A smart waiting list method and system for ticketing based on combination matching and consecutive seat priority

By dynamically calculating the admission threshold and matching integer topology combinations, combined with seat space clustering, the problems of fixed admission thresholds, disordered seat allocation, and limited scenario adaptability in the existing ticketing waitlist scheduling are solved, achieving globally optimal resource allocation and improved system stability in high-concurrency scenarios.

CN123070604APending Publication Date: 2026-11-10JIANGSU UNITED CREDIT REFERENCE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611305471.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-26
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing ticketing waitlist scheduling technology suffers from problems such as fixed admission thresholds, disordered seat allocation, locally optimal matching strategies, and limited scenario adaptability, resulting in low matching success rates, uneven allocation of seat resources, and insufficient system stability in high-concurrency scenarios.

Method used

A scheduling method based on dynamic threshold quantization constraints, integer topology combination matching, and priority of consecutive seat clusters is adopted. By dynamically calculating the admission threshold, integer global domain splitting, and seat space clustering, the system achieves global optimal matching and priority allocation of consecutive seats, and constructs a dual-scenario adaptive candidate scheduling system.

Benefits of technology

It achieves globally optimal resource allocation in high-concurrency scenarios, improves seat resource utilization and system stability, meets the continuous seat needs of multiple people traveling together, avoids ticket over-allocation and seat idleness issues, and ensures the efficiency and reliability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123070604A_ABST
    Figure CN123070604A_ABST
Patent Text Reader

Abstract

This invention discloses a smart ticketing waitlist method and system based on combination matching and consecutive seat priority, belonging to the field of intelligent ticketing scheduling technology. The method calculates the maximum admission threshold by accumulating the number of people in waitlist orders according to priority based on the single-channel vacancy capacity of a single session; it then determines whether to enter offline batch or real-time automatic waitlist mode based on time windows and trigger events. In real-time mode, using the total number of people released as the target, it generates combinations of people through integer global partitioning and filters for globally optimal matching; it clusters consecutive seats in the released seat space, prioritizes their allocation, fills remaining vacancies with discrete seats, and reclaims redundant seats. The system provides this information synchronously. This invention achieves globally optimal matching and consecutive seat priority, improving matching success rate, seat utilization rate, and the rate of multiple people obtaining consecutive seats. It features dual-scenario adaptiveness and high-concurrency security control, making it suitable for high-density ticketing scheduling scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to an adaptive intelligent standby scheduling method and system based on an integer splitting topology combination algorithm and a continuous seat cluster identification mechanism, belonging to the field of intelligent ticketing big data scheduling technology. Background Technology

[0002] With the large-scale development of offline cultural, entertainment, and sports industries, high-density ticketing lottery confirmation, dynamic cancellation and seat release, and waitlist scheduling have become core technological requirements for ticketing platforms. Currently, the mainstream ticketing waitlist scheduling systems in the industry generally adopt the basic scheduling logic of static admission threshold constraints, linear queue traversal matching, and random seat allocation.

[0003] Traditional scheduling solutions of this type have a simple architecture and low development and maintenance costs, and can meet the basic automated waiting list needs in low-concurrency, small-scale passenger flow scenarios, effectively replacing the manual query and manual matching operation mode, and are widely used in traditional ticketing operation scenarios. However, when facing the current entertainment and sports ticketing scenarios with high passenger flow, high concurrency, and concentrated demand for multiple people traveling together, this type of existing technology has exposed significant technical limitations: First, the matching mechanism has shortcomings. It generally adopts a greedy matching strategy that processes each order one by one, which cannot split the seats released by multiple people's ticket refunds into integer combinations. It is difficult to achieve the global optimal matching of waitlist orders, and it is easy for small orders to crowd out seat resources and large orders of multiple people traveling together to be unable to be matched. The overall matching success rate is low in high-dimensional ticket refund scenarios. Second, the seat allocation logic is simplistic, allocating seats only according to the order of the waiting list without considering the spatial continuity of seats. Waiting orders with multiple people traveling together are often assigned scattered seats, which fails to meet users' core need for continuous seats. Third, the capacity for managing the waiting list is insufficient. The waiting list is often limited by fixed quotas or fixed queue lengths. It is impossible to dynamically adjust the admission threshold based on the number of sessions and the real-time number of vacant seats in the channel, which can easily lead to problems such as over-allocation of tickets and long-term idle seats. Fourth, the stability of high-concurrency scenarios is insufficient. There is no independent channel isolation mechanism and concurrent security control strategy. When multiple orders and multiple ticket refunds trigger the waitlist scheduling concurrently, abnormal problems such as order status conflicts and repeated preemption of seat resources are likely to occur. Summary of the Invention

[0004] Purpose of the invention: To address the technical shortcomings of existing ticketing waitlist scheduling technologies, such as fixed admission thresholds, disordered seat allocation, locally optimal matching strategies, and limited scenario adaptability, this invention provides a dual-scenario adaptive intelligent ticketing waitlist method based on combination matching and consecutive seat priority, which is based on dynamic threshold quantization constraints, integer topology combination matching, and priority scheduling of consecutive seat clusters.

[0005] Technical solution: To achieve the above objectives, the technical solution adopted by this invention is as follows: A smart waiting list method for ticketing based on combination matching and consecutive seat priority includes the following steps: Step S1: Based on the capacity of empty seats in a single session and single channel, the number of candidates in the waiting list is accumulated and iterated in order of priority to obtain the maximum admission threshold used to define the priority boundary of valid waiting list orders.

[0006] Step S2: After obtaining the maximum admission threshold, determine whether to enter the offline batch waiting scheduling mode or the real-time automatic waiting matching mode based on the system time window status and the trigger event type.

[0007] Step S3: If it is determined that the offline batch waiting list scheduling mode has been entered, then the valid waiting list orders are filtered with the maximum admission threshold as the boundary, and the tickets are issued in order of priority with delayed issuance.

[0008] Step S4: If it is determined that the real-time automatic waiting list matching mode has been entered, the total number of people released in a single release event is taken as the target. All possible combinations of waiting list order number splits are generated by integer global splitting. The global optimal combination that perfectly matches the current waiting list order pool is selected according to priority weight.

[0009] Step S5: Following the global optimal combination, spatial clustering is performed on all seats released in this seat release event to identify continuous seat clusters, and continuous seats are preferentially allocated to the candidate orders in the global optimal combination, with the remaining demand filled by discrete seats.

[0010] Step S6: Perform a set difference operation on all seats released in this seat release event and the allocated seats, and return the resulting redundant seats to the idle seat resource pool.

