A charging service reservation scheduling method based on three-dimensional dynamic weighted scoring and related equipment

Abstract

This application provides a charging service reservation and scheduling method and related equipment based on a three-dimensional dynamic weighted scoring system, belonging to the field of new energy vehicle charging scheduling and energy management technology. The application includes: calculating a three-dimensional index score; the three-dimensional index score includes a user credit score, a demand urgency score, and an economic value score; dynamically configuring weights based on the charging station's operational status data to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight; weighting and fusing the three-dimensional index scores based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score; sorting the three-dimensional dynamic weighted scores of all charging request users, and executing charging resource scheduling decisions based on the sorting results. This application can pre-process complex calculations into a parallel scoring system, simplifying the final decision into an efficient weighted summation and sorting, reducing charging pile vacancy and improving resource utilization efficiency.

Description

Technical Field

[0001] This application relates to the field of new energy vehicle charging scheduling and energy management technology, and in particular to a charging service reservation scheduling method and related equipment based on three-dimensional dynamic weighted scoring. Background Technology

[0002] In related technologies, service areas generally adopt a "first-come, first-served" on-site queuing charging mode. This mode has revealed its inherent defects during peak traffic periods: First, rigid resource allocation can lead to safety accidents caused by vehicles running out of power; second, poor user experience can easily cause "charging anxiety" and long waiting times, reducing the overall efficiency and quality of highway travel; third, low operational efficiency, as the utilization of charging piles depends entirely on randomly arriving vehicles, lacking global scheduling, resulting in low overall utilization.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this application is to propose a charging service reservation and scheduling method and related equipment based on three-dimensional dynamic weighted scoring, which can minimize the vacancy of charging piles and improve resource utilization efficiency.

[0005] To achieve the above objectives, one aspect of this application proposes a charging service reservation and scheduling method based on three-dimensional dynamic weighted scoring, the method comprising the following steps:

[0006] Calculate a three-dimensional indicator score; the three-dimensional indicator score includes user credit score, demand urgency score, and economic value score; The weights are dynamically configured based on the operational status data of the charging stations to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight. The three-dimensional index scores are weighted and fused based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score. The three-dimensional dynamic weighted scores of all users requesting charging are sorted, and charging resource scheduling decisions are made based on the sorting results.

[0007] In some embodiments, the calculation of the three-dimensional index score includes: Acquire user's historical reservation behavior data, current charging request urgency data, and order economic value data; Calculate the user's credit score based on the aforementioned historical booking behavior data; Calculate the urgency score based on the urgency data of the current charging request; Based on the economic value data of the orders, calculate the economic value score; The user credit score, the urgency score, and the economic value score are used as the three-dimensional indicator scores.

[0008] In some embodiments, calculating a user credit score based on the historical appointment behavior data includes: Based on the historical reservation behavior data, the type of fulfillment result for each historical reservation behavior is determined according to preset rules; the type of fulfillment result includes cancellation, attendance, and no-show; the historical reservation behavior data includes reservation result, arrival delay, cancellation advance, whether it is a peak period, system fault identifier, site reliability, weather severity, and sample validity period; Based on the timeliness of the samples, a time decay weight is assigned to each historical booking behavior; Using the time decay weight, the performance result types are weighted and counted, and the probability estimate of the performance result type is obtained through probability smoothing and confidence contraction. Based on the probability estimate, interpretability sub-scores are calculated; the interpretability sub-scores include performance sub-scores, on-time sub-scores, cancellation sub-scores, no-show sub-scores, and reservation sub-scores; The interpretability sub-scores are weighted and aggregated according to preset weights to obtain the original credit score. Calculate the uncertainty of the original credit score; Based on the aforementioned uncertainty, the original credit score is scaled back to the platform's benchmark credit score to obtain the final user credit score.

[0009] In some embodiments, calculating the urgency score based on the urgency data of the current charging request includes: Based on the urgency data of the current charging request, a basic urgency is calculated using a piecewise function; the urgency data of the current charging request includes the vehicle's current state of charge, battery capacity, average power consumption rate, remaining driving range, travel demand intensity, and task tag; Based on the remaining mileage and the preset safe mileage threshold, the urgency of the remaining mileage is calculated using an S-shaped function. Based on the task tags, it is confirmed whether the vehicle meets the preset priority scheduling conditions, and the task urgency of the vehicle is obtained. The basic urgency, the remaining mileage urgency, and the task urgency are weighted and summed to obtain the demand urgency score.

[0010] In some embodiments, calculating the economic value score based on the economic value data of the order includes: The estimated charging time is calculated based on the economic value data of the order, and the charging time factor is calculated based on the estimated charging time using a normalization function; the economic value data of the order includes the charging energy requested by the user, the power of the charging pile, the user's membership level, and the percentage of additional premium paid by the user for the current order; Set a membership value factor based on the user's membership level; Calculate the dynamic premium factor based on the percentage of additional premium paid by the user for the current order; The economic value score is obtained by weighting and summing the charging duration factor, the membership value factor, and the dynamic premium factor.

[0011] In some embodiments, the step of dynamically configuring weights based on the operating status data of the charging station to obtain three-dimensional weight information includes: Set benchmark weights; the benchmark weights include user credit benchmark weights, demand urgency benchmark weights, and economic value benchmark weights; Obtain operational status data of charging stations; the operational status data of charging stations includes available power and actual power demand of the station; The system load rate of the charging station is calculated based on the operating status data of the charging station; the system load rate is the ratio of the actual power demand to the available power of the station. Set weight adjustment conditions; The baseline weights are dynamically adjusted and normalized based on the weight adjustment conditions and the operational status data of the charging stations to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight.

[0012] In some embodiments, the step of weighting and fusing the three-dimensional index scores based on the three-dimensional weight information to obtain a three-dimensional dynamically weighted score includes: The three-dimensional indicator scores are normalized to obtain normalized three-dimensional indicator scores; the normalized three-dimensional indicator scores include normalized credit score, normalized urgency score, and normalized economic value score. The normalized three-dimensional index score and the three-dimensional weight information are used to perform weighted fusion to obtain a three-dimensional dynamic weighted score.

[0013] To achieve the above objectives, another aspect of this application proposes a charging service reservation and scheduling system based on three-dimensional dynamic weighted scoring, used to implement the aforementioned method. The system includes: The first module is used to calculate a three-dimensional indicator score; the three-dimensional indicator score includes a user credit score, a demand urgency score, and an economic value score. The second module is used to dynamically configure weights based on the charging station's operational status data to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight. The third module is used to perform weighted fusion of the three-dimensional index scores based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score. The fourth module is used to sort the three-dimensional dynamic weighted scores of all charging request users and to make charging resource scheduling decisions based on the sorting results.

[0014] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0016] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0017] The embodiments of this application include at least the following beneficial effects: This application provides a charging service reservation and scheduling method, system, electronic device, storage medium, and program product based on three-dimensional dynamic weighted scoring. The solution includes: calculating a three-dimensional index score; the three-dimensional index score includes a user credit score, a demand urgency score, and an economic value score; dynamically configuring weights based on the charging station's operational status data to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight; weighting and fusing the three-dimensional index scores based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score; sorting the three-dimensional dynamic weighted scores of all charging request users, and executing charging resource scheduling decisions based on the sorting results. The embodiments of this application can pre-process complex calculations into a parallel-processable scoring system, simplifying the final decision into an efficient weighted summation and sorting, thereby reducing charging pile vacancy and improving resource utilization efficiency. Attached Figure Description

[0018] Figure 1 This is a flowchart of the charging service reservation and scheduling method based on three-dimensional dynamic weighted scoring provided in the embodiments of this application; Figure 2 This is an overall flowchart provided in the embodiments of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of systems and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0022] 1) SOC, State of Charge, indicates the percentage of remaining charge in the battery of an electric vehicle relative to its total capacity; 2) The S-shaped function, the Sigmoid function, has a graph that is a smooth "S"-shaped curve that transitions from 0 to 1; 3) Min-Max scaling, also known as minimum-maximum normalization or deviation normalization, is used to linearly transform the original data to a specific range (usually [0,1]). 4) Z-Score standardization, standard deviation standardization, or zero mean standardization: Data after Z-Score standardization has a mean of 0 and a standard deviation of 1. 5) Beta distribution, a continuous probability distribution defined in the interval [0,1], is determined by two positive parameters α (alpha) and β (beta).

