beta memory consumption costs can be obtained by real-time monitoring of node memory usage. Then, the product of the left and right node's Beta memory consumption costs and the node matching success rate is determined as the second configuration matching cost function. Finally, the product of the node matching cost and the node matching success rate is determined as the node Alpha memory consumption cost function. The node matching success rate mentioned above can be obtained through historical node statistics, and is the ratio of the number of successfully matched node instances to the total number of incoming node instances. Finally, the first configuration matching cost function, the node matching cost, the second configuration matching cost function, and the node alpha memory consumption cost function are determined as the configuration matching cost model.

[0051] Step 8: Based on the historical vehicle product configuration information set and the aforementioned rule configuration triplet set, generate a fuzzy rule configuration triplet set and a dynamic fuzzy matching threshold. The fuzzy rule configuration triplets in the aforementioned fuzzy rule configuration triplet set represent the importance of the rule configuration triplet set. The dynamic fuzzy matching threshold can be the adjacency value that matches the rule configuration triplet set to determine whether a match is successful. In practice, the executing entity uses the fuzzy C-means clustering algorithm to generate the fuzzy rule configuration triplet set based on the aforementioned historical vehicle product configuration information set. Then, the data missing rate of the vehicle product configuration information is evaluated to obtain the missing data value. Finally, the sum of the products of the preset matching threshold, the difference between the threshold and the missing data value, and the weight set corresponding to the rule configuration triplet set is determined as the dynamic fuzzy matching threshold.

[0052] Step 9: Based on the above configuration matching cost model, the above fuzzy rule configuration triplet set, and the above dynamic fuzzy matching threshold, the above product configuration triplet set, the above updated single-input matching network, and the above multi-input matching network are matched and sorted to obtain the target vehicle product configuration information. Based on the above target vehicle product configuration information, the resource control terminal is controlled to perform resource scheduling operations. In practice, the above execution entity can first distribute the above fuzzy rule configuration triplet set to the nodes corresponding to the single-input matching network for Alpha node matching, and during the matching process, prioritize matching the node with the highest cost through the configuration matching cost model to obtain a matching weight set. Secondly, the above matching weight set is input to the multi-input matching network to perform multi-mode joint matching according to combinational logic to obtain a node cumulative weight set. Then, in response to determining that there is a node set in the node cumulative weight set that is greater than or equal to the dynamic fuzzy matching threshold, the rule configuration triplet set corresponding to the node set in the consequent is determined as the initial vehicle product configuration information set, and the sum of the weights corresponding to the node cumulative weight set and the rule configuration triplet set in the consequent is determined as the sorting weight value set. Finally, the initial vehicle product configuration information set is sorted from largest to smallest by sorting the weight values, and the initial vehicle product configuration information located at the initial position is taken as the target vehicle product configuration information.

[0053] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem of "low matching efficiency and accuracy, low precision in vehicle resource scheduling, resulting in waste of vehicle resources and prolonged resource scheduling time." The factors leading to low matching efficiency and accuracy, low precision in vehicle resource scheduling, waste of vehicle resources, and prolonged resource scheduling time are often as follows: In industrial scenarios, the vehicle product configuration rule engine includes a large amount of configuration rule information, and the vehicle product configuration information to be matched is personalized and has diverse data formats. This requires full matching of each rule in the vehicle product configuration rule engine, resulting in low matching efficiency and accuracy, low matching response performance, and consequently, low precision in vehicle resource scheduling based on the target vehicle product configuration information, leading to waste of vehicle resources and prolonged resource scheduling time. Solving these factors can improve matching efficiency and accuracy, increase the precision of vehicle resource scheduling, reduce waste of vehicle resources, and shorten resource scheduling time. To achieve this effect, this disclosure firstly eliminates data heterogeneity by performing ternary structuring on the automotive product configuration information, and enables atomic splitting of the configuration rule information set, providing foundational data for subsequent network-based matching. Secondly, an initial single-input matching network is constructed using the generated condition-sharing model. The condition-sharing model quantifies the degree of condition sharing and sorts the data accordingly, reducing invalid and repetitive single-condition matching operations. Thirdly, the initial single-input matching network is genetically updated using the generated matching fitness function, and a multi-input matching network is generated. The matching fitness function achieves multi-objective evaluation by balancing matching efficiency and accuracy, improving the comprehensiveness of subsequent matching. Genetic updating effectively solves the local optima problem of static networks through global search capabilities, optimizing the node layout of the single-input matching network to better suit the scenario of automotive product configuration information, improving the quality of the updated single-input matching network. The multi-input matching network can cover situations where multiple antecedents are satisfied simultaneously, achieving complete rule matching. Furthermore, receiving the output of the efficiently updated single-input matching network avoids processing a large number of invalid matching results, further improving joint matching efficiency. Next, a node index is constructed and a configuration matching cost model is generated. The node index creates a fast location link between nodes, avoiding a full network traversal and significantly improving node retrieval efficiency. The configuration matching cost model evaluates the matching across multiple dimensions, quantifying the resource consumption and time cost of each node's matching operation, thus avoiding entry into cost nodes and shortening the overall matching time. Afterward, a fuzzy rule configuration triplet set and a dynamic fuzzy matching threshold are generated. Fuzzy C-means clustering guides subsequent matching to focus on core rules, weakening the influence of secondary rules. Furthermore, replacing hard matching with weighted soft matching effectively avoids matching failures caused by missing data in automotive product configuration information.Finally, the matching and sorting process obtains the target vehicle product configuration information and performs vehicle resource scheduling. This can improve the accuracy and efficiency of matching the target vehicle product configuration information, reduce the waste of computing resources, and thus improve the accuracy and timeliness of vehicle resource scheduling, reduce the waste of vehicle resources, and shorten the resource scheduling time.

[0054] In addressing the technical problems mentioned above, the application scenario—where a large amount of new automotive product configuration information is added, dynamically bound to a page, and then automotive resource scheduling is executed—often presents the following technical challenges: Performing full conflict detection on the newly added large amount of automotive product configuration information and the stored configuration information set is too extensive and labor-intensive, leading to lower stability and security of the stored configuration information, resulting in lower configuration information quality, wasted storage resources, and consequently, lower accuracy of the dynamic page and automotive resource scheduling, wasting significant automotive resources and extending scheduling time. Considering the following requirements for this application scenario—adaptability to frequent configuration information conflict detection, dynamic and static conflict detection, bucketed detection, and multi-dimensional conflict redundancy detection—we have decided to adopt the following solution:

[0055] Optionally, the above-mentioned use of an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain the target automotive product configuration information, and controlling the resource control terminal to perform resource scheduling operations based on the target automotive product configuration information, may include the following steps:

[0056] The first step involves detecting a rule information update operation in the aforementioned automotive product configuration rule engine. A snapshot verification is performed on the configuration rule information set and the updated configuration rule information set corresponding to the rule information update operation to obtain the current configuration rule snapshot information and the verified updated rule information set. The updated configuration rule information in the updated configuration rule information set can be newly added configuration rule information submitted by business personnel. The current configuration rule snapshot information can refer to a read-only, immutable data copy created for the currently stored configuration rule information set when the update operation occurs, serving as a stable benchmark for comparison and verification in subsequent processing to prevent data inconsistency caused by concurrent operations. The verified updated rule information in the verified updated rule information set can be an updated configuration rule information set that satisfies syntactic integrity and business validity. Syntactic integrity refers to whether there are any semantic deficiencies in the trigger conditions of the updated configuration rule information and whether the execution action is executable. Business validity refers to whether the execution action meets the requirements of the automotive product configuration.