[0011] Preferably, the vacancy capacity mentioned in step S1 is calculated using the following formula: ; In the formula: For empty slot capacity, This is a constant for single-game, single-channel allocation. The total number of people who have valid rights and payments for all sessions through this channel.

[0012] The maximum admission threshold is calculated using the following formula: ; In the formula, For the first channel i The number of people that can be accommodated in a standby order. k This represents the number of iterations for the candidate orders in this channel. N To meet the maximum admission threshold of the cumulative number of participants constraint, This is the threshold for the lottery winning order. This represents the capacity for dynamically vacant slots.

[0013] Preferably, step S2 includes: When the payment confirmation window is detected to switch from open to closed, it is determined that the system has entered the offline batch standby scheduling mode.

[0014] When it is detected that the payment confirmation window has been closed and a cancellation or order timeout event has occurred, the system will enter the real-time automatic waiting list matching mode.

[0015] Preferably, step S3, which involves sequentially delaying ticket issuance, includes generating a sorting sequence number for each order according to its priority. After calculating the delay time for each order, tickets are issued sequentially.

[0016] Preferably, in step S4, all splitting combinations generated by the integer global splitting satisfy the following conditions: the sum of the elements equals the total number of people in the banquet, the elements are arranged in a non-increasing order, the number of elements is between the configurable minimum and maximum values, and each element is a positive integer.

[0017] Preferably, in step S4, the priority weight is calculated using the following function: ; The baseline coefficients are corrected based on real-time scene factors: ; After normalization, the final dynamic coefficients are obtained: ; in: This represents the original weight values. Indicates priority weight. The number of elements in the combination. The maximum value of the element in the combination. To determine the dispersion of the internal elements of the combination. , , For scene-adaptive dynamic weighting coefficients, , , These are configurable baseline weighting coefficients, derived based on system performance constraints and business operation data. , , This is a configurable correction sensitivity coefficient, configured according to system stability requirements. This represents the concurrent pressure factor for the current channel. This is a factor representing the proportion of demand from multiple people. As a factor for seat fragmentation, , , To correct the dynamic coefficient.

[0018] Preferably, in step S4, the step of selecting the globally optimal combination specifically includes: traversing each split combination in descending order of priority weight, searching in the current valid candidate order pool for a candidate order set whose element values ​​correspond one-to-one, and determining the first perfectly matching set as the globally optimal combination.

[0019] Preferably, in step S5, the spatial clustering includes: generating clustering grouping keys based on the region code and rank code of the seats, and grouping seats with the same clustering grouping key into the same seat group. The consecutive seat cluster consists of at least two seats located in the same seat group and with consecutive column numbers.

[0020] When the total capacity of the consecutive seat clusters is insufficient to meet the number of people required for a certain waiting order, all consecutive seat clusters are first allocated to that order, and the remaining demand is filled by discrete seats.

[0021] Preferred approach: Apply an exclusive lock to the candidate orders participating in the matching process; during the lock period, the order cannot be accessed by other processes. Apply a pre-lock to the seat to be allocated; the lock is held in place upon successful matching and released immediately upon failure. Perform atomic verification on the order status before matching is executed.

[0022] Another objective of this invention is to provide a ticketing intelligent waiting list system based on combination matching and consecutive seat priority, comprising a dynamic admission threshold calculation module, a scenario diversion module, an offline batch waiting list scheduling module, a combination matching module, a consecutive seat priority allocation module, and a resource closed-loop update module, wherein: The dynamic admission threshold calculation module is used to accumulate and iterate the number of candidates in the waiting list according to the vacancy capacity of a single channel in a single session, so as to obtain the maximum admission threshold used to define the priority boundary of valid waiting list orders.

[0023] The scenario triage module is used to determine whether to enter the offline batch waiting scheduling mode or the real-time automatic waiting matching mode based on the system time window status and the trigger event type after obtaining the maximum admission threshold.

[0024] The offline batch waitlist scheduling module is invoked when the scene diversion module determines that it has entered the offline batch waitlist scheduling mode. The offline batch waitlist scheduling module filters valid waitlist orders with the maximum admission threshold as the boundary and delays ticket issuance in order of priority.

[0025] The combination matching module is invoked when the scene diversion module determines that it has entered the real-time automatic waitlist matching mode. The combination matching module takes the total number of people released in a single release event as the target, generates all possible waitlist order number splitting combinations through integer global splitting, and filters out the global optimal combination that completely matches the current waitlist order pool according to priority weight, and outputs the global optimal combination and its corresponding waitlist order set.

[0026] The continuous seat priority allocation module is used to receive the globally optimal combination and its corresponding candidate order set output by the combination matching module, and to perform spatial clustering on all seats released in this seat release event to identify continuous seat clusters, and to prioritize the allocation of continuous seats to each candidate order in the candidate order set, with the remaining demand being filled by discrete seats.

[0027] The resource closed-loop update module is used to receive the set of allocated seats output by the continuous seat priority allocation module, perform a set difference operation on all seats released in this seat release event and the set of allocated seats, and return the resulting redundant seats to the idle seat resource pool.

[0028] Compared with the prior art, the present invention has the following advantages: 1. Fundamental upgrade to the matching mechanism: Traditional technologies generally employ a greedy strategy of linear, order-by-order matching, which can only achieve local optima. This invention completes combination matching based on an integer global topology splitting algorithm, which can achieve globally optimal adaptation of seat resources.

[0029] 2. Innovation in seat allocation: Traditional technologies lack the ability to identify seat space features and have a simplistic allocation logic. This invention constructs a priority allocation mechanism for continuous seat clusters, fundamentally solving the problem of scattered seats in orders placed by multiple people traveling together.

[0030] 3. Dynamic management of the waiting list: Traditional technologies rely on static rules to limit the scope of candidate admissions; this invention achieves adaptive and precise control through a dynamic admission threshold N, effectively avoiding problems such as ticket over-allocation and idle seats. The weighting coefficients support adaptive dynamic adjustment for different scenarios, adapting to ticketing scenarios with varying loads and passenger flow characteristics. The algorithm's versatility and scenario adaptability are significantly superior to fixed-weight schemes.

[0031] 4. Unified architecture for both scenarios: Traditional technologies only support a single standby operation mode, which limits their ability to adapt to different scenarios. This invention adopts a unified architecture for both scenarios, which is compatible with both batch standby and real-time standby modes, and fully meets the needs of various business operations.