[0023] In related technologies, although the scheduled charging function has been successfully verified in urban road networks, its direct application in highway scenarios is expected to bring new challenges: for example, if users default or are late after making a reservation, valuable charging resources will be wasted; if a simple "first come, first served" rule is used, it will be impossible to distinguish user value, which will prevent high-credit users, members or users with high willingness to pay from obtaining priority services that match them, thereby weakening their willingness to use it.

[0024] In view of this, this application provides a smart charging queuing scheduling method specifically designed for highway service areas (a charging service reservation scheduling method based on three-dimensional dynamic weighted scoring). This application does not simply introduce a reservation function, but aims to construct a comprehensive evaluation and dynamic scheduling mechanism based on multiple factors such as user creditworthiness, urgency of demand, and willingness to pay a premium. This solution can intelligently sort and allocate resources when a user initiates a reservation, thereby achieving a leap from "passive waiting" to "active control." Its advantages are: it can minimize the vacancy of charging piles and improve resource utilization efficiency; it guides users to fulfill their obligations through positive incentives, optimizing user behavior; and it can meet diverse needs by providing tiered services, thereby improving the platform's service quality and profitability.

[0025] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0026] Figure 1 This is an optional flowchart of a charging service reservation and scheduling method based on three-dimensional dynamic weighted scoring provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S104.

[0027] Step S101: Calculate the three-dimensional indicator score; the three-dimensional indicator score includes user credit score, demand urgency score, and economic value score. Step S102: Dynamically configure weights based on the charging station's operational status data to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight. Step S103: The three-dimensional index scores are weighted and fused according to the three-dimensional weight information to obtain the three-dimensional dynamic weighted score; Step S104: Sort the three-dimensional dynamic weighted scores of all charging request users, and make charging resource scheduling decisions based on the sorting results.

[0028] In steps S101 to S104 of the embodiments of this application, step S101 quantifies the core indicators of the three objectives (credit representing fairness and efficiency, urgency representing safety and efficiency, and economic value representing benefits), and integrates them in step S103 to maximize the overall utility of the system. Step S102 can identify peak periods (high load) and automatically enter the "ensuring safety and improving efficiency" mode (increasing the urgency weight); it can identify off-peak periods (low load) and enter the "promoting benefits" mode (increasing the economic value weight). This dynamic adaptability improves the intelligence level and resource utilization efficiency of the system. Once the scores and weights are determined, the ranking (S104) and decision-making are highly efficient (second-level response), overcoming the high computational complexity of traditional optimization models.

[0029] In some embodiments, step S101 may include, but is not limited to, steps S111 to S115: Step S111: Obtain the user's historical reservation behavior data, the urgency data of the current charging request, and the economic value data of the order; Step S112: Calculate the user's credit score based on historical appointment behavior data; Step S113: Calculate the urgency score based on the urgency data of the current charging request; Step S114: Calculate the economic value score based on the economic value data of the order; Step S115: Use user credit score, demand urgency score and economic value score as three-dimensional indicators for scoring.

[0030] In some embodiments, steps S111 to S115 clearly define three major data dimensions, and in steps S112 to S114, a calculation method is designed for each dimension, which can simultaneously consider the user's historical credit, current emergency situation, and order economic contribution.

[0031] In some embodiments, step S112 may include, but is not limited to, steps S1121 to S1127: Step S1121: Based on historical reservation behavior data, determine the type of fulfillment result for each historical reservation behavior according to preset rules; the type of fulfillment result includes cancellation, attendance, and no-show; historical reservation behavior data includes reservation result, arrival delay, cancellation advance, whether it is a peak period, system fault identifier, site reliability, weather severity, and sample timeliness; Step S1122: Based on the sample timeliness, assign a time decay weight to each historical booking behavior; Step S1123: Using time decay weights, perform weighted counting of performance result types, and obtain probability estimates of performance result types through probability smoothing and confidence contraction. Step S1124: Calculate the interpretability sub-scores based on the probability estimates; the interpretability sub-scores include performance sub-scores, on-time sub-scores, cancellation sub-scores, no-show sub-scores, and reservation sub-scores. Step S1125: The interpretability sub-scores are weighted and aggregated according to preset weights to obtain the original credit score; Step S1126: Calculate the uncertainty of the original credit score; Step S1127: Based on the uncertainty, the original credit score is scaled back to the platform's benchmark credit score to obtain the final user credit score.

[0032] In steps S1121 to S1127 of some embodiments, the scoring can quickly reflect the latest changes in user behavior, providing timely feedback for users whose behavior has improved or deteriorated. Based on a set of rules (such as grace period G, arrival buffer B, and early cancellation threshold Tearly), a user's booking journey is determined as a clear fulfillment outcome type. Environmental factors such as fault exemption (fault_flag) and site reliability are introduced to reflect the fairness of the system—users are not penalized due to platform problems. A weight is assigned to each historical behavior sample, with time decay weight controlling the rate of forgetting; events that happened yesterday (age_days=1) have a much higher weight than events from a year ago (age_days=365). Probability smoothing introduces a beta distribution prior, which can effectively prevent extreme probability estimation under small samples. Credibility contraction shrinks the smoothed probability towards a more stable platform baseline rate, so that the system will not make absurd or extreme judgments when facing new users or users with sparse data, improving the system's versatility and reliability. Step S1124 converts the probability value into an easily understandable score and breaks it down into specific dimensions to achieve scoring transparency. Users and operators can see not only an overall score, but also a clear view of user performance across various dimensions such as "punctuality," "cancellation habits," and "resource utilization," facilitating problem diagnosis and behavioral guidance. Step S1125 integrates multi-dimensional insights into a comprehensive score, with the weighting reflecting the platform's value orientation and business strategy for different credit dimensions. Steps S1126-S1127 quantify the uncertainty of the scoring based on Bayesian posterior variance. The less data available and the more volatile the behavior, the greater the variance and the higher the uncertainty, ensuring the overall robustness and fairness of the scoring system.

[0033] In some embodiments, step S113 may include, but is not limited to, steps S1131 to S1134: Step S1131: Based on the urgency data of the current charging request, calculate the basic urgency using a piecewise function; the urgency data of the current charging request includes the vehicle's current state of charge, battery capacity, average power consumption rate, remaining driving range, travel demand intensity, and task label. Step S1132: Based on the remaining mileage and the preset safe mileage threshold, calculate the urgency of the remaining mileage using an S-shaped function; Step S1133: Based on the task tag, confirm whether the vehicle meets the preset priority scheduling conditions to obtain the task urgency of the vehicle. Step S1134: The basic urgency, remaining mileage urgency, and task urgency are weighted and summed to obtain the demand urgency score.