[0057] The second step involves performing bucketed multi-dimensional semantic deduplication on the aforementioned verified updated rule information set and the aforementioned current configuration rule snapshot information to obtain a deduplicated configuration rule information set. In practice, the aforementioned execution entity can first use a hash bucketing algorithm to bucket the aforementioned verified updated rule information set and the aforementioned current configuration rule snapshot information to obtain a configuration rule bucket information set. The configuration rule bucket information in the aforementioned configuration rule bucket information set can be obtained by dividing the configuration information included in the verified updated rule information set and the current configuration rule snapshot information into different logical groups, i.e., buckets, according to key features (e.g., car model, configuration category), so that the configuration rule information included in each bucket is highly related and the coupling between rules in buckets is low. Secondly, for each configuration rule bucket information, the following checking steps are performed: syntactic redundancy detection and semantic redundancy detection are performed on the configuration rule information set included in the configuration rule bucket information to obtain syntactic redundancy detection information and semantic redundancy detection information. The aforementioned syntactic redundancy detection information can be obtained by performing implication verification using the built-in conflict detection function of the rule engine (e.g., Drools' Rule Conflict Detection tool) to detect whether there are any inclusion, opposition, or overlap relationships in the formal logic of the configuration rule information set. The aforementioned semantic redundancy detection information can be obtained by determining the semantic cosine similarity between configuration rule information, with a cosine similarity greater than or equal to 80% indicating semantic redundancy. Finally, using the Drools rule engine, based on multiple syntactic redundancy detection information and multiple semantic redundancy detection information, the aforementioned verified updated rule information set and the aforementioned current configuration rule snapshot information are deduplicated to obtain the deduplicated configuration rule information set.

[0058] The third step involves identifying rule associations within the deduplicated configuration rule information set to obtain a configuration rule association information set. This association information set can represent whether there are conflicts between the deduplicated configuration rules, and the execution order and priority of the rules. In practice, the executing entity can first extract association information from the deduplicated configuration rule information set using regular expressions or a DSL (Dynamic Script Language) parser to obtain an initial association information set. Then, the deduplicated configuration rule information set and the initial association information set are input into a fine-tuned SpanBERT model to obtain an initial configuration triplet set and a confidence set. The SpanBERT model can be a model fine-tuned using rule information from sample automotive products. Next, using the confidence set, at least one initial configuration triplet set with a confidence threshold greater than or equal to a preset confidence threshold is selected from the initial configuration triplet set. This preset confidence threshold can be a pre-defined critical value used to determine whether to remove initial association information. For example, the preset confidence threshold could be 0.85. Finally, the initial association information set corresponding to at least one initial configuration triple is determined as the configuration rule association information set.

[0059] The fourth step involves constructing a relational graph by combining the aforementioned configuration rule association information set with the corresponding deduplicated configuration rule information set, resulting in a configuration context relational graph. This configuration context relational graph can be a graph representing the configuration rule association information set and rule semantic attribute information among the deduplicated configuration rule information. Nodes in the configuration context relational graph can be the deduplicated configuration rule information corresponding to the configuration rule association information set, and connecting edges can be the aforementioned configuration rule association information set.

[0060] The fifth step involves performing dynamic and static conflict detection on the aforementioned configuration context association graph to obtain a conflict detection result set. The conflict detection results in this set characterize whether conflicts exist in the deduplicated configuration rule information set; if conflicts exist, the conflict type detection result is output. In practice, the execution entity can first use the Z3 solver (Satisfiability Modulo Theory solver) to perform pre- and post-contradictory detection on the deduplicated configuration rule information set, obtaining a pre- and post-contradictory detection result set. This pre- and post-contradictory detection result can indicate whether there is a contradiction between the pre-condition and post-execution parts of the deduplicated configuration rule information. Secondly, a depth-first traversal algorithm is used to perform a static traversal of the first configuration graph, obtaining a static graph detection result set. This static graph detection result can indicate whether there are loops or strongly connected components. The first configuration graph can be the graph corresponding to the deduplicated configuration rule information after removing the pre- and post-contradictory detection results that characterize conflicts. Then, using the Drools rule engine, dynamic rule conflict detection is performed on the deduplicated configuration rule information set corresponding to the second configuration graph, resulting in a dynamic rule detection result set. The second configuration graph can be the graph corresponding to the deduplicated configuration rule information after removing the static graph detection results that represent conflicts. Finally, using the Choco Solver constraint solver, dynamic scene conflict detection is performed on the third configuration graph, resulting in a conflict detection result set. The third configuration graph can be the graph corresponding to the deduplicated configuration rule information after removing the dynamic rule detection results that represent conflicts.

[0061] Step 6: Perform logical implication detection on the deduplicated configuration rule information set corresponding to the conflict detection result set to obtain a logical implication detection result set. The logical implication detection results in this set characterize whether the deduplicated configuration rule information is overwritten by other deduplicated configuration rule information. The deduplicated configuration rule information set corresponding to the conflict detection result set can be obtained by removing conflicting deduplicated configuration rule information corresponding to the conflict detection result set from the deduplicated configuration rule information set. In practice, the execution entity can first use the Prolog engine to transform the corresponding deduplicated configuration rule information set and then perform targeted backtracking to obtain the logical implication detection result set. This transformation can be a transformation of the corresponding deduplicated configuration rule information set into Horn clauses.

[0062] Step 7: Based on the aforementioned logical implication detection result set, incremental structuring processing is performed on the aforementioned deduplicated configuration rule information set to obtain a rule configuration triplet set. Similarly, triplet structuring processing is performed on the aforementioned automotive product configuration information to obtain a product configuration triplet set. Based on the product configuration triplet set and the rule configuration triplet set, configuration rule matching is performed on the automotive product configuration information corresponding to the aforementioned resource call request to obtain the target automotive product configuration information. Finally, based on the target automotive product configuration information, the resource control terminal executes resource scheduling operations. In practice, the executing entity can remove the deduplicated configuration rule information set corresponding to the logical implication detection results representing anomalies from the aforementioned deduplicated configuration rule information set to obtain the target configuration rule information set. Then, triplet extraction is performed on the target configuration rule information set to obtain a rule configuration triplet set, and triplet structuring processing is performed on the aforementioned automotive product configuration information to obtain a product configuration triplet set. Finally, based on the rule configuration triplet set and the product configuration triplet set, the configuration rule matching is performed on the vehicle product configuration information corresponding to the above resource call request to obtain the target vehicle product configuration information. The implementation method of controlling the resource control terminal to perform resource scheduling operation based on the above target vehicle product configuration information can refer to the implementation method of steps two to nine in the previous "Some optional implementation methods in some embodiments", which will not be repeated here.