[0032] 5. Innovation in Concurrency Safety Mechanisms: Traditional technologies lack data isolation and concurrency control mechanisms, which can easily lead to anomalies such as order status conflicts and resource duplication in high-concurrency scenarios. This invention completely solves the hidden dangers of concurrent operation through multiple protections such as single-event single-channel data isolation, order exclusive locking, and seat pre-locking, significantly improving the reliability of the scheduling system. It is suitable for dynamic optimization and allocation scenarios of ticketing resources with high density, high concurrency, and high precision requirements, such as performing arts and sports events. Attached Figure Description

[0033] Figure 1 This is a flowchart illustrating the intelligent waiting list method for ticketing based on combination matching and consecutive seat priority.

[0034] Figure 2 This is a schematic diagram of the intelligent waiting list system for ticketing based on combination matching and consecutive seat priority.

[0035] Figure 3 This is a schematic diagram of the overall process of intelligent standby scheduling for a single field, single channel, and dual scenarios according to the present invention.

[0036] Figure 4 The flowchart shows the refined algorithm for integer topology combination matching and priority allocation of consecutive seat clusters in this invention. Detailed Implementation

[0037] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0038] Example 1 To address the shortcomings of existing ticketing waitlist scheduling technologies, such as fixed admission thresholds, disordered seat allocation, locally optimal matching strategies, and limited scenario adaptability, this embodiment provides an intelligent ticketing waitlist method based on combination matching and priority for consecutive seats. It achieves adaptive iterative optimization of the waitlist admission boundary by establishing a dynamic quantification model of seat capacity; constructs a spatially structured identification model for consecutive seats to achieve priority scheduling of high-quality seat resources; designs a global integer combination matching algorithm to overcome the limitations of local greedy matching and achieve globally optimal adaptation; and builds a dual-scenario intelligent traffic splitting architecture to achieve unified algorithm adaptation for differentiated scenarios, such as... Figure 1 As shown, it includes the following steps: Step S1: Based on the capacity of empty seats in a single session and single channel, the number of candidates in the waiting list is accumulated and iterated in order of priority to obtain the maximum admission threshold used to define the priority boundary of valid waiting list orders.

[0039] In some embodiments of the present invention, a channel isolation and concurrency control mechanism is also included. This mechanism achieves physical data isolation based on a single channel per session, ensuring that quotas, seats, and orders from different channels do not interfere with each other. Furthermore, a concurrent exclusive lock mechanism is constructed to prevent state conflicts and resource preemption. Channel isolation: Using session and channel identifier as the sole dimension, quota, seat, and waiting list order data are stored in partitions to enable independent scheduling of multiple channels within a single session.

[0040] Exclusive Lock for Orders: An exclusive lock is applied to the waiting orders for matching. During the lock period, the order cannot be refunded, canceled, or accessed by other matching processes.

[0041] Seat pre-lock: Pre-lock the seats to be allocated, and the lock is fixed after the matching is completed. If the matching fails, the lock is released immediately.

[0042] Status atomic verification: Forcefully verify the order status before matching; only valid candidate statuses can enter the matching process.

[0043] In some embodiments of the present invention, based on the single-channel system's quota confirmation constant and the real-time valid number of confirmed users in that channel, the capacity of vacant seats in that channel is iteratively calculated. Through boundary iteration constrained by the cumulative number of users, the maximum admission lottery order threshold N for that channel is adaptively generated, and the quantification result is persisted to a distributed cache, enabling global sharing and access to that channel. The capacity of vacant seats is calculated using the following formula: ; In the formula: The empty seat capacity is the total number of valid seats available for waiting list replacements based on the session and channel. The empty seat capacity is a variable representing the difference between a system configuration constant and the real-time number of confirmed holders. This is a constant for single-game, single-channel allocation. It represents the maximum allocation of seats for each game and channel preset by the system. This constant for single-game, single-channel allocation is a fixed system configuration constant. The total number of users who have made valid payments for all sessions through this channel is the total number of users who have completed payment and successfully confirmed their rights within a single session and channel. The total number of users who have made valid payments for all sessions through this channel is a dynamic statistical variable.

[0044] Iterative solution of the maximum admission threshold The maximum admission threshold is calculated according to the following formula: ; In the formula, For the first channel i The number of people allowed in a single waiting list order, that is, the number of people in a single venue with a single channel. i The total number of people who purchased tickets corresponding to each standby order. k This represents the number of iterations for the candidate orders in this channel. NTo satisfy the maximum admission threshold for the cumulative number of participants constraint, the maximum admission threshold is... Based on the constraints of session and real-time available seat capacity in each channel, a dynamically generated candidate admission threshold is used to define the upper boundary of the effective candidate user set. The priority code is a globally unique and ordered priority code generated during the random lottery process to determine the order of winning tickets. The code value is negatively correlated with the corresponding session and the waiting list priority of users in the channel; the smaller the value, the higher the priority weight. This represents the capacity for dynamically vacant slots.

[0045] The above-mentioned quantitative model enables dynamic adaptive iteration of the candidate admission boundary, thus eliminating over-allocation and under-allocation problems from a mathematical perspective.

[0046] Step S2: After obtaining the maximum admission threshold, determine whether to enter the offline batch waiting scheduling mode or the real-time automatic waiting matching mode based on the system time window status and the trigger event type.

[0047] In some embodiments of the present invention, a scene-dimensional discrimination model is constructed, and based on time window constraints and event triggering types, intelligent routing of two types of candidate scheduling modes is achieved, unifying the core algorithm and differentiated scheduling logic, specifically including: (1) Offline Batch Waiting Schedule Mode: When the confirmation payment window is detected to switch from open to closed, global batch waiting schedule is triggered, and the system is determined to enter the offline batch waiting schedule mode. The offline batch waiting schedule mode is an offline batch waiting optimal scheduling mechanism based on global seat capacity constraints after the confirmation payment window period ends.

[0048] (2) Real-time automatic waiting list matching mode: When it is detected that the payment window has been closed and a cancellation or order timeout event occurs, the system will enter the real-time automatic waiting list matching mode. The real-time automatic waiting list matching mode is a dynamic game matching mechanism that is triggered by user cancellation or order timeout after the payment window expires.

[0049] Step S3: If it is determined that the offline batch waiting list scheduling mode has been entered, then the valid waiting list orders are filtered with the maximum admission threshold as the boundary, and the tickets are issued in order of priority with delayed issuance.

[0050] In some embodiments of the present invention, sequentially delaying ticket issuance includes: generating sorting numbers for each order according to priority. After calculating the delay time for each order, tickets are issued sequentially.

[0051] In some embodiments of the present invention, screening satisfies The effective waiting list is sorted by priority, and then tickets are issued in batches based on a time-series off-peak scheduling strategy.