[0034] In steps S1131 to S1134 of some embodiments, the basic urgency is calculated using a piecewise function. An exponential function can be used in the high-risk zone (SOC < 15%) because the battery voltage may drop sharply when the battery is low, causing vehicle driving uncertainty and safety risks to increase exponentially. This function can non-linearly amplify this risk, issuing a strong warning to the system. In the low-risk zone (SOC ≥ 15%), an approximately linear function is used. In this range, the battery is sufficient, and the risk changes gradually; a linear function is sufficient to describe the trend, while avoiding unnecessary scheduling interference for vehicles with high battery levels. The remaining mileage urgency is calculated using an S-shaped function. The S-shaped function provides a smooth and continuous change in urgency near the safe mileage threshold, avoiding fractional jumps at the threshold point, thus preventing drastic fluctuations in scheduling order due to small mileage fluctuations. Task urgency is confirmed based on task tags, incorporating the vehicle's social attributes into scheduling considerations, allowing the system to go beyond simple physical state judgment. This embodiment provides a "user declaration channel" combined with an "automatic identification + manual review" audit mechanism, ensuring both response speed and authority, and preventing resource abuse. The weighted summation yields the final urgency score. Under normal circumstances, the three factors are balanced. When resources are extremely scarce, the weights of the base urgency and remaining mileage urgency can be increased through dynamic weight configuration to emphasize the physical risks of the vehicle itself.

[0035] In some embodiments, step S114 may include, but is not limited to, steps S1141 to S1144: Step S1141: Calculate the estimated charging time based on the economic value data of the order, and calculate the charging time factor based on the estimated charging time using a normalization function; the economic value data of the order includes the charging energy requested by the user, the power of the charging pile, the user's membership level, and the percentage of additional premium paid by the user for the current order; Step S1142: Set the membership value factor based on the user's membership level; Step S1143: Calculate the dynamic premium factor based on the percentage of additional premium paid by the user for the current order; Step S1144: The charging duration factor, membership value factor, and dynamic premium factor are weighted and summed to obtain the economic value score.

[0036] In steps S1141 to S1144 of some embodiments, the efficiency of order utilization of the platform's scarce resource—time—is evaluated to guide resources towards efficient turnover. The estimated charging time measures the physical time an order will occupy the charging station, and a normalized function (S-shaped function) is used to reflect the principle of diminishing marginal utility. For orders with excessively long durations (such as exceeding a preset inflection point), the marginal revenue brought to the platform by each minute occupied is diminishing, and it also hinders the access of other orders, resulting in high opportunity costs. Therefore, this embodiment can apply a value discount using an S-shaped function. The membership value factor can capture a user's long-term economic contribution and loyalty, reflecting the user's charging history and integrity. A VIP user with a high number of historical purchases is given a higher weight to encourage upgrades and maintain integrity. The dynamic premium factor incorporates the user's direct payment intention shown in the current order into the evaluation system, maximizing revenue during periods of resource scarcity.

[0037] In some embodiments, step S102 may include, but is not limited to, steps S201 to S205: Step S201: Set benchmark weights; benchmark weights include user credit benchmark weight, demand urgency benchmark weight, and economic value benchmark weight. Step S202: Obtain the operating status data of the charging station; the operating status data of the charging station includes the available power of the station and the actual power demand. Step S203: Calculate the system load rate of the charging station based on the operating status data of the charging station; the system load rate is the ratio of actual power demand to the available power of the station. Step S204: Set weight adjustment conditions; Step S205: The baseline weights are dynamically adjusted and normalized according to the weight adjustment conditions and the operating status data of the charging station to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight and economic value weight.

[0038] In steps S201 to S205 of some embodiments, baseline weights are set to establish the basic principles and default behavior patterns of system scheduling, ensuring the stability and consistency of the strategy under normal conditions. The weight adjustment conditions include: increasing urgency weights (due to resource scarcity, priority must be given to vehicles with low battery to prevent breakdowns, improve user experience and safety), increasing economic value weights (due to resource availability, the main goal is to maximize revenue per unit time), and weight reset conditions due to extreme weather. Through normalization and boundary constraints, weight adjustments prevent model collapse and ensure reasonable output results.

[0039] In some embodiments, step S103 may include, but is not limited to, steps S301 to S302: Step S301: Normalize the three-dimensional indicator scores to obtain normalized three-dimensional indicator scores; the normalized three-dimensional indicator scores include normalized credit score, normalized urgency score and normalized economic value score. Step S302: The normalized three-dimensional index scores and three-dimensional weight information are used to perform weighted fusion to obtain a three-dimensional dynamic weighted score.

[0040] In steps S301 to S302 of some embodiments, the scores of the three dimensions (e.g., credit score C, urgency U, economic value E) may have different dimensions and numerical ranges. By mathematical transformation (e.g., Min-Max scaling, Z-Score standardization, etc.), the scores of the three dimensions can be uniformly mapped to the same numerical range, which can create a fair premise for weighted fusion.

[0041] This application also provides a charging service reservation and scheduling system based on three-dimensional dynamic weighted scoring to implement the method described above. The system includes: The first module is used to calculate the three-dimensional indicator score, which includes user credit score, demand urgency score, and economic value score. The second module is used to dynamically configure weights based on the charging station's operational status data to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight. The third module is used to perform weighted fusion of the three-dimensional index scores based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score. The fourth module is used to sort the three-dimensional dynamic weighted scores of all charging request users and make charging resource scheduling decisions based on the sorting results.

[0042] As an optional implementation, the system in this application embodiment also includes a user credit scoring module (used to quantitatively assess the user's reliability and willingness to fulfill obligations), a demand urgency assessment module (used to quantify the urgency of the vehicle's current charging needs, aiming to prioritize travel safety), an economic value assessment module (used to quantify the direct and indirect economic contribution of orders to the platform), a dynamic weight configuration module (used to adaptively adjust the relative importance of credit, urgency, and economic value in the final decision in real-time operational scenarios), and a fusion scheduling decision module (used to integrate all information, generate the final scheduling queue, and ensure the interpretability of the decision-making process). The collaborative work of these modules systematically solves the complex multi-objective optimization problem of balancing efficiency, revenue, fairness, and safety under limited resources.

[0043] As an optional implementation, this application relates to the field of new energy vehicle charging scheduling and energy management technology, specifically providing a dynamic weighted scoring method and system based on the integration of credit evaluation, demand urgency, and economic value to optimize the charging reservation and resource scheduling process for electric vehicles. This method is particularly suitable for typical scenarios where multiple vehicles compete for limited charging resources, such as when there are multiple reservation requests within the same time period (e.g., five electric vehicles simultaneously competing for the right to use the same charging pile at 10:00 AM).

[0044] This application aims to systematically solve a complex multi-objective optimization problem: how to coordinate the different service needs of users of different levels (including VIP users, ordinary members and non-members) under the condition of limited charging pile resources, and simultaneously improve the overall operational efficiency and resource utilization of charging stations.

[0045] Its core concept lies in completely abandoning the traditional "first-come, first-served" or static priority single-dimensional decision-making model. Instead, it constructs a multi-dimensional, quantifiable, and dynamically evolving comprehensive evaluation system to generate a differentiated comprehensive priority score for each reservation request, thereby achieving intelligent and refined scheduling decisions. Specifically, this application reconstructs the core decision-making logic of charging scheduling priority by organically integrating and dynamically adapting the weights of three key dimensions: credit status, urgency of demand, and economic value. 1. User Credit Dimension: Quantitatively evaluate users' historical performance (such as on-time arrival rate and cancellation rate) to effectively reduce the vacancy rate of charging pile resources caused by user defaults or lateness, and improve the actual utilization efficiency of system resources by introducing a credit constraint mechanism.

[0046] 2. Urgency of Need Dimension: This dimension comprehensively considers the urgency of the user's current needs, including but not limited to objective indicators such as the vehicle's remaining battery power and the urgency of the user's subsequent journey. This dimension ensures that the system prioritizes responding to truly urgent charging needs, guaranteeing the user's basic travel experience and safety.