[0063] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem of "low accuracy of dynamic pages and vehicle resource scheduling precision, resulting in a waste of a large amount of vehicle resources and prolonged scheduling time." The factors leading to low accuracy of dynamic pages and vehicle resource scheduling precision, wasting a large amount of vehicle resources and prolonging scheduling time are often as follows: Performing full conflict detection on a large number of newly added vehicle product configuration information and the stored configuration information set results in an excessively large detection scope and workload, and also leads to low stability and security of the stored configuration information, resulting in low quality configuration information, wasting a large amount of storage resources, and consequently, low accuracy of dynamic pages and vehicle resource scheduling precision, wasting a large amount of vehicle resources and prolonging scheduling time. If the above factors are resolved, the accuracy of dynamic pages and vehicle resource scheduling precision can be improved, reducing the waste of vehicle resources and shortening scheduling time. To achieve this effect, this disclosure firstly uses snapshot verification to provide a stable and immutable baseline data version for newly added configuration information conflict detection, effectively preventing data competition, dirty reads, or data overwriting problems during rule information updates. Secondly, bucketed multi-dimensional semantic deduplication divides configuration rule information into multiple highly cohesive and loosely coupled buckets, which can narrow the data scope for conflict detection, avoid brute-force comparisons in the full dataset, and improve deduplication efficiency through semantic and logical redundancy deduplication. Thirdly, generating a configuration context association graph allows for visual and quantifiable visualization of the relationships between deduplicated configuration rule information, facilitating subsequent dynamic and static conflict detection. Then, dynamic and static conflict detection, through multi-level conflict detection and implicit conflict detection within logical implication detection, improves the accuracy of conflict detection from multiple perspectives. Finally, incremental structuring processing is performed on the detected configuration information set to obtain product configuration triplet sets. The resource control terminal executes resource scheduling operations, performing conflict detection on a large amount of newly added configuration information in a stable environment. This effectively avoids the workload of full-scale conflict detection, improves conflict detection efficiency and accuracy, enhances the quality of target vehicle product configuration information, and consequently improves the accuracy of vehicle resource scheduling based on target vehicle product configuration information, reducing vehicle resource waste and shortening scheduling time.

[0064] Step 102: Extract configuration elements from the target vehicle product configuration information to obtain the vehicle product atomic configuration information set.

[0065] In some embodiments, the aforementioned executing entity may extract configuration elements from the aforementioned target vehicle product configuration information to obtain a set of atomic configuration information for the vehicle product. The atomic configuration information in this set may be the information of the smallest, indivisible configuration element that constitutes the target vehicle product configuration information. This set of atomic configuration information may include, but is not limited to, at least one of the following: basic element information related to the vehicle product, value-determining element information (vehicle product pricing element information), distribution element information, and extended element information. The basic element information may include, but is not limited to, at least one of the following: vehicle model, user type, and vehicle value specification period (financing period). The value-determining element information may include, but is not limited to, at least one of the following: interest rate, acquisition value attribute value (handling fee) for purchasing or leasing the vehicle product, and vehicle product circulation substitution attribute value (margin offset information). The distribution element information may include, but is not limited to, at least one of the following: distribution area information and distribution dealer coverage information. The extended element information may include, but is not limited to, at least one of the following: new energy vehicle subsidy information, vehicle product added value information (additional loans), and vehicle product subsidy information.

[0066] In some optional implementations of certain embodiments, the above-mentioned extraction of configuration elements from the target vehicle product configuration information to obtain an atomic configuration information set for the vehicle product may include the following steps:

[0067] The first step is to divide the aforementioned target vehicle product configuration information into elements, resulting in a vehicle product configuration element information set. The vehicle product configuration element information in this set can be the smallest, indivisible field information that constitutes the aforementioned target vehicle product configuration information.

[0068] The second step involves matching the aforementioned set of automotive product configuration element information with the preset set of product atomic configuration information to obtain an atomic configuration matching result set. The preset product atomic configuration information in the preset set can be pre-defined, indivisible field information that constitutes the automotive product configuration information. The atomic configuration matching results in the aforementioned atomic configuration matching result set represent whether the automotive product configuration element information and the preset product atomic configuration information have successfully matched.

[0069] The third step is to create atomic configuration information for at least one automotive product configuration element corresponding to the unsuccessful matching result set, in response to the determination that there is a matching result in the atomic configuration matching result set, thereby obtaining the atomic configuration information set.

[0070] The fourth step is to determine whether at least one automotive product configuration element information corresponding to a successful matching result has undergone atomic element update processing.

[0071] Fifth, in response to the determination that atomic element update processing exists, at least one vehicle product configuration element information with atomic element update processing is identified as a version update atomic configuration information set. The version update atomic configuration information in the aforementioned version update atomic configuration information set can be the latest version of the atomic configuration information for a preset product with update processing, identified as the version update atomic configuration information.

[0072] Step 6: The above-mentioned version update atomic configuration information set and the above-mentioned creation atomic configuration information set are identified as the atomic configuration information set for automotive products.

[0073] Step 103: Perform business combination on the atomic configuration information set of automotive products to obtain automotive product template configuration information.

[0074] In some embodiments, the aforementioned executing entity can perform business combination on the aforementioned atomic configuration information set of automotive products to obtain automotive product template configuration information. The aforementioned automotive product template configuration information can be information from a configuration template composed of the aforementioned atomic configuration information set of automotive products. For example, the aforementioned automotive product template configuration information can be new energy vehicle installment template information (composed of atomic configuration information for new energy vehicle models, atomic configuration information for 36 / 48-month terms, atomic configuration information for floating interest rates, and atomic configuration information for subsidies). The template parameters of the aforementioned automotive product template configuration information can be configured. For example, the atomic configuration information for floating interest rates in the aforementioned automotive product template configuration information can be set as a configurable item to achieve reuse and standardization of the automotive product atomic configuration information set, reducing repetitive configuration workload.

[0075] As an example, the aforementioned executing entity can use business rule information to perform business combination on the aforementioned atomic configuration information set of automotive products to obtain automotive product template configuration information. Here, the aforementioned business rule information can be rule information determined by the target user (e.g., business personnel) through business policies.

[0076] Step 104: Fill in the configuration parameters for the automotive product template configuration information to obtain the automotive product information.

[0077] In some embodiments, the executing entity can populate the vehicle product template configuration information with configuration parameters to obtain vehicle product information. This vehicle product information can be the information of the vehicle product that best matches the target vehicle product configuration information. For example, the vehicle product information can be vehicle product information (e.g., a new energy vehicle installment template information) populated with configuration parameters (e.g., an interest rate of 3.8% and a subsidy rate of 2%) (e.g., a special installment product for a new energy vehicle of brand xx). In practice, the executing entity can first obtain a set of historical parameter configuration information related to the configuration parameter set of the vehicle product template configuration information. Then, the historical parameter configuration information set and related parameter document information are input into a large language model for statistical analysis to obtain a target parameter configuration information set. This parameter document information can be national policy document information related to the configuration parameter set. Finally, the target parameter configuration information set is input into the vehicle product template configuration information to obtain the vehicle product information.

[0078] Step 105: Using the product value determination rule engine, perform value decoupling and matching processing on the automotive product information to obtain the target product value determination information.

[0079] In some embodiments, the aforementioned executing entity can utilize a product value determination rule engine to perform value decoupling and matching processing on the aforementioned automotive product information to obtain target product value determination information. The target product value determination information (automotive product interest subsidy information) can be information used to determine the value subsidy of the aforementioned automotive product information. The aforementioned product value determination rule engine can be a rule engine that determines the target product value determination information that best matches the aforementioned automotive product information. The aforementioned product value determination rule engine may include: a value determination rule base, a value determination working memory, and a value determination inference engine. The aforementioned value determination rule base can be configured by matching rules through the following steps: using the Phreak algorithm, performing pattern matching between the stored set of product value configuration rule information related to automotive product information and the aforementioned automotive product information to obtain a set of product value configuration rule information. Secondly, in response to the determination that the set of product value configuration rule information includes multiple sets of product value configuration rule information, the earliest product value configuration rule information is selected from the set of product value configuration rule information in a first-in-first-out (FIFO) time order as the target product value determination information.