[0052] Timing staggered delay formula:

[0053] In the formula: Sort orders by priority number, based on The valid candidate orders are generated in ascending order, with smaller numbers indicating higher priority. This reduces the delay time for single order scheduling, enabling time decoupling and orderly scheduling in high-concurrency scenarios. In batch queue scenarios, this refers to the latency from triggering scheduling to executing matching for a single order, used for peak-shifting and decoupling in high-concurrency scenarios. The system provides a customizable base latency threshold that can be adaptively adjusted based on concurrency pressure and system performance parameters, and is not limited to a single fixed value. It is a configurable time-series scheduling basic delay constant used for peak-shifting and decoupling scheduling in high-concurrency batch candidate scenarios, and supports custom adaptation.

[0054] After the staggered scheduling message is delivered, the consumer side confirms the execution status of compliant orders and solidifies the seats, completing the batch waiting loop.

[0055] Step S4: If it is determined that the real-time automatic waiting list matching mode has been entered, the total number of people released in a single release event is taken as the target. All possible combinations of waiting list order number splits are generated by integer global splitting. The global optimal combination that perfectly matches the current waiting list order pool is selected according to priority weight.

[0056] This invention proposes a topology combination matching algorithm based on positive integer global partitioning. Taking the number of people released in batches as the target dimension, it enumerates all suitable combination topology structures and selects the globally optimal matching solution based on priority weight, thus avoiding the defects of traditional greedy local matching.

[0057] In some embodiments of the present invention, the combined global domain generation method is as follows: The target number of attendees, n, is recursively split into positive integer global domains to generate a set of all unordered split combinations. The present invention provides a dedicated constraint model and recursive iteration formula, strictly defining the global unordered splitting rules: Let the target number of attendees be a positive integer. Number of people in a single sub-dimension Number of combined elements At the same time, set a minimum order quantity within a single combination. Maximum value And satisfy Its global unordered set of positive integer partitions and combinations. The complete constraint formula is as follows:

[0058] Among them, total amount constraint This ensures that the sum of the number of people in each group after splitting is equal to the total number of people released in a single refund, thus meeting the basic business requirement of matching the number of people. This represents the total number of seats available due to a single user's ticket refund or order timeout. (Non-incremental sorting constraint) Automatically removes repeated ordered combinations such as (1,2) and (2,1) to ensure that the resulting topological combinations are unique and free of redundancy. Quantity boundary constraints. , This represents the minimum number of orders within a single combination, i.e., the minimum number of orders allowed in a single matching combination that the system can configure. This represents the maximum number of orders within a single combination, i.e., the maximum number of orders allowed in a single matching combination that the system can configure. It limits the range of order quantities within a single combination based on actual ticketing business rules. This is a logical indicator function. It takes the value 1 when the condition in parentheses is true and 0 when it is false. It is used to determine the legality of splitting elements from a mathematical perspective. The number of people in a single split is the number of people in the sub-dimension, that is, the number of people carried by the order corresponding to the j-th element in the integer split combination. This represents the number of elements in a single group that can be split and combined.

[0059] The recursive generation formula combined with business pruning optimization:

[0060] in, Let be the set of all unordered positive integer partitions and combinations, that is, the set of all unique unordered partitions and combinations generated after recursively partitioning the number of people n across the entire positive integer domain. This is a logical indicator function, i.e., a mathematical logic function. It takes the value 1 when the condition inside the parentheses is true and takes the value 0 when it is false. It is used to determine the validity of an element. This is for recursively generating subsets, that is, the set of sub-split combinations generated during the recursive process of integer splitting. This represents the maximum value of the element within the combination, i.e., the maximum number of people that an order can carry for all elements in a single split combination.

[0061] Recursive logic derivation: 1. Baseline condition: When the number of people attending the event is... When there is only one split combination[1], it serves as the recursive termination condition, ensuring that the algorithm converges normally.

[0062] 2. Iteration condition: For any Select the first element to split For the remaining number of people Continue recursively generating subsets .constraint Continue the non-increasing sorting rule to avoid duplicate combinations.

[0063] 3. Pruning constraints During the recursive process, the upper limit of the number of combined elements is determined in advance, and oversized combinations are actively eliminated, effectively reducing the overall computational cost.

[0064] In some embodiments of the present invention, a multi-level weighted calculation equation system is constructed to prioritize the global splitting and combination, thereby achieving global optimal matching. All splitting combinations generated by the integer global splitting satisfy the following conditions: the sum of the elements equals the total number of people in the banquet; the elements are arranged in a non-increasing order; the number of elements is between the configurable minimum and maximum values; and each element is a positive integer.

[0065] The priority weight is calculated using the following function: ; in: This represents the original weight values. Indicates priority weight. The number of elements in the combination. The maximum value of the element in the combination. To determine the dispersion of the internal elements of the combination. , , The adaptive dynamic weighting coefficients are derived from system performance constraints, business operation data, and real-time scenario factors. They are not fixed parameters preset by humans and satisfy the following conditions: , , The derivation is as follows: (1) Definition of input scene factors: Three types of quantization factors that can be acquired in real time are introduced as inputs for coefficient correction: Concurrent stress factors The ratio of the current channel real-time request QPS to the system's rated scheduling QPS threshold, with a value range of [0,1], represents the system load pressure.

[0066] Multiple demand percentage factor The percentage of valid candidate orders with ≥2 people, ranging from [0,1], represents the intensity of demand for multiple people traveling together.

[0067] Seat fragmentation factor The ratio of the average dispersion of currently available seats to the theoretical maximum dispersion, with a value range of [0,1], represents the degree of fragmentation of seat resources.

[0068] (2) Derivation of benchmark coefficients (general scenario) Based on the actual application scenario of "a certain supermarket" in this application, and combined with the existing ticketing data analysis results, the following baseline weights for general sports event scenarios are derived: Scheduling efficiency is a fundamental prerequisite for high system availability. Inadequate scheduling performance can directly lead to system crashes and order timeouts. Therefore, a baseline coefficient is set. , accounting for the highest proportion.

[0069] User experience is a core business objective and a core competitive advantage of a product, directly impacting user retention and reputation. (Baseline coefficient) .

[0070] Seat fragmentation is a long-term hidden cost, secondary priority, benchmark coefficient. ; Satisfying in general scenarios + =1, which is consistent with the logic of weight normalization.

[0071] (3) Scene dynamic correction and normalization formula The baseline coefficients are corrected based on real-time scene factors, and the final dynamic coefficients are obtained after normalization. ; In the formula This is the efficiency weight sensitivity coefficient, which controls the response magnitude of the scheduling efficiency weight to changes in concurrency pressure. To assess the sensitivity coefficient of the user experience weight, we control the response of the user experience weight to changes in the proportion of multiple users' needs. The resource weight sensitivity coefficient controls the response of the discrete penalty weight to changes in the degree of seat fragmentation. The default value of the three coefficients is 0.5 in general scenarios, and they can be dynamically configured by technical personnel according to the requirements of system performance and business stability.