[0047] 3. Economic Value Dimension: This dimension incorporates users' economic contributions into the evaluation system, such as membership level, historical spending, and willingness to pay a premium for current orders. This dimension aims to guide scarce charging resources towards high-contributing users, thereby directly improving the overall profitability of the operating platform.

[0048] Based on the above core concepts, the technical problem this application aims to solve is to provide a charging service reservation and scheduling method and system for users of different levels. This method should be based on a three-dimensional dynamic weighted scoring model, achieving synergistic optimization of resource utilization efficiency, differentiated user service experience, and platform operational revenue while ensuring the real-time nature and interpretability of scheduling decisions. Specifically, this application aims to solve the following sub-technical problems: 1. Establish a quantitative and transparent priority assessment model: How to design specific quantitative indicators, standardized methods, and dynamic weight allocation strategies for the three dimensions of credit, urgency, and economic value to overcome the rigidity of rule-based methods and generate a comprehensive, fair, and interpretable overall priority score.

[0049] 2. Achieve efficient and real-time scheduling decisions: How to combine dynamic scoring models with efficient scheduling algorithms (such as priority-based queue management) to ensure that the system can respond quickly (in seconds) to massive concurrent reservation requests and overcome the high computational complexity of optimization models.

[0050] 3. Balancing multi-objective optimization and fairness: How to dynamically adapt operational objectives in different scenarios (such as ensuring efficiency during peak periods and promoting revenue during off-peak periods) by adjusting the weight configuration of the three dimensions, while meeting the service expectations of high-value users and users with urgent needs, and avoiding systemic discrimination against low-credit or ordinary users, thus maintaining basic fairness.

[0051] 4. Enhance the system's robustness to uncertainty: How to use user credit dimensions as a reference to predict their future performance behavior (such as the probability of on-time arrival) so that the scheduling strategy can inherently cope with the uncertainty of user arrival time, thereby improving the overall robustness and reliability of the system.

[0052] The technical solution of this application will be described in detail below with reference to specific implementation methods. For ease of understanding, the content of this application will be unfolded according to functional modules, including a user rating module, an urgency assessment module, an economic value assessment module, and a fusion scheduling rating module. Through a hierarchical and modular approach, the overall concept and core innovations of this application can be made clearer and more explicit, thereby facilitating implementation and reproduction by those skilled in the art.

[0053] I. User Credit Scoring System (Credit): This application systematically describes the implementation scheme of a user credit scoring system (execution step S112). The scheme first clarifies the system's data input sources and business tag definitions, laying the foundation for core calculations. Subsequently, it focuses on introducing professional data processing methods for rate-related indicators, including smoothing using a Beta-binomial distribution and applying credibility theory for shrinkage correction to enhance statistical stability. Based on this, the document details the calculation of sub-scores, the synthesis logic of the comprehensive credit score, methods for handling data uncertainty, and an interpretable output mechanism designed to ensure model transparency.

[0054] 1. Core parameter definition: Refer to Table 1 for the output items included in the system.

[0055] Table 1. System Output

[0056] Refer to Table 2 for the system's input content.

[0057] This application provides the following outcome determination rule: Let the start time of the reservation plan be... The actual charging start time is (If none, leave blank), cancellation time is (If none, then empty). The following are the relevant parameters and rules for the determination: = Grace period (default 5 minutes) = Arrival buffer (default 10 minutes). = Early cancellation threshold (default 30 minutes).

[0058] (1) Fault exemption: when At that time, negative labels are not included in the penalty, and only positive statistics are recorded; however, labels are still generated according to the rules to retain records.

[0059] (2) Cancel: If Calculate the amount of early cancellations: ; And it will be determined according to the following rules: (Cancel in advance); (Canceled too late).

[0060] (3) Attendance: If Then calculate the late time: ; And it will be determined according to the following rules: (Complete on time and fulfill contractual obligations); (Delayed completion).

[0061] (4) No-show: If the event is not cancelled and the person does not attend, and Then it is determined as follows: (Breaking the promise); Boundary specification: If If so, it is considered as not cancelled (entry / no-show judgment).

[0062] 2. Timeliness Weighting and Robust Estimation: To give more weight to recent behavior in the overall evaluation results, this application first introduces the timeliness weight and the time decay weight in robust estimation, the calculation formula of which is as follows: ; To balance timeliness and stability in the statistical process of event behavior, this application performs weighted counting and frequency calculation on any target event type (i.e., performance outcome type, such as cancellation event). The definitions are as follows: ; ; ; in, It is a key parameter for controlling the rate at which the weights decay over time; Representing the How many days have passed since the current time for each sample? The timeliness weight of the event samples; This is an indicator function used to determine the first... Does each sample belong to this event type? It is the sum of the weights of all samples (regardless of whether they are the target event); It is the proportion of the weighted count of the target event to the total weighted sample size.

[0063] Table 2. System Inputs

[0064] Through the aforementioned weighting process, this application assigns higher weight to recent events when calculating frequency, thereby avoiding the situation where excessive historical samples mask the latest behavioral patterns. Simultaneously, this scheme balances the contributions of samples across different time spans, making the obtained probability estimates more consistent with the user's current actual behavioral characteristics, thus improving the accuracy and adaptability of prediction and judgment.

[0065] To avoid extreme results due to insufficient sample size during event frequency estimation, this application introduces a Beta-binomial smoothing method. Specifically, it employs a Beta distribution prior. The smoothed estimate is defined as follows: ; in, It is a virtual number of positive samples; This is a virtual negative sample number. By introducing prior parameters in addition to the actual samples, this scheme can effectively add "pseudo-samples" to the sample set, avoiding the probability estimate from degenerating directly into extreme values ​​of 0 or 1 under small sample conditions. This method not only improves the robustness of the model under sparse samples, but also ensures that the estimation results are more statistically reasonable and smooth, thereby enhancing the applicability and generalization ability of the system.

[0066] In estimating event frequency, to balance individual data with the overall baseline, this application further introduces a credibility contraction mechanism. Specifically, it makes... For platform benchmark rate, The confidence coefficient (a larger coefficient indicates greater reliance on the platform average) is then defined as follows: ; This mechanism can adaptively adjust the source of the estimation results based on the sample size. When the sample size N is large, the estimation results are mainly dominated by individual data, making... This ensures the diversity and accuracy of individual behaviors; however, when the sample size N is small, the estimation results tend to reflect the overall platform baseline rate, making... This effectively avoids the instability of estimation results under small sample sizes. Therefore, this scheme achieves a balance between robustness and statistical rationality while ensuring individualized characterization.

[0067] To further enhance the stability and interpretability of the model, this application, based on the aforementioned estimation method, configures reasonable prior distributions and platform benchmark rates for various ratio-type indices. Specifically, a Beta distribution is used as the prior, and its parameter settings and corresponding benchmark rates are shown in Table 3.

[0068] Table 3. Parameter Settings and Corresponding Base Rates

[0069] By introducing the aforementioned prior and benchmark rate configurations into the model, parameters can be flexibly adjusted under different city, site, or time period conditions, thereby further improving the robustness and adaptability of the estimation.

[0070] 3. Calculation of interpretable sub-components: (1) Performance Sub-points: This application first calculates a performance sub-score based on the performance probability to reflect the user's actual performance level under a given reservation plan. By directly mapping the performance probability to a score, this scheme can highlight the importance of user performance behavior in quantitative evaluation. This method not only avoids estimation bias caused by insufficient samples but also ensures that the scoring results remain comparable and consistent across different users. The specific calculation formula is as follows: ; in, This score represents the fulfillment probability after weighting by timeliness, prior smoothing, and credibility contraction. Using this scoring method, differences in user fulfillment behavior become clearly apparent. Users with high fulfillment rates will receive near-perfect fulfillment sub-scores, thus gaining higher credit in the overall assessment; while users with low fulfillment rates will be constrained by lower sub-scores. This mechanism effectively guides users to improve their fulfillment rates, thereby enhancing platform operational efficiency and resource utilization.