[0080] In some optional implementations of certain embodiments, the above-mentioned use of a product value determination rule engine to perform value decoupling and matching processing on the above-mentioned automotive product information to obtain target product value determination information may include the following steps:

[0081] The first step involves decomposing the value determination rule information set included in the aforementioned product value determination rule engine into rule nodes, resulting in a value determination condition node set and a value determination execution node set. The value determination rule information in the aforementioned value determination rule information set can be the rule information existing in the aforementioned product value determination rule engine used to determine the calculation method for the interest subsidy information of automobile products. The value determination condition nodes in the aforementioned value determination condition node set can be the smallest logically decidable unit of the condition part in the value determination rule information. The value determination execution nodes in the aforementioned value determination execution node set can be the nodes that execute the interest subsidy information of the automobile product after the corresponding value determination condition is successfully matched. For example, the aforementioned value determination rule information could be "Purchase a new energy passenger vehicle, and receive a subsidy of 12% of the new vehicle sales price, up to a maximum of 20,000 yuan." Then, the aforementioned value determination condition node could be "Vehicle type: New energy passenger vehicle." The aforementioned value determination execution node could be "Subsidy information: 12% of the sales price" and "Maximum subsidy amount: 20,000 yuan."

[0082] The second step involves determining the set of value-determining related edges and the initial edge weight set for the aforementioned set of value-determining condition nodes and the set of value-determining execution nodes. The value-determining related edges in the set of value-determining related edges can be directed edges connecting value-determining condition nodes and value-determining execution nodes. The initial edge weights in the initial edge weight set can be determined by calculating the ratio of the number of times each node appears to the total number of occurrences. These initial edge weights characterize the importance of the value-determining related edges.

[0083] The third step involves inputting the aforementioned set of value determination condition nodes, the aforementioned set of value determination execution nodes, the aforementioned set of value determination related edges, and the aforementioned set of initial edge weights into a graph database to obtain a value determination rule dependency graph. This value determination rule dependency graph can be a graph representing the relationships between the set of value determination condition nodes, the set of value determination execution nodes, and the aforementioned set of value determination related edges. The graph database can be a pre-defined Neo4j graph database.

[0084] The fourth step involves performing a graph traversal matching on the aforementioned value determination rule dependency graph and the aforementioned automotive product information to determine whether the automotive product information and each value determination condition node encountered during the traversal should execute a node activation operation. This node activation operation can be used to determine nodes with uncertainties and real-time requirements, to handle nodes with non-Boolean logic, or to ensure traceability of the matching state. The operation to determine nodes with uncertainties and real-time requirements can be an activation operation where the node needs to acquire external data in real time to make a determination. For example, the operation to ensure traceability of the matching state could be that node A and node B are activated, but value determination execution node C is not activated, while value determination execution node D is activated; in this case, the final determined target product value determination information is located at value determination execution node D. The graph traversal matching can be performed using a depth-first search algorithm. The execution of node activation operations can be a dynamic execution process for graph traversal matching, improving the real-time performance, complexity, and uncertainty handling capabilities of the matching, thus avoiding degradation into a static path query.

[0085] Fifth, in response to the determination of unexecuted node activation operations, the set of value-determining traversal paths corresponding to the unexecuted node activation operations is pruned according to the aforementioned value-determining rule dependency graph, resulting in a value-determining pruned dependency graph. This value-determining pruned dependency graph can be a graph structure obtained by removing the set of value-determining traversal paths from the aforementioned value-determining rule dependency graph. The value-determining traversal paths in the set of value-determining traversal paths can be paths located after the nodes corresponding to the unexecuted node activation operations.

[0086] Step 6: In response to the activation operation of the determined execution node, perform path optimization on the value-determined traversal path set to obtain the optimized value-determined traversal path set. This path optimization can be performed using the initial edge weight set as the initial value for the heuristic algorithm.

[0087] Step 7: In response to the completion of the graph traversal matching, generate the target product value determination information based on the aforementioned value-determined pruned dependency graph or optimized value-determined traversal path set. The completion of the graph traversal matching can be defined as the end of matching when the executed node has no subsequent nodes. In practice, the executing entity can, in response to the graph traversal matching result, determine the rule information corresponding to the last value-determined execution node in the aforementioned value-determined pruned dependency graph or optimized value-determined traversal path set as the target product value determination information.

[0088] Step 106: Based on the target product value determination information, dynamically call the pre-set product value determination algorithm library to perform product value determination processing and obtain the automobile product value information.

[0089] In some embodiments, the aforementioned executing entity may dynamically invoke a pre-defined product value determination algorithm library to perform product value determination processing based on the aforementioned target product value determination information, thereby obtaining automotive product value information. The aforementioned automotive product value information may be the most relevant value determined by matching multiple dimensions, including vehicle model, region, and time period, as included in the aforementioned target product value determination information (e.g., equal principal and interest repayment information, equal principal repayment information, interest-only repayment information, tiered loan repayment information, financial lease repayment information (including direct lease, sale-leaseback, residual value calculation), and flexible repayment information). The aforementioned automotive product value information may be the "2024 B-class vehicle interest subsidy scheme in North China."

[0090] In some optional implementations of certain embodiments, the above-mentioned dynamic invocation of a pre-defined product value determination algorithm library to perform product value determination processing based on the target product value determination information to obtain automobile product value information may include the following steps:

[0091] The first step, in response to the detected drag-and-drop operation by the target user targeting the aforementioned preset product value determination algorithm library, is to determine whether the product value determination algorithm set corresponding to the drag-and-drop operation includes a personalized value determination formula. The aforementioned preset product value determination algorithm library can be a pre-defined database composed of algorithms used to determine the value information of automotive products. This library can be an algorithm library for relevant value calculation algorithms (equal principal and interest payments, equal principal payments, interest-only payments followed by principal payments, tiered loans, financial leasing, flexible repayment). The target user can be a salesperson used to calculate the value information of automotive products. The aforementioned personalized value determination formula can be a value determination formula that does not exist in the aforementioned preset product value determination algorithm library and needs to be customized by the target user. For example, the personalized value determination formula could be a formula that returns a value attribute value (a dealer rebate formula). The aforementioned product value determination algorithm set consists of algorithms stored in the aforementioned preset product value determination algorithm library.

[0092] The second step involves, in response to the determination of the personalized value determination formula, invoking the personalized value determination editor to obtain the target personalized value determination formula. This personalized value determination editor can be an editor that allows the target user to edit the formula visually. This personalized value determination editor can be a MathType editor. The target personalized value determination formula can be a formula generated by the target user using the personalized value determination editor.

[0093] The third step involves using the product value determination interface to combine and call the product value determination algorithm set corresponding to the drag-and-drop operation, thereby obtaining value determination algorithm combination information. The product value determination interface can be an interface used to call product value determination algorithm sets included in a preset product value determination algorithm library. The value determination algorithm combination information (for example, a new energy vehicle installment calculation scheme could be a scheme consisting of an equal principal and interest repayment algorithm, an interest subsidy deduction rule algorithm, and a new energy subsidy algorithm) can be a combination formula obtained by combining the above product value determination algorithm set.

