A method, apparatus and equipment for warehouse collaborative scheduling data processing

Abstract

This invention provides a method, apparatus, and equipment for warehouse collaborative scheduling data processing, belonging to the field of warehousing technology, and solves the problems of low warehouse scheduling efficiency and cargo backlog losses. The method includes: acquiring warehouse operation data; preprocessing the warehouse operation data to obtain a basic parameter set, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehouse resource parameters; determining cargo scheduling coefficients based on the basic parameter set, which include timeliness urgency coefficients, cargo damage risk coefficients, order priority coefficients, and operation resource adaptation coefficients; determining dynamic scheduling priority scoring data based on the cargo scheduling coefficients; determining a hierarchical scheduling strategy based on the dynamic scheduling priority scoring data, which includes at least two levels of scheduling strategies; and allocating and scheduling operations for goods according to the hierarchical scheduling strategy. This solution achieves hierarchical scheduling of warehousing, improves scheduling efficiency, and reduces cargo losses.

Description

Technical Field

[0001] This invention relates to the field of warehousing technology, and in particular to a method, apparatus and equipment for collaborative scheduling data processing in warehousing. Background Technology

[0002] With the rapid development of modern logistics and supply chain industries, warehousing, as a core node of the logistics network, directly impacts the overall operating costs and service quality of the logistics system through its scheduling efficiency and control precision. Currently, significant progress has been made in warehousing scheduling technology. The industry has widely achieved automated collection of warehousing data across all dimensions, utilizing technologies such as IoT sensors and RFID (Radio Frequency Identification) tags to obtain real-time information on goods location, environmental temperature, and humidity. Warehouse management systems and control systems have been widely adopted, supporting digital control of inbound and outbound processes and enabling basic route planning and task allocation. Despite these advancements in warehousing digitization and automation, several insurmountable shortcomings remain.

[0003] First, the rigidity of scheduling rules is a significant problem. Most existing scheduling methods rely on static, time-based rules such as First-In-First-Out (FIFO), while a few employ linearly weighted multi-factor priority ranking schemes. Essentially, these methods still involve linearly combining multiple factors with fixed weights. These methods are clearly inadequate when handling goods with high time sensitivity and high risk of damage. When a large influx of regular orders occurs, the priority of high-risk goods is easily diluted and squeezed out, leading to overstocking and losses. This is a long-standing and unresolved core problem in the industry.

[0004] Secondly, existing technologies have failed to establish a refined dynamic hierarchical scheduling mechanism. Most hierarchical rules are based on the static attributes of orders and are not deeply linked to the real-time status of goods and the risk of damage. When goods experience changes in risk such as environmental anomalies or approaching expiration dates during storage, existing scheduling systems cannot dynamically adjust their priorities, resulting in a disconnect between scheduling decisions and the actual status of the goods. Summary of the Invention

[0005] This invention provides a warehouse collaborative scheduling data processing method, apparatus and equipment, which solves the problems of low warehouse scheduling efficiency and easy loss of goods due to backlog.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: This invention provides a warehouse collaborative scheduling data processing method, including: Obtain warehouse operation data; The warehousing operation data is preprocessed to obtain a set of basic parameters, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters. Based on the aforementioned set of basic parameters, a cargo scheduling coefficient is determined, which includes a timeliness urgency coefficient, a cargo damage risk coefficient, an order priority coefficient, and an operational resource adaptation coefficient. Based on the cargo scheduling coefficient, determine the dynamic scheduling priority score data; Based on the dynamic scheduling priority scoring data, a hierarchical scheduling strategy is determined, which includes at least two levels of scheduling strategy. According to the hierarchical scheduling strategy, the goods are allocated and scheduled.

[0007] Optionally, the warehousing operation data is preprocessed to obtain a set of basic parameters, including: The warehousing operation data is cleaned to obtain cleaned warehousing operation data; The cleaned warehousing operation data is standardized to obtain standardized warehousing operation data; The standardized warehousing operation data is then subjected to time and space alignment processing to obtain aligned warehousing operation data; The aligned warehousing operation data is processed by feature extraction to obtain a basic parameter set.