[0071] (2) Precise point sub-scores (only for on-site samples): This application is for samples that have already been presented (including...) FULFILLED and LATE_FULFILLED The system assesses punctuality and maps lateness to score deductions to characterize users' adherence to appointment schedules. This sub-score distinguishes between "fulfilling appointments but not being punctual" situations within the overall scoring system, thus reflecting service order and resource allocation efficiency in a more granular way.

[0072] First, this application sets segmented penalty points for lateness, defining a base penalty point for each arriving sample based on the duration of lateness (in minutes). : ; A monotonically segmented, tiered penalty system is adopted to balance interpretability and execution stability: minor lateness (≤5 minutes) is considered a tolerable disturbance and no points are deducted; moderate (5–15) and severe lateness (15–30) are subject to progressively stricter fixed penalty points to avoid over-amplifying single-point anomalies; major lateness (>30) has a capped penalty to prevent extreme values ​​from dominating the overall evaluation. In real-world operations, this design effectively suppresses the impact of long-tail lateness on the mean, while providing users with clear behavioral boundaries and directions for improvement.

[0073] By introducing a multiplicative context factor into the base penalty, we obtain the modified penalty: ; The negative externalities of being late during peak hours are greater (occupancy, queuing, chain congestion), so their impact should be appropriately amplified; when station stability is insufficient, penalties for the same lateness should be reduced to reflect the impact of factors controllable by the platform; under severe weather conditions, the intensity of penalties should be reduced to ensure fairness. Contextualized correction enables dynamic adaptation to external objective conditions, reducing two types of distortions: "over-penalty due to uncontrollable factors" and "under-penalty for lateness with high externalities."

[0074] Time-sensitive weights were applied to the samples presented. By performing a weighted average, we obtain the comprehensive penalty score at the sample level: ; This application employs a time-weighted framework consistent with the overall methodology, ensuring that recent behavior is more representative of the current assessment, while maintaining consistency with other sub-components to improve consistency and comparability within the system. This applies when behavioral patterns change (e.g., recent improvements). It can reflect the latest status more quickly and has better sensitivity and response speed.

[0075] Finally, the total penalty points are mapped to the exact sub-points: ; With 100 points as the maximum score, negative scores will be awarded. The deduction is applied directly to ensure linearity and intuitiveness; at the same time, a lower limit of 40 points is set to avoid excessive penalties when the sample size is limited or there are short-term fluctuations, ensuring the robustness and operability of the evaluation. The final score can effectively distinguish between different levels of lateness, while avoiding excessively low scores due to occasional occurrences, thus stabilizing the expected incentives for users and the allocation of platform resources.

[0076] (3) Cancel sub-points: This application differentiates user cancellation behavior, distinguishing between early and late cancellations, and achieves fair evaluation through a weighted scoring system. To avoid double counting, the following equivalent formula is used for calculation: ; ; Late cancellations have a greater impact on platform resource allocation, therefore they are assigned a weight of 100; early cancellations, while disrupting user experience, have a relatively minor impact, and are assigned a weight of 60; at the same time, double counting is ensured to avoid excessive penalties. This scheme can distinguish the severity of different cancellation behaviors, incentivizing users to cancel as early as possible to reduce platform losses, while ensuring the fairness and rationality of the scoring.

[0077] (4) No-show sub-cents: To address situations where users fail to fulfill or cancel their reservations, this application establishes a no-show sub-score to reflect the direct impact of such behavior on platform operations. The calculation formula is as follows: ; No-show behavior directly leads to resource idleness and operational losses. A linear penalty is used to ensure that the score decreases proportionally to the no-show rate. Furthermore, to prevent frequent no-shows from masking the problem through averaging, this solution adds a combo penalty: if the number of no-shows in the most recent M=5 times is ≥ 2, the sub-score is multiplied by an additional 0.85 to further constrain high-frequency no-show behavior. This mechanism can significantly reduce the incidence of high-frequency no-shows, ensure resource utilization and fairness among users, and improve the overall reliability of the platform's services.

[0078] (5) Occupying a slot (only for performance samples): For situations where resources are underutilized despite fulfillment of obligations, this application sets up a "slot-holding score" to measure the effective utilization rate of reserved resources. First, the effective utilization rate of a single fulfillment sample is defined. : ; in, and These represent the weights of time utilization and energy utilization, respectively.

[0079] Set threshold If r ∈ [0,1] (e.g., 0.5), then: ; Obtained through smooth contraction Then, the points for occupying the pit are calculated as follows: ; While reserving a spot isn't entirely equivalent to cancellation or no-show, it still negatively impacts resource utilization efficiency. Therefore, by using effective utilization rates and thresholds to determine underutilization, appropriate deductions are made from the score. This mechanism encourages users to use reserved resources rationally, avoiding situations of "formal fulfillment but actual waste," ultimately improving overall resource utilization efficiency and user experience.

[0080] 4. Overall credit score and uncertainty reduction: (1) Weighting and Aggregation: This application generates an initial value for the comprehensive credit score by weighting and aggregating the sub-scores. The sub-score weights are set to satisfy... Each weight can be adjusted according to business needs. An example setting is as follows: ; ; in, Weighting of sub-company members for performance. Weighting is done based on the accuracy of the points. To cancel sub-weights, Weighting of no-shows The weighting of "slot reservation" behavior is determined by a weighted aggregation method. This method balances the impact of different behavioral dimensions, with fulfillment, punctuality, cancellation, and no-shows being the primary dimensions, and "slot reservation" behavior being secondary. This ensures the comprehensiveness and reasonableness of the final score. Through weighted aggregation, the overall credit score can comprehensively reflect a user's multi-dimensional behavioral performance, avoiding score distortion caused by fluctuations in a single indicator.

[0081] (2) Baseline score: To enhance the robustness of the scoring, this application constructs a baseline score based on the platform's overall benchmark rate. The baseline values ​​for each sub-score are defined as follows: ; ; ; ; ; in .

[0082] ; in, This is the baseline value for the performance sub-unit. The baseline value of the standard point sub-score, To cancel the baseline value of the sub-segment, This is the baseline value for the no-show sub-value. This is the baseline value for occupying the pit. This is an estimate of the probability of performance. To cancel the probability estimate in advance, This is an estimate of the probability of late cancellation. This is an estimate of the probability of no-show. This is an estimate of the probability of occupying a slot. To cancel the total estimate. The baseline value for type k (e.g., performance sub-segment, on-time sub-segment, etc.). The weights are of type k. The platform baseline score serves as a reference for the overall level, providing a stable reference point when the sample size is insufficient or individual scores fluctuate significantly. This mechanism ensures fairness in comparability among different users, and even in data-sparse scenarios, the scoring results will not deviate excessively from the industry or platform average level.

[0083] (3) Uncertainty With retraction strength : To overcome the scoring estimation bias caused by small sample size or behavioral fluctuations, this application introduces an uncertainty measure based on Bayesian posterior variance and a corresponding shrinkage mechanism to improve the robustness of the scoring system.