[0094] The fourth step involves verifying the combination of the aforementioned target personalized value determination formula and the aforementioned value determination algorithm to obtain a verification result information set. The verification result information in this set can be used to verify the rationality of the combination logic and formula syntax of the aforementioned target personalized value determination formula and the aforementioned value determination algorithm combination.

[0095] Fifth, in response to the determination that the above verification result information set represents that the verification has passed, the above value determination algorithm combination information and the above target personalized value determination formula are determined as the value information of the automobile product.

[0096] Step 107: Based on the vehicle product information and vehicle product value information, dynamically render and bind the product configuration terminal to obtain the vehicle product configuration page, and based on the vehicle product configuration page, control the resource control terminal to perform resource scheduling operations.

[0097] In some embodiments, the aforementioned execution entity can dynamically render and bind the product configuration terminal to obtain a vehicle product configuration page based on the aforementioned vehicle product information and vehicle product value information. Furthermore, based on the aforementioned vehicle product configuration page, it can control a resource control terminal to perform resource scheduling operations, wherein the aforementioned resource scheduling operations include at least one of the following: transportation of vehicle parts and transportation of complete vehicles. The aforementioned product configuration terminal can be a terminal used to configure relevant configuration information of vehicle products. The aforementioned vehicle product configuration page can be a personalized page designed based on the aforementioned vehicle product information and vehicle product value information. When the vehicle product information includes the atomic configuration information of "new energy vehicle subsidies," the aforementioned vehicle product configuration page can automatically display the relevant subsidy fields. When the vehicle product information does not include the atomic configuration information of "GPS installation requirements," the aforementioned vehicle product configuration page can automatically hide the relevant input fields. The aforementioned resource control terminal can be a hardware or software terminal used to dynamically adjust the resource scheduling of the entire automotive industry chain. For example, the aforementioned resource control terminal can be an automotive supplier sourcing platform. The aforementioned resource scheduling operations can promote the sales of corresponding vehicle products by leveraging the vehicle subsidy level corresponding to the vehicle product configuration page, thereby promoting the scheduling of automotive resources across the automotive industry chain. The aforementioned vehicle resource scheduling operations may also include: scheduling the inventory quantity of different vehicle models, and increasing subsidies for vehicles with a large number of models. In practice, the executing entity can utilize React Hooks (a web development framework) to dynamically render and bind the product configuration page based on the aforementioned vehicle product information and value information, thereby obtaining the vehicle product configuration page. Furthermore, using simulated annealing algorithms, the resource control terminal can be controlled to execute vehicle resource scheduling operations based on the vehicle product configuration page.

[0098] In some optional implementations of certain embodiments, the process of dynamically rendering and binding the product configuration page based on the aforementioned vehicle product information and vehicle product value information to obtain the vehicle product configuration page may include the following steps:

[0099] The first step is to perform hierarchical constraint decomposition on the aforementioned automotive product information and automotive product value information to obtain a hierarchical automotive product information set and a hierarchical subsidy value information set. The hierarchical automotive product information set can be derived by decomposing the automotive product information into explicit / implicit field information and field association information. The hierarchical subsidy value information set can be derived by decomposing the automotive product value information into field rendering configuration information, subsidy value calculation logic information, and subsidy value verification information.

[0100] The second step involves generating front-end and back-end interface association constraint information based on the aforementioned tiered automotive product information set and tiered subsidy value information set. This constraint information serves to constrain data interaction between the front-end and back-end using the tiered automotive product information set and the tiered subsidy value information set. It may include: fields, types, optional or mandatory options, default values, and dynamic data interaction configuration information returned by the back-end. In practice, the executing entity can utilize scripts and template engines to execute the received constraint file for the tiered automotive product information set and the tiered subsidy value information set to obtain the front-end and back-end interface association constraint information. This constraint file can be a constraint file generated through expert experience, mapping the front-end and back-end interactions of the tiered automotive product information set and the tiered subsidy value information set.

[0101] The third step involves generating a set of reusable constraint functions for both the front-end and back-end interfaces based on the aforementioned front-end and back-end interface association constraint information. This set of reusable constraint functions can be used for metadata judgment, explicit / implicit rule determination, subsidy value information calculation, and debugging logs. The explicit / implicit rule determination function can be an integrated LRU (Least Recently Used) cache, where the cache key is the vehicle product information number and configuration fields, and it is recalculated only when the configuration fields or explicit / implicit rules change. In practice, the executing entity can input the aforementioned front-end and back-end interface association constraint information into a large language model to obtain the set of reusable constraint functions.

[0102] The fourth step involves dynamically rendering the layered automotive product information set and the layered subsidy value information set based on the aforementioned set of reusable front-end and back-end constraints and the associated constraints of the front-end and back-end interfaces, resulting in an initial automotive product configuration page. This initial automotive product configuration page can be a preliminary page generated for front-end page rendering based on the layered automotive product information set and the layered subsidy value information set. In practice, the executing entity can utilize the Vue 3 Composition API to dynamically render the layered automotive product information set and the layered subsidy value information set based on the aforementioned set of reusable front-end and back-end constraints and the associated constraints of the front-end and back-end interfaces, thereby obtaining the initial automotive product configuration page.

[0103] The fifth step involves applying page debouncing to the initial car product configuration page to obtain the debounced version. This page debouncing can be achieved using debouncing technology. Specifically, when an event is triggered on the initial car product configuration page, the execution of the event handler function is delayed. If the event is triggered again within the delay period, the timer restarts. The event handler function will only execute if the event is not triggered again within the specified time. Therefore, page debouncing prevents certain high-frequency operations from being triggered frequently, thereby improving page performance.

[0104] Step 6: Perform test case testing on the initial vehicle product configuration page after image stabilization to obtain the vehicle product configuration page. This test case testing can be performed using generated page test cases to test the initial vehicle product configuration page after image stabilization, thereby improving page performance. These page test cases can be a set of test inputs, execution conditions, and expected results compiled to improve the performance of the initial vehicle product configuration page after image stabilization, in order to test whether the program or function of the initial vehicle product configuration page after image stabilization meets the requirements.

[0105] In addressing the technical problems mentioned above, the application scenario—dynamic scheduling of vehicle resources via dynamically bound pages during peak holiday periods—often presents the following challenges: Increased vehicle purchases during peak holiday periods, coupled with varying and personalized configuration information for each user, lead to reduced load and performance on vehicle configuration pages. This results in numerous redundant and erroneous vehicle resource scheduling requests and prolonged response times, leading to lower accuracy in resource adjustments, increased response time, and resource waste and stockpiling. To meet the specific requirements of this application scenario—adapting to dynamic page rendering and binding, high concurrency, personalized page display, high page performance, timely page response, and precise vehicle resource scheduling—we have decided to adopt the following solution:

[0106] Optionally, the above-mentioned dynamic rendering and binding of the product configuration terminal to obtain the vehicle product configuration page based on the aforementioned vehicle product information and vehicle product value information, and the control of the resource control terminal to perform resource scheduling operations based on the vehicle product configuration page, may include the following steps:

[0107] The first step is to dynamically bind the product configuration page based on the aforementioned vehicle product information and value information, resulting in a vehicle product configuration binding page. This binding page can be a page that dynamically displays the vehicle product information and value information. The implementation method for this step can be found in the section above, "Some Optional Implementations in Some Embodiments," and will not be repeated here.