[0072] The corrected coefficients are normalized to ensure that the sum of the coefficients is always 1.

[0073] After normalization, the final dynamic coefficients are obtained: ; in: , , This is a configurable correction sensitivity coefficient. This represents the concurrent pressure factor for the current channel. This is a factor representing the proportion of demand from multiple people. As a factor for seat fragmentation, , , To correct the dynamic coefficient.

[0074] (4) Adaptation logic 1. The higher the concurrent pressure, The larger the value, the more the algorithm prioritizes scheduling performance. The higher the proportion of users requiring the service, the better. The larger the number of orders, the more likely the algorithm will prioritize matching orders from multiple users. The more fragmented the seating arrangement, the better. The larger the value, the stronger the penalty for dispersion, and the algorithm actively avoids fragmented splitting.

[0075] 2. Item The quadratic power function is used to strengthen the priority effect of the number of orders in the combination. The fewer the number of orders in the combination, the higher this value is.

[0076] 3. Item When the number of elements in a combination is the same, the larger the maximum number of people a combination can accommodate, the higher this value will be, prioritizing orders with multiple people traveling together.

[0077] 4. Item : is the dispersion penalty term, where This is used to quantify the dispersion of elements within a combination. The more fragmented the combination structure, the stronger the negative correction effect of this item, which can effectively suppress fragmented small orders from crowding out high-quality seat resources.

[0078] 5. To normalize the weight values, the weight value range is uniformly converged to the [0,1] interval through overall scaling, since the dynamic coefficients always satisfy... After normalization, the weight range is consistent with the original weight logic, which facilitates the system to perform unified threshold determination and sorting.

[0079] Sorting rules: The fewer the number of elements in a combination, the higher the normalized weight value. When the number of elements is the same, the larger the value of the largest child element within the combination, the higher the weight value, and the system prioritizes high-weight combinations for matching.

[0080] In some embodiments of the present invention, the step of selecting the globally optimal combination specifically includes: traversing each split combination in descending order of priority weight, searching in the current valid candidate order pool for a candidate order set whose element values ​​correspond one-to-one, and determining the first perfectly matching set found as the globally optimal combination.

[0081] In some embodiments of the present invention, the optimal solution matching method includes: Concurrency lock verification: Before traversing the combination, acquire an exclusive lock on the target candidate order and atomically verify the order status. Once locked, the order cannot be preempted by other ticket refund processes.

[0082] A multidimensional fit function is introduced to quantify the degree of matching between the combination and the current candidate order pool:

[0083] in, This is a multidimensional fit function, which is used to quantify the degree of matching between a single split / combination and the current candidate order pool. This is the set of currently valid waiting orders, that is, the set of all orders in a valid waiting state within a single session and single channel.

[0084] By normalized weights Traverse all combined topologies in descending order and calculate the comprehensive matching value. The combination with the highest overall matching value is selected as the global optimal solution. Simultaneously, the dimensional adaptability of the current candidate order resource pool is verified, and the first fully compatible combination set is returned. If no valid compatible combination is found, the lock resources are released and the seat resources are returned to the pool for release.

[0085] Step S5: Following the global optimal combination, spatial clustering is performed on all seats released in this seat release event to identify continuous seat clusters, and continuous seats are preferentially allocated to the candidate orders in the global optimal combination, with the remaining demand filled by discrete seats.

[0086] This invention constructs a spatial structured identification model for seats, clusters discrete seat resources, identifies high-priority continuous seat clusters, and realizes intelligent priority allocation of spatial resources.

[0087] In some embodiments of the present invention, the spatial clustering includes: generating clustering grouping keys based on the region code and rank code of the seats, and grouping seats with the same clustering grouping key into the same seat group. The consecutive seat cluster consists of at least two seats located in the same seat group and with consecutive column numbers.

[0088] When the total capacity of the consecutive seat clusters is insufficient to meet the number of people required for a certain waiting order, all consecutive seat clusters are first allocated to that order, and the remaining demand is filled by discrete seats.

[0089] Seat clustering grouping formula:

[0090] in, The seat clustering group key is a unique identifier used for seat spatial clustering, which is generated by concatenating the region code and the number code. Area code, which is a unique identifier for different viewing areas within the venue. The row number code is a unique identifier for different rows within the same area of ​​the venue.

[0091] General clustering is performed based on regional coding and number coding to divide seats with the same source into the same resource group.

[0092] Constraints for determining consecutive seat clusters:

[0093] in, is the column number of the i-th seat, that is, the column number identifier of different seats in the same row in the venue.

[0094] They are determined to be in the same continuous seat cluster if and only if the difference between the column numbers of adjacent seats is 1 and the spatial clustering codes are the same.

[0095] Two-level allocation strategy model: 1. High-quality matching: If the capacity of consecutive seat clusters is greater than or equal to the number of people required by the order, consecutive seat cluster resources will be allocated first.

[0096] 2. Adaptive Filling of Remaining Space: If the total capacity of consecutive seat clusters is insufficient to cover order demand, all consecutive seat cluster resources will be used, and the remaining dimensional gaps will be adaptively filled by discrete seat resources.

[0097] In some embodiments of the present invention, an exclusive lock is applied to the candidate orders participating in the matching process, and the order cannot be invoked by other processes during the lock period. A pre-lock is applied to the seat to be allocated, which is fixed upon successful matching and released immediately upon failure. An atomic check is performed on the order status before matching is executed.

[0098] Step S6: Perform a set difference operation on all seats released in this seat release event and the allocated seats, and return the resulting redundant seats to the idle seat resource pool.

[0099] In some embodiments of the present invention, after the seat resource matching and allocation are completed, the order status is confirmed and solidified, and redundant seat resources are quantified through set difference operation:

[0100] In the formula: This refers to the total set of seats available for this refund, which is the set of all seats released due to a single refund or order expiration. It is a set difference operator, which removes the assigned seats from the total set of seats. This is the set of allocated waitlist seats, which is the set of all seats successfully allocated to users in this waitlist scheduling. Unoccupied redundant seats are obtained through differential operations, and resource updates are uniformly performed back to the pool to achieve closed-loop scheduling of seat resources.

[0101] This invention constructs a dynamic quantitative model for seat capacity, which replaces the traditional static and fixed configuration by calculating vacant seats and admission thresholds in real time. It completely eliminates the problems of ticket over-allocation and seat waste from the perspective of mathematical constraints, significantly improves the effective utilization rate of seat resources, realizes dynamic adaptive iteration of the candidate admission threshold, and maximizes the accuracy of resource scheduling.