[0084] The core idea is that when individual behavioral data is insufficient or performance is unstable, the reliability of the original score is low. Therefore, the system introduces a prior platform benchmark score (anchor point). The reliability of the original data and the benchmark anchor point is dynamically adjusted by calculating the uncertainty: when the uncertainty is high, the score results converge towards the platform benchmark anchor point to avoid inaccurate assessments due to accidental behavior; when the uncertainty is low, the score calculated from the original behavioral data is primarily adopted to preserve individual specificity. The specific implementation steps are as follows: First, the uncertainty of the assessment is quantified using the variance of the Beta posterior distribution in Bayesian statistics. For each core behavior type (such as fulfillment, cancellation, delayed cancellation, no-show, etc.), the variance is calculated as follows: ; The posterior parameters are determined by both the observed data and prior knowledge. ; The average variance is obtained by combining the variances of each core behavioral type. And normalize to obtain the final uncertainty: ; Here It is between The larger the scalar value, the higher the overall uncertainty. Actual pullback strength. From the maximum shrinkage coefficient and uncertainty The product of these two products eventually shrinks: ; in, The original credit score. This serves as a benchmark score for the platform. Uncertainty reflects sample size and volatility. Smaller sample sizes or greater volatility result in higher variance, and consequently, greater uncertainty. By controlling the intensity of the pullback, the score will gradually approach the platform's baseline score, avoiding instability due to insufficient data. This pullback mechanism can improve the robustness of the scoring system in small sample and high-volatility scenarios while ensuring individual differences, making the overall credit score both sensitive to real behavior and not overly affected by random events.

[0085] II. Urgency of Demand: This module reflects the user's current charging urgency (execution step S113) and is one of the key factors in the scheduling process. Urgency is determined not only by the battery level but also by factors such as remaining driving range, estimated waiting time, and task attributes. The core task of this module is to calculate a numerical value (comprehensive urgency index) to quantify the "urgency" of each vehicle needing charging. The scheduling system will then prioritize all vehicles based on this "urgency," arranging charging for those with higher urgency levels.

[0086] The core parameters are defined as shown in Table 4.

[0087] (1) Basic urgency function driven by SOC: This application focuses on the mapping relationship between vehicle battery status and mission urgency. It is important to emphasize that this relationship is not a simple linear correspondence, but rather a curve with phased characteristics, particularly showing a significant upward trend in risk during the low battery range. When the vehicle's remaining battery level is below 15%, the vehicle may break down at any time due to insufficient energy, resulting in a high risk of mission interruption or failure. Based on this, this application sets the sense of urgency to exhibit an exponentially rapid increase during the low battery range, highlighting the reality of a sharp increase in risk at low battery levels. When the vehicle's battery level is above 15%, its operational safety and sustainability are generally guaranteed, and the risk level is low. Therefore, within this range, the sense of urgency decreases gradually and approximately linearly as the battery level increases, reflecting the trend of reduced risk at high battery levels.

[0088] ; in It is the attenuation constant. , The normalization coefficient is used. Through the above design, the urgency function constructed in this application can more realistically reflect the task risk characteristics under different power ranges: it strengthens the sudden constraint at the low power critical point to avoid the scheduling system ignoring potential breakdown risks; and it remains stable in the high power range to avoid excessive interference to resource scheduling, thereby improving the overall scheduling safety and rationality.

[0089] Table 4. Core Parameter Definitions

[0090] (2) Remaining mileage driving function: This application further introduces the vehicle's remaining mileage as a urgency-driven factor. Considering that a vehicle's remaining mileage being too low during mission execution could prevent it from completing the mission or returning to a refueling point, risk modeling of the remaining mileage is necessary during the scheduling process. Specifically, this application sets a safe mileage threshold. (For example, 50 kilometers). When the remaining mileage of the vehicle is much higher than this threshold, the risk of task execution is low, and the urgency remains at a low level; however, when the remaining mileage is close to or lower than the threshold, the risk rises rapidly, and the urgency of vehicle dispatching increases significantly. To avoid abrupt changes at the threshold, this application uses a sigmoid function to smoothly describe this relationship. This function has an "S"-shaped curve, which is sensitive to changes in mileage near the threshold, but maintains a smooth transition overall, thus effectively avoiding drastic fluctuations in the sorting results due to slight variations.

[0091] ; in This represents the vehicle's remaining driving range. The safe mileage threshold (e.g., 50 km); Parameters for controlling the width of the transition zone.

[0092] This function enables the scheduling system to issue an urgent signal in a timely manner when the remaining mileage is about to run out, allowing for early intervention in resource allocation; at the same time, it maintains stability when the remaining mileage is sufficient, avoiding ineffective scheduling due to oversensitivity. Overall, it improves the system's ability to ensure task completion rate and operational safety.

[0093] (3) User application channel: To ensure that special mission vehicles (such as emergency vehicles, public transport vehicles, and government vehicles) can receive priority access to energy and dispatch support in emergency situations, this application has designed a user application channel mechanism.

[0094] Specifically, the platform provides an "Emergency Mission" application portal on the client side, allowing users to directly submit vehicle mission types (such as emergency medical services, public transportation support, law enforcement, and official duties) when necessary. Upon receiving an application, the platform uses a combination of automatic identification and manual review for verification. Vehicles that pass the review will have their mission tag weighted accordingly. It will be dynamically increased to the 0.8–1.0 range to significantly improve its priority in scheduling. This priority only takes effect during task execution and automatically reverts to its normal weight after the task is completed. The rationale for this setting can be summarized as follows: 1. Emergency Response Support: Through the application channel mechanism, the platform can quickly identify vehicles with emergency missions and ensure that they receive priority support under limited resource conditions.

[0095] 2. Flexible management: The "automatic + manual" review process ensures both response efficiency and avoids false declarations or resource abuse.

[0096] 3. Dynamic Priority: Priority is only increased within the task's validity period to ensure overall system fairness and reasonable resource allocation.

[0097] This mechanism effectively enhances the platform's service capabilities in emergency scenarios, ensuring that emergency vehicles such as ambulances and buses can complete their missions smoothly despite energy shortages. Simultaneously, it avoids prolonged resource occupation in normal operating environments, balancing public interest with platform efficiency, and significantly enhancing the system's reliability and social value.

[0098] The final overall urgency level is: .

[0099] III. Economic Value: This module measures the direct and indirect economic contribution of orders to the platform. In scheduling, orders with high economic value are prioritized to improve overall revenue and resource utilization.

[0100] (1) Charging time effect: This application first considers the impact of order charging time on the platform's economic value. Let the user's required energy be... (kWh), charging power is (kW), the expected charging time is defined as: ; The platform exhibits diminishing marginal value for excessively long orders. To reflect this characteristic, the normalization function is defined as follows: ; in, This is an inflection point (e.g., 90 minutes). This is for transitional bandwidth. The design allows short-term orders (<60 minutes) to receive higher scores, while longer orders receive progressively lower scores.

[0101] (2) Value of Membership: To encourage long-term user contributions, this application sets reward factors for different membership levels, defined as follows: ; in, This parameter reflects the differences in long-term value contribution among different user groups.

[0102] (3) Dynamic premium: If users choose to pay an additional premium during peak periods, the platform can increase their economic value score proportionally to the premium. The definition is as follows: ; in, Additional payment for users, Basic rate, It is a regulating factor.

[0103] (4) Overall economic value score: Taking into account the charging time effect, membership value, and dynamic premium factors, the economic value index is defined as follows: ; in, , This represents the relative importance of each dimension in the overall economic value assessment. . To assess the relative importance of charging time in overall economic value evaluation, To assess the relative importance of membership value in overall economic value evaluation, This relates to the relative importance of dynamic premiums in overall economic value assessment.

[0104] IV. Dynamic Weighted Fusion Scoring Model (Execution Steps S102-S104): In the final scheduling priority, three types of scores need to be integrated: credit ( ), urgency ( ), economic value ( To maintain comparability, the three values ​​were first normalized to obtain... .

[0105] Subsequently, the dynamic weighted fusion scoring model of this application is used to adaptively adjust under different operational scenarios. Three weighting categories are used to balance fairness, profitability, and security. The core idea is: under hard constraints (such as prioritizing low battery levels), a soft adjustment factor is used for real-time weighting.