[0008] Optionally, based on the aforementioned set of basic parameters, the cargo scheduling coefficient is determined, including: Based on the cargo timeliness parameters in the aforementioned basic parameter set, a timeliness urgency coefficient is determined, wherein the cargo timeliness parameters include at least the cargo-related order identifier, remaining fulfillment time, total order fulfillment time, remaining shelf life, and total shelf life of the goods; Based on the cargo damage risk parameters in the aforementioned basic parameter set, a cargo damage risk coefficient is determined, wherein the cargo damage risk parameters include at least the environmental deviation risk value, the storage duration risk value, and the inherent cargo damage risk value. Based on the order fulfillment parameters in the basic parameter set, an order priority coefficient is determined, wherein the order fulfillment parameters include at least an order type identifier; Based on the warehousing resource parameters in the basic parameter set, the operational resource adaptation coefficient is determined, wherein the warehousing resource parameters include at least the storage location adaptation value and the equipment resource adaptation value.

[0009] Optionally, based on the cargo scheduling coefficient, dynamic scheduling priority scoring data is determined, including: The cargo scheduling coefficients are input into a preset priority scoring model for processing to obtain dynamic scheduling priority scoring data. The priority scoring model is as follows: ; in, This is for dynamic scheduling priority scoring data; This is the timeliness and urgency coefficient; This refers to the risk coefficient of cargo damage. This is the order fulfillment priority coefficient; The resource adaptation coefficient for the task; , , The preset weighting coefficients, and ; It is a non-linear risk amplification factor; To obtain the aforementioned timeliness coefficient and cargo damage risk coefficient The maximum value in.

[0010] Optionally, the nonlinear risk amplification factor can be determined as follows: The nonlinear risk amplification factor is compared with a preset critical threshold. When the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is greater than or equal to the critical threshold, the nonlinear risk amplification factor is set to a first value; when the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is less than the critical threshold, the nonlinear risk amplification factor is set to a second value. The first value is greater than the second value.

[0011] Optionally, based on the dynamic scheduling priority scoring data, a hierarchical scheduling strategy is determined, including: The dynamic scheduling priority scoring data is divided into levels according to at least two preset grading thresholds to obtain scheduling level data; Based on the scheduling level data, scheduling execution rules are determined, including resource allocation priority, job path selection strategy, and task interruption strategy.

[0012] Optionally, according to the hierarchical scheduling strategy, the cargo allocation and scheduling operation includes: Based on the scheduling execution rules corresponding to different levels in the hierarchical scheduling strategy, scheduling task instructions for each cargo are generated. The scheduling task instructions are sent to the warehouse control system to control the warehouse equipment to perform corresponding operations; The status data during the job execution process is collected in real time, and the corresponding parameters in the basic parameter set are updated accordingly.

[0013] This invention also provides a warehouse collaborative scheduling data processing device, comprising: The acquisition module is used to acquire warehouse operation data; The processing module is used to preprocess the warehousing operation data to obtain a basic parameter set, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters; based on the basic parameter set, determine cargo scheduling coefficients, which include timeliness urgency coefficients, cargo damage risk coefficients, order priority coefficients, and operation resource adaptation coefficients; based on the cargo scheduling coefficients, determine dynamic scheduling priority scoring data; based on the dynamic scheduling priority scoring data, determine a hierarchical scheduling strategy, which includes at least two levels of scheduling strategy; and allocate and schedule cargo according to the hierarchical scheduling strategy.

[0014] This invention also provides a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when run by the processor, executes the above-described method.

[0015] This invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described method.