[0108] The second step involves generating a set of page test cases for the aforementioned vehicle product information and value information, specifically for the vehicle product configuration binding page. These page test cases can encompass various aspects of testing related to the vehicle product configuration binding page, including functional testing, exception testing, boundary condition testing, load testing, stress testing, resource utilization testing, and security testing. As an example, the executing entity can utilize an automated test case generation tool to generate the set of page test cases for the aforementioned vehicle product configuration binding page based on the vehicle product information and value information. This automated test case generation tool can be the TestNG automated testing framework.

[0109] The third step is to determine the text similarity matrix and statement coverage similarity matrix for the aforementioned page test case set. Each element in the text similarity matrix represents the Euclidean distance similarity between any two page test cases. Each element in the statement coverage similarity matrix represents the degree of overlap in the test statements executed when the corresponding two page test cases are tested. The statement coverage similarity matrix can be obtained through the following steps: First, determine the test statement group included in each page test case to obtain a test statement group set. Then, using Jaccard distance similarity, determine the similarity between any two page test cases to obtain the statement coverage similarity matrix.

[0110] The fourth step involves clustering the page test case set based on the aforementioned text similarity matrix and statement coverage similarity matrix, resulting in a set of page test case clusters. Each page test case cluster can be a set of multiple highly similar page test cases, determined by the Davies-Bouldin Index (DB). The DB index characterizes the clustering effect; a smaller DB index indicates better clustering. For example, the execution entity can first perform a weighted summation of the text similarity matrix and statement coverage similarity matrix to obtain a test case similarity matrix. The weights in this weighted summation can be manually determined, for example, 0.6 and 0.4. Then, using a hierarchical clustering algorithm, the page test case set is clustered based on the test case similarity matrix to obtain the set of page test case clusters.

[0111] Step 5: For each page test case cluster in the aforementioned page test case cluster set, perform intra-cluster test case defect prediction on the set of page test cases included in the aforementioned page test case cluster to obtain an intra-cluster defect prediction information set. The intra-cluster defect prediction information in the aforementioned intra-cluster defect prediction information set can be information characterizing whether a page test case is defective and its corresponding probability value. A defect can characterize that the execution result of the page test case testing the automotive product configuration binding page is inconsistent with the preset execution result. The preset execution result can be the expected result before the page test case is executed. In practice, the aforementioned execution entity can first extract semantic features from each page test case to obtain a set of page test feature vectors. The page test feature vectors can characterize the attribute information of the page test case. Then, the aforementioned page test feature vector set is input into the test case defect prediction classification model to obtain intra-cluster defect prediction information. The aforementioned test case defect prediction classification model can be a model that classifies and identifies defects in the input set of page test feature vectors. For example, the aforementioned test case defect prediction classification model can be a model composed of a Transformer, a Long Short-Term Memory Neural Network model, and an SVM (Support Vector Machine) cascaded together. The aforementioned test case defect prediction classification model can be a model trained using a sample set of defective and non-defective test cases based on the hinge loss function.

[0112] Step 6: Based on the aforementioned cluster-based defect prediction information set, sort the aforementioned page test case clusters within each cluster to obtain a cluster-based page test case sequence set. As an example, the executing entity can first sort the page test case clusters with the defective classification label from largest to smallest using the probability value set of defects included in the aforementioned cluster-based defect prediction information set, thus obtaining a cluster-based page test case sequence set.

[0113] Step 7: Based on the aforementioned intra-cluster defect prediction information set and the aforementioned intra-cluster page test case sequence set, perform inter-cluster test case sorting on the aforementioned page test case cluster set to obtain an inter-cluster page test case sequence. The inter-cluster page test case sequence can be a sequence obtained by inter-cluster sorting of the page test case cluster set. This inter-cluster sorting can be based on the number of test cases with the classification label "defective" in each page test case cluster, ordered from largest to smallest. When multiple page test case clusters include the same number of defective page test cases, the inter-cluster sorting is performed based on the probability values ​​of the intra-cluster page test cases located at the initial position in the intra-cluster test case sequence, ordered from largest to smallest. When multiple page test case clusters do not include defective page test cases, the inter-cluster sorting is performed based on the intra-cluster distances located at the termination position in the intra-cluster distance set, ordered from largest to smallest. As an example, the executing entity can perform the following inter-cluster sorting steps for each page test case cluster: First, determine the Euclidean distance between the group of page test cases with the classification label "no defect" and the cluster center, as the intra-cluster distance set. The cluster center can be the location information of the page test case located at the center. Next, the number of page test cases with the category label "defective" in each intra-cluster page test case sequence set is determined, resulting in a test case number set. Then, this test case number set is sorted from largest to smallest to obtain a test case number sequence. Next, in response to the determination that the page test case cluster set includes a cluster of defective page test cases with the same number of test cases, the probability value set of the page test cases located at the starting position in the multiple intra-cluster page test case sequences corresponding to the defective page test case cluster set is determined as the defect probability dataset. Then, this defect probability value set is sorted from largest to smallest among clusters. Finally, in response to the determination that the page test case cluster set does not include a cluster of defective page test cases, the page test case cluster sets are sorted from largest to smallest among inter-cluster distances based on the corresponding multiple intra-cluster distances, resulting in an inter-cluster page test case sequence.

[0114] Step 8: Perform test case sampling on the above inter-cluster page test case sequence to obtain a sampled page test case sequence. In practice, the execution entity can select one page test case from each of the page test case clusters categorized as "defect-free" according to the order of the above inter-cluster page test case sequence, and then perform sampling adjustment to obtain the sampled page test case sequence.

[0115] Step nine involves performing test case checks on the aforementioned vehicle product configuration binding page based on the sampled page test case sequence. This yields the vehicle product configuration page, and based on this page, the resource control terminal executes resource scheduling operations. The vehicle product configuration page is one that, after testing, shows no performance issues. Therefore, this step, through precise test case checks, ensures the accuracy and consistency of vehicle product configuration information, effectively avoiding low page conversion rates due to page errors, improving vehicle product conversion rates, enhancing the accuracy of resource scheduling, shortening vehicle product launch cycles, rapidly responding to market changes, improving the precise scheduling of vehicle resources, and reducing resource hoarding and waste.

[0116] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem of "leading to reduced load and performance of the automotive product configuration page, lower accuracy of resource adjustment, increased response time of resource scheduling, and waste and hoarding of automotive resources." The factors leading to reduced load and performance of the automotive product configuration page, lower accuracy of resource adjustment, increased response time of resource scheduling, and waste and hoarding of automotive resources are often as follows: During peak holiday periods, users purchase more automotive products, and each user's configuration information for automotive products is different and somewhat personalized, leading to reduced load and performance of the automotive product configuration page. This results in a large number of redundant and erroneous automotive resource scheduling demands and longer resource scheduling response times, leading to lower accuracy of resource adjustment, increased response time of resource scheduling, and waste and hoarding of automotive resources. To achieve this effect, this disclosure firstly, automatically generating page test case sets can avoid the subjectivity of generating test cases based on expert experience, improve the objectivity and comprehensiveness of the page test case sets, and facilitate the accuracy of subsequent testing of the automotive product configuration binding page. Secondly, determining the text similarity matrix and sentence coverage similarity matrix, considering the similarity of page test cases from both static text and dynamic sentence coverage perspectives, is beneficial for improving the subsequent optimization of the test case sequence. Thirdly, clustering the page test case set, grouping similar test cases into the same cluster, facilitates faster identification of page defects. Subsequently, predicting defects within each cluster and performing intra-cluster and inter-cluster sorting prioritizes defective page test cases, facilitating faster defect identification, improving the detection rate, and reducing the page's testing load. Next, sampling and adjusting the inter-cluster page test case sequence, by pre-testing the page test cases most effective at detecting page performance issues, reduces the number of page test cases used for page detection, improving the detection rate, reducing the page's testing load, and improving page performance. Finally, controlling the resource control terminal to perform vehicle resource scheduling operations can improve the accuracy of resource adjustments, shorten resource scheduling response time, and reduce vehicle resource waste and hoarding.