[0106] (1) Weighting benchmark: Under normal, stable conditions, the recommended baseline weights are: ; That is, it takes into account credit (40%), urgency (30%), and economic value (30%).

[0107] (2) Peak-hour regulation (load constraint driven): If the system load rate If the actual power demand reaches more than 80% of the available power at the site, the urgency weight is increased to ensure the safety of vehicles on the verge of power outages. ; At the same time maintain .in This is the sensitivity coefficient. It enables the prioritization of limited charging resources to vehicles with the most urgent needs (such as those on the verge of running out of power) when resources are extremely scarce. This prevents vehicles from breaking down due to complete power loss, ensuring users' basic travel safety and core experience, and preventing the worst-case scenario of service failure.

[0108] (3) Off-peak period adjustment (yield-driven): when And queue length The platform prioritizes increasing the weight of economic value: ; in As a regulating factor, ensure It can reach around 0.5 during off-peak hours, thereby optimizing the return per unit of time.

[0109] (4) Special event handling (security-driven): The system automatically increases the urgency weight when the following conditions are met: 1. Extreme weather: Temperatures <-15°C or >40°C, with a high risk of battery degradation; 2. Public Safety: Emergency vehicles successfully registered (ambulances, police cars, government vehicles); 3. Regional power grid pressure: The dispatching system requires limiting peak load.

[0110] At this point, set directly: .

[0111] (5) Operational strategy adjustment (manual intervention): The platform management interface allows manual weight adjustments based on strategic objectives: 1. Attract new users and increase user activity: Increase credit weight. This encourages good users to receive faster responses; 2. Increase revenue and improve efficiency: Increase the weight of economic value ; 3. Service backup: Increase the weight of urgency level. To protect vulnerable groups or critical vehicles.

[0112] Manual adjustments must be submitted through the management interface and gradually transitioned within a 15-minute window to avoid sudden changes that could cause sorting oscillations.

[0113] (6) Weight normalization and boundary conditions: Normalization must be performed after each adjustment: ; and restrictions This prevents a certain dimension from being completely ignored.

[0114] (7) Final output: The final output includes the following: 1. Overall score ; 2. Three-dimensional subdivision ; 3. Weight and its dynamic sources; 4. Contribution breakdown: This helps explain why a certain type of factor either increases or decreases the overall score.

[0115] To facilitate understanding of the technical solution of this application, the process is described below. Figure 2 The implementation process is explained in detail.

[0116] The user credit scoring system in this embodiment: 1. Data Input Stage: The platform first receives detailed data on users' appointment events, including the start time of the appointment schedule. Actual arrival time Cancellation time This step obtains essential information such as whether it is during peak hours. This step ensures the integrity of the basic data required for subsequent algorithm execution.

[0117] 2. Sample cleaning stage: Normalize the input data. If the event is marked as a fault exemption ( If the data is positive, only positive records will be retained, and negative labels will not be included in the penalty calculation; at the same time, all timestamps will be standardized to ensure that data from different sources are comparable in the time dimension.

[0118] 3. Time-weighted stage: For different historical samples, a time decay weighting method is used, defined as follows: .

[0119] 4. Weighted Counting Phase: We perform weighted statistics on various events to obtain the total weighted sample size: ; And the weighted number of the corresponding events; ; This step lays the foundation for subsequent probability estimation. A Beta prior distribution is then introduced to avoid extreme probability estimates caused by small samples. .

[0120] 5. Credibility contraction phase: The estimation results are further scaled back to the platform benchmark rate to calculate: .

[0121] 6. Subtotal Calculation Stage: Based on the contracted index probabilities, calculate the sub-fractions for each category: Sub-compliance: ; Accurate score: ; Cancel sub-points: ; No-show bonus: ; Occupying a spot: .

[0122] 7. Aggregation Phase: The original comprehensive score is obtained by weighting and summing the sub-scores according to preset weights. .

[0123] 8. Uncertainty Discount Stage: Uncertainty is calculated using the beta posterior variance. and with strength parameters Discount the raw score to the platform baseline score: .

[0124] 9. Output and Interpretation Stage: The system ultimately outputs the user's comprehensive credit score. At the same time, each sub-part is given The report also includes a breakdown of the contributions of each indicator. Furthermore, it provides trend changes and improvement suggestions to help users and the platform understand the reasons behind score changes.

[0125] The urgency calculation module in this embodiment: 1. Data Input Stage: Collect vehicle SOC, battery capacity, average power consumption rate, and remaining driving range. Task Tags Parameters such as these.

[0126] 2. SOC driver function: When the car At that time, the urgency level increased exponentially; when At that time, the urgency level decreases approximately linearly: .

[0127] 3. Remaining mileage function: Set a threshold ( ), calculated using the Sigmoid function: .

[0128] 4. User Application Channel: Provide an emergency mission application portal; vehicles that pass the review will be assigned... Dynamically increased to the 0.8–1.0 range, significantly improving priority during the task.

[0129] 5. Overall urgency level: .

[0130] The economic value assessment module in this embodiment: 1. Charging time effect: ; .

[0131] 2. Value of Membership: .

[0132] 3. Dynamic premium: .

[0133] 4. Overall economic value: .

[0134] The dynamic weight fusion model in this embodiment: 1. Normalization process: In the final scheduling priority, three types of scores need to be integrated: credit ( ), urgency ( ), economic value ( To maintain comparability, the three values ​​were first normalized to obtain... .

[0135] 2. Weighting benchmark: In typical scenarios: .

[0136] 3. Peak-hour regulation: If the system load rate If the actual power demand reaches more than 80% of the available power at the site, the urgency weight is increased to ensure the safety of vehicles on the verge of power outages. .

[0137] 4. Off-peak season adjustment: when And queue length The platform prioritizes increasing the weight of economic value: .

[0138] 5. Special event handling: In case of extreme weather, emergency vehicle reporting, or power grid stress, directly set: .

[0139] 6. Manual strategy adjustment: The platform management interface allows manual weight adjustments based on strategic objectives: 1. Attract new users and increase user activity: Increase credit weight. This encourages good users to receive faster responses; 2. Increase revenue and improve efficiency: Increase the weight of economic value ; 3. Service backup: Increase the weight of urgency level. To protect vulnerable groups or critical vehicles.

[0140] Manual adjustments must be submitted through the management interface and gradually transitioned within a 15-minute window to avoid sudden changes that could cause sorting oscillations.

[0141] 7. Weight normalization and boundary conditions: Normalization must be performed after each adjustment: ; and restrictions This prevents a certain dimension from being completely ignored.

[0142] 8. Final output: The final output includes the following: 1. Overall score ; 2. Three-dimensional subdivision ; 3. Weight and its dynamic sources; 4. Contribution breakdown: This helps explain why a certain type of factor either increases or decreases the overall score.

[0143] Through the above embodiments, this application constructs a quantitative system based on three dimensions: credit score, urgency, and economic value. By dynamically weighting these dimensions, it achieves an adaptive ranking mechanism under different operational scenarios. This solution ensures both security and fairness while improving platform revenue and resource utilization efficiency, demonstrating good engineering feasibility and widespread application value.

[0144] The beneficial effects of this application are: Compared with related technologies, this application proposes a comprehensive scheduling scoring method that integrates user credit, task urgency, and economic value. This method overcomes the shortcomings of traditional scheduling systems in terms of single-dimensional decision-making, data sparsity sensitivity, and lack of interpretability, and has the following significant advantages and beneficial effects: 1. Multi-dimensional integration for more comprehensive decision-making: Traditional scheduling methods often rely on a single indicator (such as electricity consumption or reservation order), which can easily lead to imbalances in resource allocation. This application integrates user credit scores, task urgency, and economic value indices in a weighted manner, enabling scheduling decisions to balance security, service fairness, and platform revenue, thereby improving the overall rationality of scheduling.