[0102] This invention utilizes a continuous seat clustering identification model. Through spatial-dimensional clustering and priority scheduling, it accurately identifies high-quality continuous seat resources, prioritizing the needs of multiple collaborative candidates. This addresses the technical pain point of traditional algorithms' discrete seat allocation and significantly optimizes the candidate user experience. It also employs global topology combination matching to achieve globally optimal resource adaptation.

[0103] This invention unifies the core quantization algorithm, is compatible with both offline batch time-series scheduling and real-time dynamic adaptive matching scenarios, and constructs a fully covered ticketing waitlist scheduling system. Its scenario adaptability far exceeds that of traditional single scheduling modes, significantly improving system fault tolerance and versatility. Its dual-scenario traffic-sharing architecture offers extremely strong algorithm versatility and robustness.

[0104] This invention introduces a new single-channel isolation and concurrent exclusive lock mechanism, fundamentally preventing issues such as order refunds and multiple ticket cancellations concurrently occupying the same order during waitlist matching. This ensures the atomicity and consistency of the scheduling process, significantly improving system stability and data accuracy. Concurrency security management eliminates state conflicts and resource preemption.

[0105] The core logic of this invention is defined through standardized mathematical formulas, constraint models, and weighting algorithms, with no ambiguous business logic, possessing extremely high technical barriers and meeting the inventiveness, novelty, and practicality requirements of patent application review. The entire process involves mathematical modeling, making it quantifiable, reproducible, and iterative.

[0106] Based on the global recursive decomposition of positive integers to generate all combination sets, and combined with scenario-adaptive dynamic weight sorting rules and traversal verification logic, the globally optimal solution is selected. Seat clustering is completed through spatial encoding, and continuous seat clusters are identified based on column number constraints. A differentiated allocation strategy is executed to ensure that orders from multiple people traveling together receive consecutive seats. Combining the vacancy quantification formula and the cumulative constraint of the number of people on the waiting list, the admission threshold is dynamically iteratively solved to achieve precise control over the scale of the waiting list. A unified core algorithm logic is reused, adapting to both offline batch time-series scheduling and real-time dynamic adaptive matching waiting list business scenarios. Data physical isolation is achieved by session and channel dimensions, combined with order exclusive locks, seat pre-locking, and state atomic verification to avoid conflicts and anomalies under concurrent scheduling. Redundant seats are quantified using set difference operators to complete seat resource return to the pool and state updates, achieving closed-loop scheduling throughout the entire process.

[0107] Example 2 To address the shortcomings of existing ticketing waitlist scheduling technologies, such as fixed admission thresholds, disordered seat allocation, locally optimal matching strategies, and limited scenario adaptability, this embodiment provides a smart ticketing waitlist system based on combined matching and consecutive seat priority. The system employs a smart ticketing waitlist method based on combined matching and consecutive seat priority, such as... Figure 2As shown, it includes a dynamic admission threshold calculation module, a scene triage module, an offline batch waitlist scheduling module, a combination matching module, a continuous seat priority allocation module, and a resource closed-loop update module, wherein: The dynamic admission threshold calculation module is used to accumulate and iterate the number of candidates in the waiting list according to the vacancy capacity of a single channel in a single session, so as to obtain the maximum admission threshold used to define the priority boundary of valid waiting list orders.

[0108] The scenario triage module is used to determine whether to enter the offline batch waiting scheduling mode or the real-time automatic waiting matching mode based on the system time window status and the trigger event type after obtaining the maximum admission threshold.

[0109] The offline batch waitlist scheduling module is invoked when the scene diversion module determines that it has entered the offline batch waitlist scheduling mode. The offline batch waitlist scheduling module filters valid waitlist orders with the maximum admission threshold as the boundary and delays ticket issuance in order of priority.

[0110] The combination matching module is invoked when the scene diversion module determines that it has entered the real-time automatic waitlist matching mode. The combination matching module takes the total number of people released in a single release event as the target, generates all possible waitlist order number splitting combinations through integer global splitting, and filters out the global optimal combination that completely matches the current waitlist order pool according to priority weight, and outputs the global optimal combination and its corresponding waitlist order set.

[0111] The continuous seat priority allocation module is used to receive the globally optimal combination and its corresponding candidate order set output by the combination matching module, and to perform spatial clustering on all seats released in this seat release event to identify continuous seat clusters, and to prioritize the allocation of continuous seats to each candidate order in the candidate order set, with the remaining demand being filled by discrete seats.

[0112] The resource closed-loop update module is used to receive the set of allocated seats output by the continuous seat priority allocation module, perform a set difference operation on all seats released in this seat release event and the set of allocated seats, and return the resulting redundant seats to the idle seat resource pool.

[0113] Example 3 This embodiment provides a batch candidate dynamic threshold scheduling method, used to characterize the overall execution chain of the entire candidate scheduling system of the present invention, such as... Figure 3As shown, the system first calculates the maximum admission lottery order threshold N based on a constant single-channel allocation quota and the real-time valid number of confirmed users for that channel using a dynamic capacity constraint model, and then caches the quantification results. Subsequently, the system performs intelligent traffic diversion based on the time window status and trigger event type. If the payment window is detected as closed, an offline batch waiting list scheduling mode is triggered. Valid orders are filtered based on the admission threshold N, and after order exclusivity locking and status verification, a configurable time-series staggered scheduling mechanism completes the batch ticketing loop. If the payment window is detected as closed and a ticket refund / seat release event occurs, a real-time automatic waiting list matching mode is triggered. After concurrency security verification, the system enters the entire process of integer topology combination matching, priority allocation of consecutive seat clusters, and differential release of redundant seat resources, ultimately completing the entire waiting list scheduling loop for that channel. Specifically, this includes the following steps: Business parameters: For a certain session, the single-channel allocation constant LOTTERY_DRAW_NUM = 1000 people, the total number of valid payment confirmations ΣACCEPTED_PAY_NUM = 920 people, the vacant seat capacity VACANT_SEAT_COUNT = 80 seats, and there are 20 waiting list orders for this channel. Execution process: 1. Based on the vacancy slot capacity formula, we get: VACANT_SEAT_COUNT = 1000 920 = 80.

[0114] 2. Traverse the waiting orders in ascending order of SIGN_ORDER, iteratively accumulate the number of people, and solve for the maximum admission threshold N.

[0115] 3. Filter valid candidate orders with SIGN_ORDER≤N, execute exclusive locks on the orders and atomic status checks to avoid concurrent status conflicts.