[0117] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a rule engine-based automotive product resource scheduling device. These device embodiments are similar to... Figure 1 Corresponding to the method embodiments shown, this rule engine-based automotive product resource scheduling device can be specifically applied to various electronic devices.

[0118] like Figure 2As shown, a rule engine-based automotive product resource scheduling device 200 includes: a configuration rule matching unit 201, a configuration element extraction unit 202, a business combination unit 203, a configuration parameter filling unit 204, a value decoupling matching unit 205, a call combination unit 206, and a control unit 207. The configuration rule matching unit 201 is configured to: in response to detecting a resource call request for an automotive product configuration page, use an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain target automotive product configuration information. The configuration element extraction unit 202 is configured to: extract configuration elements from the target automotive product configuration information to obtain an automotive product atomic configuration information set. The business combination unit 203 is configured to: perform business combination on the automotive product atomic configuration information set to obtain automotive product template configuration information. The configuration parameter filling unit 204 is configured to: fill configuration parameters into the automotive product template configuration information to obtain automotive product information. The value decoupling and matching unit 205 is configured to: utilize the product value determination rule engine to perform value decoupling and matching processing on the aforementioned automotive product information to obtain target product value determination information. The invocation and combination unit 206 is configured to: dynamically invoke a pre-defined product value determination algorithm library to perform product value determination processing based on the aforementioned target product value determination information to obtain automotive product value information. The control unit 207 is configured to: dynamically render and bind a page to the product configuration terminal based on the aforementioned automotive product information and automotive product value information to obtain an automotive product configuration page; and, based on the aforementioned automotive product configuration page, control the resource control terminal to perform resource scheduling operations, wherein the aforementioned resource scheduling operations include at least one of the following: transportation of automotive parts and transportation of complete vehicles.

[0119] It is understandable that the units and references recorded in the rule-based automotive product resource scheduling device 200 are... Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the rule engine-based automotive product resource scheduling device 200 and the units contained therein, and will not be repeated here.

[0120] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device (e.g., an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0121] like Figure 3As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0122] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0123] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0124] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0125] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0126] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: respond to detecting a resource call request for a vehicle product configuration page; utilize an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain target automotive product configuration information; extract configuration elements from the target automotive product configuration information to obtain an automotive product atomic configuration information set; perform business combination on the automotive product atomic configuration information set to obtain automotive product template configuration information; and populate the automotive product template configuration information with configuration parameters to obtain automotive product configuration information. Product information; using a product value determination rule engine, the above automotive product information is subjected to value decoupling and matching processing to obtain target product value determination information; based on the above target product value determination information, a pre-set product value determination algorithm library is dynamically invoked to perform product value determination processing to obtain automotive product value information; based on the above automotive product information and the above automotive product value information, the product configuration terminal is dynamically rendered and bound to obtain an automotive product configuration page, and based on the above automotive product configuration page, the resource control terminal is controlled to perform resource scheduling operations, wherein the above resource scheduling operations include at least one of the following: transportation of automotive parts and transportation of complete vehicles.

[0127] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0129] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a configuration rule matching unit, a configuration element extraction unit, a business combination unit, a configuration parameter filling unit, a value decoupling matching unit, a call combination unit, and a control unit. The names of these units do not necessarily limit the unit itself; for example, the configuration rule matching unit may also be described as "a unit that, in response to detecting a resource call request for an automotive product configuration page, uses an automotive product configuration rule engine to perform configuration rule matching on the automotive product configuration information corresponding to the resource call request to obtain the target automotive product configuration information."

[0130] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0131] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for scheduling automotive product resources based on a rules engine, comprising: In response to a detected resource request for a vehicle product configuration page, the system utilizes a vehicle product configuration rule engine to perform configuration rule matching on the vehicle product configuration information corresponding to the resource request to obtain target vehicle product configuration information. This includes: performing multi-dimensional semantic extraction on the vehicle product configuration information to obtain a configuration multi-dimensional semantic information set; performing conditional bitmasking and feature encoding on the configuration multi-dimensional semantic information set and the configuration rule information set included in the vehicle product configuration rule engine to obtain configuration bitwise operation feature values ​​and a configuration semantic feature vector set; and performing decision tree matching on the configuration semantic feature vector set and the vehicle product configuration decision tree corresponding to the configuration rule information set to obtain a matching result. The decision node set, wherein the automotive product configuration decision tree is obtained through the following steps: performing rule structured parsing on the configuration rule information set to obtain a configuration rule structured information set; generating an initial configuration decision tree based on the configuration rule structured information set; performing bit-mapping on the initial configuration decision tree to obtain an automotive product configuration decision tree, wherein the automotive product configuration decision tree includes: a rule bit-mask set; performing bit-operation matching on the configuration bit operation feature value and the rule bit-mask set corresponding to the matching decision node set to obtain a matching product configuration rule information set; sorting the matching product configuration rule information set to obtain target automotive product configuration information; The configuration elements of the target vehicle product configuration information are extracted to obtain the vehicle product atomic configuration information set. The atomic configuration information set of the automotive products is combined with business functions to obtain the template configuration information of the automotive products. The configuration parameters are filled into the automotive product template configuration information to obtain automotive product information; Using a product value determination rule engine, the automotive product information is subjected to value decoupling and matching processing to obtain target product value determination information. This includes: decomposing the value determination rule information set included in the product value determination rule engine into rule nodes to obtain a value determination condition node set and a value determination execution node set; determining the value determination related edge set and initial edge weight set for the value determination condition node set and the value determination execution node set; inputting the value determination condition node set, the value determination execution node set, the value determination related edge set, and the initial edge weight set into a graph database to obtain a value determination rule dependency graph; and performing value determination rule dependency... The graph and the automotive product information are matched using a graph traversal to determine whether the automotive product information and each value determination condition node traversed have performed a node activation operation. In response to determining that no node activation operation has been performed, the value determination traversal path set corresponding to the non-executed node activation operation is pruned according to the value determination rule dependency graph to obtain a value determination pruned dependency graph. In response to determining that a node activation operation has been performed, the value determination traversal path set is optimized to obtain an optimized value determination traversal path set. In response to determining that the graph traversal matching has ended, the target product value determination information is generated according to the value determination pruned dependency graph or the optimized value determination traversal path set. Based on the target product value determination information, a pre-defined product value determination algorithm library is dynamically invoked to perform product value determination processing, thereby obtaining automobile product value information. Based on the vehicle product information and the vehicle product value information, the product configuration terminal is dynamically rendered and bound to obtain a vehicle product configuration page. Based on the vehicle product configuration page, the resource control terminal is controlled to perform resource scheduling operations, wherein the resource scheduling operations include at least one of the following: transportation of vehicle parts and transportation of complete vehicles.