[0145] 2. Robust modeling to avoid extreme fluctuations: By introducing time decay weights, Beta-binomial smoothing, and confidence contraction, this application effectively alleviates the estimation bias problem in small sample scenarios and avoids excessive extremes in user scores, thereby ensuring the stability and robustness of the model under different data scales.

[0146] 3. Dynamic Adaptation and Flexible Scenarios: This application introduces a dynamic weight adjustment mechanism, which can flexibly adjust the scoring weights according to real-time operating scenarios (such as peak congestion, sudden disasters, and emergency rescue missions), thereby achieving adaptive adjustment of scheduling priorities and enhancing the system's adaptability in complex environments.

[0147] 4. High interpretability and excellent user experience: This application provides three-dimensional sub-scores and a breakdown of the contribution of each factor along with the overall score, helping the platform and users intuitively understand the scoring results and the reasons for changes. This mechanism not only improves transparency but also provides a clear reference path for users to improve their behavior and enhance their credit.

[0148] 5. Balancing social and economic benefits: On the one hand, this application can prioritize emergency vehicles (such as ambulances and public transportation vehicles), improving public services and emergency response capabilities; on the other hand, by introducing economic value factors, the platform can maximize revenue while ensuring fairness, forming a virtuous cycle.

[0149] In summary, this application is superior to related technologies in terms of scientific rigor, robustness, flexibility, and interpretability. It can significantly improve the security, efficiency, and social value of scheduling systems and has broad prospects for widespread application.

[0150] Key points of this application: 1. User credit assessment method based on historical performance behavior; 2. Methods for assessing the urgency of users' charging needs; 3. A charging service reservation and scheduling method based on a three-dimensional dynamic weighted scoring model.

[0151] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0152] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0153] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0154] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0155] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0156] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0157] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0158] The charging service reservation and scheduling method, system, electronic device, storage medium, and program product based on three-dimensional dynamic weighted scoring provided in this application embodiment systematically solves the multi-objective optimization problem under limited charging resources by constructing a comprehensive methodology that integrates credit constraints, risk perception, value assessment, dynamic weighting, and fusion decision-making. Ultimately, it achieves synergistic improvements in operational efficiency, platform revenue, user fairness, and travel safety.

[0159] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0160] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0161] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A charging service reservation and scheduling method based on three-dimensional dynamic weighted scoring, characterized in that, The method includes the following steps: Calculate a three-dimensional indicator score; the three-dimensional indicator score includes user credit score, demand urgency score, and economic value score; The weights are dynamically configured based on the operational status data of the charging stations to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight. The three-dimensional index scores are weighted and fused based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score. The three-dimensional dynamic weighted scores of all users requesting charging are sorted, and charging resource scheduling decisions are made based on the sorting results.

2. The method according to claim 1, characterized in that, The calculation of the three-dimensional index score includes: Acquire user's historical reservation behavior data, current charging request urgency data, and order economic value data; Calculate the user's credit score based on the aforementioned historical booking behavior data; Calculate the urgency score based on the urgency data of the current charging request; Based on the economic value data of the orders, calculate the economic value score; The user credit score, the urgency score, and the economic value score are used as the three-dimensional indicator scores.

3. The method according to claim 2, characterized in that, The calculation of the user credit score based on the historical appointment behavior data includes: Based on the historical reservation behavior data, the type of fulfillment result for each historical reservation behavior is determined according to preset rules; the type of fulfillment result includes cancellation, attendance, and no-show; the historical reservation behavior data includes reservation result, arrival delay, cancellation advance, whether it is a peak period, system fault identifier, site reliability, weather severity, and sample validity period; Based on the timeliness of the samples, a time decay weight is assigned to each historical booking behavior; Using the time decay weight, the performance result types are weighted and counted, and the probability estimate of the performance result type is obtained through probability smoothing and confidence contraction. Based on the probability estimate, interpretability sub-scores are calculated; the interpretability sub-scores include performance sub-scores, on-time sub-scores, cancellation sub-scores, no-show sub-scores, and reservation sub-scores; The interpretability sub-scores are weighted and aggregated according to preset weights to obtain the original credit score. Calculate the uncertainty of the original credit score; Based on the aforementioned uncertainty, the original credit score is scaled back to the platform's benchmark credit score to obtain the final user credit score.

4. The method of claim 2, wherein, The calculation of the urgency score based on the urgency data of the current charging request includes: Based on the urgency data of the current charging request, a basic urgency is calculated using a piecewise function; the urgency data of the current charging request includes the vehicle's current state of charge, battery capacity, average power consumption rate, remaining driving range, travel demand intensity, and task tag; Based on the remaining mileage and the preset safe mileage threshold, the urgency of the remaining mileage is calculated using an S-shaped function. Based on the task tags, it is confirmed whether the vehicle meets the preset priority scheduling conditions, and the task urgency of the vehicle is obtained. The basic urgency, the remaining mileage urgency, and the task urgency are weighted and summed to obtain the demand urgency score.

5. The method of claim 2, wherein, The calculation of the economic value score based on the economic value data of the order includes: The estimated charging time is calculated based on the economic value data of the order, and the charging time factor is calculated based on the estimated charging time using a normalization function; the economic value data of the order includes the charging energy requested by the user, the power of the charging pile, the user's membership level, and the percentage of additional premium paid by the user for the current order; Set a membership value factor based on the user's membership level; Calculate the dynamic premium factor based on the percentage of additional premium paid by the user for the current order; The economic value score is obtained by weighting and summing the charging duration factor, the membership value factor, and the dynamic premium factor.

6. The method of claim 1, wherein, The process of dynamically configuring weights based on the operational status data of charging stations to obtain three-dimensional weight information includes: Set benchmark weights; the benchmark weights include user credit benchmark weights, demand urgency benchmark weights, and economic value benchmark weights; Obtain operational status data of charging stations; the operational status data of charging stations includes available power and actual power demand of the station; The system load rate of the charging station is calculated based on the operating status data of the charging station; the system load rate is the ratio of the actual power demand to the available power of the station. Set weight adjustment conditions; The baseline weights are dynamically adjusted and normalized based on the weight adjustment conditions and the operational status data of the charging stations to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight.

7. The method of claim 1, wherein, The step of weighting and fusing the three-dimensional indicator scores based on the three-dimensional weight information to obtain a three-dimensional dynamically weighted score includes: The three-dimensional indicator scores are normalized to obtain normalized three-dimensional indicator scores; the normalized three-dimensional indicator scores include normalized credit score, normalized urgency score, and normalized economic value score. The normalized three-dimensional index score and the three-dimensional weight information are used to perform weighted fusion to obtain a three-dimensional dynamic weighted score.

8. A charging service reservation and scheduling system based on three-dimensional dynamic weighted scoring, used to implement the method as described in any one of claims 1 to 7, characterized in that, The system includes: The first module is used to calculate a three-dimensional indicator score; the three-dimensional indicator score includes a user credit score, a demand urgency score, and an economic value score. The second module is used to dynamically configure weights based on the charging station's operational status data to obtain three-dimensional weight information; the three-dimensional weight information includes user credit weight, demand urgency weight, and economic value weight. The third module is used to perform weighted fusion of the three-dimensional index scores based on the three-dimensional weight information to obtain a three-dimensional dynamic weighted score. The fourth module is used to sort the three-dimensional dynamic weighted scores of all charging request users and to make charging resource scheduling decisions based on the sorting results.

9. An electronic device, comprising: The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images