[0116] 4. Based on the time-series peak-shifting delay formula DELAY_TIME=INDEX×BASE_DELAY, the scheduling is performed, the basic delay threshold BASE_DELAY=200ms is set, and the sorting sequence number INDEX starts from 1 and increments, so as to realize the delay of ticket issuance for each order and the decoupling of time sequence in high-concurrency scenarios.

[0117] Example 4 This embodiment provides an automatic candidate two-dimensional combination matching method, used to characterize the refinement of the core innovative algorithm of this invention, such as... Figure 4As shown, using the number of passengers (n) who cancel tickets and release seats in a single session, single channel, and single time as input parameters, a set of all unordered topological combinations Ω(n) is generated through a global recursive decomposition formula for positive integers. All combinations are sorted in descending order of priority using a two-layer weighted sorting model. After order exclusive locking and state atomic verification, the suitability of the candidate order resource pool is verified group by group according to the sorting results, and the globally optimal matching order combination is selected. After successful matching, the same-origin seat groups are divided using a seat space clustering formula, and high-priority continuous seat resources are identified by combining continuous seat cluster judgment constraints. A two-layer allocation strategy of prioritizing continuous seats and filling in discrete seats is executed to complete seat allocation. After allocation, the process returns to Figure 1, and seat resource closed-loop update and redundancy release are performed, including the following steps: Business parameters: Number of passengers released per refund (n=2); waitlist orders include several 2-person and 1-person configurations. Execution process: 1. Generate Ω(2)={[2],[1,1]} based on the integer global recursive decomposition formula.

[0118] 2. Calculate the combined weights according to the multi-level weight equation system, and complete the sorting by combining the normalization results. The weight of combination [2] is higher than that of combination [1,1], and the 2-person specification candidate orders are matched first.

[0119] 3. Add an exclusive lock to the target order and complete the atomic status verification.

[0120] 4. Cluster identification of consecutive seat clusters, and assign consecutive seats to successfully matched orders.

[0121] 5. After allocation, return to the overall map and execute S6 set difference operation to update redundant seat resources.

[0122] Example 5 This embodiment provides a high-dimensional complex combination matching method, including the following steps: Parameters for this scenario: Concurrency pressure factor Multiple demand percentage factor Seat fragmentation factor The final weighting coefficients are calculated using the dynamic adjustment formula. , , .

[0123] Business parameters: Number of passengers released per refund (n=4); waitlist orders include multiple options: 4 people, 3 people, 2 people, and 1 person. Execution process: 1. Generate the unordered split set Ω(4)={[4],[3,1],[2,2],[2,1,1],[1,1,1,1]}.

[0124] 2. Traverse all combined topologies in descending order of weight priority.

[0125] 3. Perform concurrent exclusive lock verification and status verification on candidate orders.

[0126] 4. Select the optimal combination globally and complete the fixed allocation of seats.

[0127] Example 6 This embodiment provides a method for differentiated allocation of consecutive seat clusters, including the following steps: Business Parameters: This seating release includes 3 consecutive seats in row 1 of section A and 2 scattered seats in row 2 of section B. Waiting lists are for 3 people, 2 people, and 1 person. Execution Process: 1. Perform seat clustering and grouping based on GROUP_KEY=AREA_CODE+'_'+ROW.

[0128] 2. Identify row 1 of area A as a high-priority continuous seat cluster based on the continuous seat cluster determination constraint and perform seat pre-locking.

[0129] 3. For orders with 3 people, priority is given to allocating complete and continuous seat clusters, and the remaining discrete seats are adaptively filled for orders with 2 people.

[0130] 4. After allocation is completed, release the lock resources and update the seat pool through set difference operation.

[0131] Simulation test To objectively verify the technical advantages of this invention compared to traditional ticketing waitlist scheduling technology, standardized simulation control tests (controlled experimental environment, comparison of controlled variables) and online actual operation data statistics (real business scenarios) were conducted. Throughout the testing, the mainstream traditional ticketing waitlist solution in the industry was used as the benchmark: the traditional solution adopts a static admission threshold, linear greedy matching, random seat allocation, and an architecture without independent channels and concurrent locking.

[0132] I. Basic Test Instructions 1.1 Hardware and software testing environment (1) Simulation test environment Server configuration: 8-core, 16GB cloud server, CentOS 7.9 operating system. Simulation tool: A self-developed ticketing traffic simulation platform that allows customization of total seat capacity, refund frequency, order type, and concurrency level. Database: MySQL 8.0 + Redis distributed cache (2) Online production environment Deployment Scenario: "Supermarket" sports event ticketing system Server Cluster: Distributed microservice architecture, multi-node load balancing 1.2 Definition of Core Evaluation Indicators (1) Order matching success rate: The proportion of waiting orders with successful matching of the number of people to the total number of waiting orders.

[0133] (2) Consecutive seat acquisition rate: The proportion of orders placed by multiple people in the same group that obtain complete consecutive seats.

[0134] (3) Seat utilization rate: The proportion of the number of seats that have been successfully allocated to the total number of seats released this time.

[0135] (4) Concurrency conflict rate: The percentage of times an order status is abnormal or a seat is repeatedly occupied under high-concurrency scheduling.

[0136] II. Standardized Simulation Comparison Test All tests used the industry-standard traditional ticketing waiting list solution (static threshold + linear greedy matching + random seating) as a control group, with each scenario running independently for 50 rounds and the average value taken.

[0137] Scenario 1: Batch standby scheduling scenario Business parameters: 1000 seats per session, 920 seats confirmed, 80 seats vacant, and 20 waiting list orders.

[0138]

[0139] Scenario 2: High-dimensional real-time scene completion Business parameters: Single ticket refund releases seats for 4 people; waitlist orders include various configurations for 4, 3, 2, and 1 people.

[0140] Scenario 3: High-concurrency extreme scenario Business parameters: Simultaneously trigger 30 sets of ticket refunds and 80 waitlist orders in a single channel to simulate peak traffic before the start of the performance.

[0141]

[0142] III. Actual Online Operation Data (Reflecting Only the System of This Invention) This invention has been officially launched and is running in the ticketing system of a certain sports event. The following is a statistical summary of 30 consecutive days of operation data (daily average data): Average daily standby orders on match days: 4150 Average daily seat release due to ticket refunds: 1587 seats Peak daily concurrency on match days: 1100 QPS (QPS is the query rate per second, which is the number of requests the system can process per second and is used to measure the system's concurrent processing capacity).

[0143]

[0144] IV. Overall Conclusion Simulation tests show that, through core technologies such as integer global combination matching, priority allocation of consecutive seats, and dynamic threshold control, the present invention achieves orders of magnitude improvements in matching efficiency, resource utilization, and concurrency stability compared to traditional solutions.