2. The method according to claim 1, wherein, The step of using an automotive product configuration rule engine to match configuration rules with the automotive product configuration information corresponding to the resource call request to obtain the target automotive product configuration information includes: Obtain a multi-source product association dataset associated with the initial automotive product configuration information set corresponding to the resource call request; A behavioral graph is constructed from the multi-source product association dataset to obtain a heterogeneous graph of object behavioral structure; Cross-modal feature extraction is performed on the heterogeneous graph of the object's behavioral structure to obtain a cross-modal graph behavioral feature vector set; Based on the cross-modal graph behavior feature vector set, risk causal tracing identification is performed on the heterogeneous graph of object behavior structure to obtain risk causal tracing identification information; Based on the cross-modal graph behavior feature vector set, multi-dimensional risk identification is performed on the heterogeneous graph of object behavior structure to obtain a set of configuration object risk information. Based on the risk causal tracing and identification information and the risk information set of the configuration object, risk interception and control are performed on the initial automotive product configuration information set to obtain automotive product configuration information; Using an automotive product configuration rule engine, the automotive product configuration information is matched with configuration rules to obtain the target automotive product configuration information.

3. The method according to claim 1, wherein, The step of extracting configuration elements from the configuration information of the target vehicle product to obtain an atomic configuration information set for the vehicle product includes: The target vehicle product configuration information is divided into elements to obtain a vehicle product configuration element information set. The set of automotive product configuration element information is matched with the preset set of product atomic configuration information to obtain an atomic configuration matching result set; In response to determining that there is a matching result in the atomic configuration matching result set that represents an unsuccessful match, atomic configuration information is created for at least one automotive product configuration element information corresponding to the unsuccessful match result, and an atomic configuration information set is created. Determine whether at least one automotive product configuration element information corresponding to a successfully matched result has undergone atomic element update processing. In response to the determination that there is an atomic element update process, at least one vehicle product configuration element information with atomic element update process is determined as the version update atomic configuration information set; The version update atomic configuration information set and the creation atomic configuration information set are determined as the automotive product atomic configuration information set.

4. The method according to claim 1, wherein, The step of dynamically invoking a pre-defined product value determination algorithm library based on the target product value determination information to perform product value determination processing and obtain automobile product value information includes: In response to detecting a drag-and-drop operation by a target user on the preset product value determination algorithm library, determine whether the product value determination algorithm set corresponding to the drag-and-drop operation includes a personalized value determination formula; In response to determining the personalized value determination formula, the personalized value determination editor is invoked to obtain the target personalized value determination formula; By using the product value determination interface, the set of product value determination algorithms corresponding to the drag-and-drop operation is combined and invoked to obtain value determination algorithm combination information. The combination of the target personalized value determination formula and the value determination algorithm is verified to obtain a verification result information set. In response to the determination that the verification result information set represents a successful verification, the value determination algorithm combination information and the target personalized value determination formula are determined as the automotive product value information.

5. The method according to claim 1, wherein, The step of dynamically rendering and binding the product configuration page based on the vehicle product information and the vehicle product value information to obtain the vehicle product configuration page includes: The automotive product information and the automotive product value information are decomposed into hierarchical constraints to obtain a hierarchical automotive product information set and a hierarchical subsidy value information set. Based on the hierarchical automotive product information set and the hierarchical subsidy value information set, generate front-end and back-end interface association constraint information; Based on the aforementioned front-end and back-end interface association constraint information, a set of front-end and back-end reusable constraint functions is generated; Based on the set of front-end and back-end reuse constraint functions and the set of front-end and back-end interface association constraint information, dynamic form rendering is performed on the set of hierarchical automobile product information and the set of hierarchical subsidy value information to obtain the initial automobile product configuration page. The initial vehicle product configuration page is subjected to page stabilization processing to obtain the stabilized initial vehicle product configuration page; Test cases were performed on the initial vehicle product configuration page after image stabilization to obtain the vehicle product configuration page.

6. A rule-based automotive product resource scheduling device, comprising: The configuration rule matching unit is configured to, in response to detecting a resource call request for a car product configuration page, utilize a car product configuration rule engine to perform configuration rule matching on the car product configuration information corresponding to the resource call request to obtain target car product configuration information. This includes: performing multi-dimensional semantic extraction on the car product configuration information to obtain a configuration multi-dimensional semantic information set; performing conditional bitmasking and feature encoding processing on the configuration multi-dimensional semantic information set and the configuration rule information set included in the car product configuration rule engine to obtain configuration bitwise operation feature values ​​and a configuration semantic feature vector set; and constructing a decision tree from the car product configuration decision tree corresponding to the configuration semantic feature vector set and the configuration rule information set. The matching process yields a matching decision node set. The automotive product configuration decision tree is obtained through the following steps: performing rule structured parsing on the configuration rule information set to obtain a configuration rule structured information set; generating an initial configuration decision tree based on the configuration rule structured information set; performing bit-mapping on the initial configuration decision tree to obtain an automotive product configuration decision tree, wherein the automotive product configuration decision tree includes: a rule bit-mask set; performing bit-operation matching on the configuration bit operation feature values ​​and the rule bit-mask set corresponding to the matching decision node set to obtain a matching product configuration rule information set; and sorting the matching product configuration rule information set to obtain the target automotive product configuration information. The configuration element extraction unit is configured to extract configuration elements from the configuration information of the target vehicle product to obtain an atomic configuration information set of the vehicle product. The service combination unit is configured to perform service combination on the atomic configuration information set of the automotive product to obtain automotive product template configuration information. The configuration parameter filling unit is configured to fill the configuration parameters into the automotive product template configuration information to obtain automotive product information. The value decoupling matching unit is configured to use a product value determination rule engine to perform value decoupling matching on the automotive product information to obtain target product value determination information. The invocation combination unit is configured to dynamically invoke a pre-defined product value determination algorithm library to perform product value determination processing based on the target product value determination information, thereby obtaining automotive product value information. This includes: decomposing the value determination rule information set included in the product value determination rule engine into rule nodes to obtain a value determination condition node set and a value determination execution node set; determining the value determination associated edge set and initial edge weight set for the value determination condition node set and the value determination execution node set; inputting the value determination condition node set, the value determination execution node set, the value determination associated edge set, and the initial edge weight set into a graph database to obtain a value determination rule dependency graph; and... The value determination rule dependency graph and the vehicle product information are matched by graph traversal to determine whether the vehicle product information and each value determination condition node traversed have performed a node activation operation. In response to determining that no node activation operation has been performed, the value determination traversal path set corresponding to the unexecuted node activation operation is pruned according to the value determination rule dependency graph to obtain a value determination pruned dependency graph. In response to determining that a node activation operation has been performed, the value determination traversal path set is optimized to obtain an optimized value determination traversal path set. In response to determining that the graph traversal matching has ended, the target product value determination information is generated according to the value determination pruned dependency graph or the optimized value determination traversal path set. The control unit is configured to dynamically render and bind a product configuration terminal to obtain a vehicle product configuration page based on the vehicle product information and the vehicle product value information, and to control a resource control terminal to perform resource scheduling operations based on the vehicle product configuration page, wherein the resource scheduling operations include at least one of the following: transportation of vehicle parts and transportation of the complete vehicle.

7. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-5.

8. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-5.

Images