[0145] Online operational data verification: This invention fully meets the high-density, high-concurrency ticketing scheduling needs of a certain super event, with a seat utilization rate of nearly 100%. It completely solves the core pain points of traditional systems, such as batch idle seats and scattered seats for multiple orders, and the engineering implementation effect is excellent.

[0146] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A smart waiting list method for ticketing based on combination matching and consecutive seat priority, characterized in that, Includes the following steps: Step S1: Based on the capacity of empty seats in a single session and single channel, the number of candidates in the waiting list is accumulated and iterated in order of priority to obtain the maximum admission threshold used to define the priority boundary of valid waiting list orders; Step S2: After obtaining the maximum admission threshold, determine whether to enter the offline batch waiting scheduling mode or the real-time automatic waiting matching mode based on the system time window status and the trigger event type. Step S3: If it is determined that the offline batch waiting list scheduling mode has been entered, then the valid waiting list orders are filtered with the maximum admission threshold as the boundary, and the tickets are issued in order of priority with delayed time. Step S4: If it is determined that the real-time automatic waiting list matching mode has been entered, the total number of people released in a single release event is taken as the target. All possible combinations of waiting list order number splits are generated by integer global splitting. The global optimal combination that perfectly matches the current waiting list order pool is selected according to priority weight. Step S5: Following the global optimal combination, perform spatial clustering on all seats released in this seat release event to identify consecutive seat clusters, and prioritize the allocation of consecutive seats to the candidate orders in the global optimal combination, with the remaining demand filled by discrete seats; Step S6: Perform a set difference operation on all seats released in this seat release event and the allocated seats, and return the resulting redundant seats to the idle seat resource pool.

2. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 1, characterized in that: The vacancy capacity mentioned in step S1 is calculated using the following formula: ; In the formula: For empty slot capacity, This is a constant for single-game, single-channel allocation. The total number of people who have valid rights and payments for all sessions in this channel; The maximum admission threshold is calculated using the following formula: ; In the formula, For the first channel i The number of people allowed in a standby order; k This represents the number of iterations for the candidate orders in this channel. N To meet the maximum admission threshold of the cumulative number of participants constraint, This is the threshold for the lottery winning order. This represents the capacity for dynamically vacant slots.

3. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 2, characterized in that: Step S2 includes: When the payment confirmation window is detected to switch from open to closed, it is determined that the system will enter the offline batch standby scheduling mode. When it is detected that the payment confirmation window has been closed and a cancellation or order timeout event has occurred, the system will enter the real-time automatic waiting list matching mode.

4. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 3, characterized in that: Step S3, which involves sequentially delaying ticket issuance, includes: generating sorting numbers for each order according to priority. After calculating the delay time for each order, tickets are issued sequentially.

5. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 4, characterized in that: In step S4, all splitting combinations generated by the integer global splitting satisfy the following conditions: the sum of the elements equals the total number of people in the banquet, the elements are arranged in a non-increasing order, the number of elements is between the configurable minimum and maximum values, and each element is a positive integer.

6. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 5, characterized in that: In step S4, the priority weight is calculated using the following function: ; The baseline coefficients are corrected based on real-time scene factors: ; After normalization, the final dynamic coefficients are obtained: ; in: This represents the original weight values. Indicates priority weight. The number of elements in the combination. The maximum value of the element in the combination. To determine the dispersion of the internal elements of the combination. , , For scene-adaptive dynamic weighting coefficients, , , These are configurable baseline weighting coefficients, derived based on system performance constraints and business operation data. , , This is a configurable correction sensitivity coefficient, configured according to system stability requirements. This represents the concurrent pressure factor for the current channel. This is a factor representing the proportion of demand from multiple people. As a factor for seat fragmentation, , , To correct the dynamic coefficient.

7. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 6, characterized in that: In step S4, the step of selecting the global optimal combination specifically includes: traversing each split combination in descending order of priority weight, searching in the current valid candidate order pool for a candidate order set whose element values ​​correspond one-to-one, and determining the first perfectly matching set as the global optimal combination.

8. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 7, characterized in that: In step S5, the spatial clustering includes: generating clustering grouping keys based on the regional code and serial number code of the seats, and dividing seats with the same clustering grouping key into the same seat group; the continuous seat cluster consists of at least two seats located in the same seat group and with consecutive column numbers; When the total capacity of the consecutive seat clusters is insufficient to meet the number of people required for a certain waiting order, all consecutive seat clusters are first allocated to that order, and the remaining demand is filled by discrete seats.

9. The intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 8, characterized in that: An exclusive lock is applied to the candidate orders participating in the matching process, and the order cannot be invoked by other processes during the lock period; A pre-lock is applied to the seat to be allocated, which is fixed upon successful matching and released immediately upon failure; the order status is atomically verified before matching is executed.

10. A system employing the intelligent waiting list method for ticketing based on combination matching and consecutive seat priority as described in claim 1, characterized in that, It includes a dynamic admission threshold calculation module, a scene triage module, an offline batch waitlist scheduling module, a combination matching module, a continuous seat priority allocation module, and a resource closed-loop update module, among which: The dynamic admission threshold calculation module is used to accumulate and iterate the number of candidates in the waiting list according to the priority order based on the empty seat capacity of a single channel in a single session, so as to obtain the maximum admission threshold used to define the priority boundary of valid waiting list orders. The scenario triage module is used to determine whether to enter the offline batch waiting scheduling mode or the real-time automatic waiting matching mode based on the system time window status and the trigger event type after obtaining the maximum admission threshold. The offline batch waitlist scheduling module is invoked when the scene diversion module determines that it has entered the offline batch waitlist scheduling mode. The offline batch waitlist scheduling module filters valid waitlist orders with the maximum admission threshold as the boundary and delays ticket issuance in order of priority. The combination matching module is invoked when the scene diversion module determines that it has entered the real-time automatic waiting match mode. The combination matching module takes the total number of people released in a single release event as the target, generates all possible combinations of waiting order number splits through integer global split, and filters out the global optimal combination that perfectly matches the current waiting order pool according to priority weight, and outputs the global optimal combination and its corresponding waiting order set. The continuous seat priority allocation module is used to receive the global optimal combination and its corresponding candidate order set output by the combination matching module, and to perform spatial clustering on all seats released in this seat release event to identify continuous seat clusters, and to prioritize the allocation of continuous seats to each candidate order in the candidate order set, with the remaining demand being filled by discrete seats; The resource closed-loop update module is used to receive the set of allocated seats output by the continuous seat priority allocation module, perform a set difference operation on all seats released in this seat release event and the set of allocated seats, and return the resulting redundant seats to the idle seat resource pool.