[0016] The technical solution of the present invention has at least the following effects: The above-described solution of the present invention acquires warehousing operation data; preprocesses the warehousing operation data to obtain a basic parameter set, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters; determines cargo scheduling coefficients based on the basic parameter set, which include timeliness urgency coefficients, cargo damage risk coefficients, order priority coefficients, and operation resource adaptation coefficients; determines dynamic scheduling priority scoring data based on the cargo scheduling coefficients; determines a hierarchical scheduling strategy based on the dynamic scheduling priority scoring data, which includes at least two levels of scheduling strategy; and allocates and schedules cargo according to the hierarchical scheduling strategy, thereby achieving hierarchical scheduling of warehousing, improving scheduling efficiency, and reducing cargo loss. Attached Figure Description

[0017] Figure 1 This is a flowchart of the warehouse collaborative scheduling data processing method provided in the embodiments of the present invention; Figure 2 This is a structural diagram of the warehouse collaborative scheduling data processing device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the computing device provided in an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0019] like Figure 1 As shown, an embodiment of the present invention proposes a warehouse collaborative scheduling data processing method, including: Step 11, Obtain warehouse operation data; Step 12: Preprocess the warehousing operation data to obtain a basic parameter set, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters; Step 13: Determine the cargo scheduling coefficient based on the basic parameter set. The cargo scheduling coefficient includes the timeliness urgency coefficient, cargo damage risk coefficient, order priority coefficient, and operational resource adaptation coefficient. Step 14: Determine the dynamic scheduling priority score data based on the cargo scheduling coefficient; Step 15: Determine a hierarchical scheduling strategy based on the dynamic scheduling priority scoring data. The hierarchical scheduling strategy includes at least two levels of scheduling strategy. Step 16: Distribute and schedule goods according to the hierarchical scheduling strategy.

[0020] In step 11 of this embodiment, warehouse operation data, including the time of goods entering and leaving the warehouse, storage location, order information, equipment status and environmental parameters, are collected in real time through the warehouse management system, sensor equipment or manual input.

[0021] In step 12, the warehousing operation data is cleaned, transformed, and normalized. Cleaning includes removing outliers and duplicate data; transformation involves feature extraction to map the raw data into standardized parameters, resulting in a set of basic parameters. These parameters include: cargo timeliness parameter (allowed time from goods entering to leaving the warehouse), cargo damage risk parameter (probability of damage during handling or storage), order fulfillment parameter (urgency of order delivery), and warehousing resource parameter (availability of current warehousing equipment and personnel).

[0022] In step 13, based on the basic parameter set, the cargo scheduling coefficient is determined through weighted calculation. The timeliness urgency coefficient is the reciprocal of the cargo timeliness parameter, the cargo damage risk coefficient is the linear mapping value of the cargo damage risk parameter, the order priority coefficient is the weighted score of the order fulfillment parameter, and the operational resource adaptation coefficient is the degree of matching between the warehousing resource parameters and the task requirements.

[0023] In step 14, dynamic scheduling priority score data is determined by using a dynamic priority model based on the cargo scheduling coefficient.

[0024] In step 15, based on the dynamic scheduling priority scoring data, the goods are divided into multiple levels of scheduling strategies, or more levels are set according to actual needs to form a hierarchical scheduling strategy.

[0025] In step 16, based on the hierarchical scheduling strategy, warehousing resources are allocated in order of priority, high-priority goods are processed first, and medium and low-priority tasks are coordinated synchronously to achieve parallel scheduling of multiple tasks.

[0026] The embodiments of this invention construct a priority scoring model with a nonlinear risk amplification factor. When the timeliness or damage risk of goods exceeds a critical value, its priority is amplified, ensuring that high-risk goods are handled preferentially from the scheduling rule level. Based on the scoring results, a four-level differentiated scheduling mechanism is established to achieve precise on-demand allocation of warehousing resources. Simultaneously, the accompanying digital twin simulation and emergency scheduling module can pre-verify the scheduling plan and handle risks in a closed-loop manner. This technical solution effectively reduces the overdue loss rate and improves the overall operational efficiency and resource utilization of warehousing.

[0027] In an optional embodiment of the present invention, step 12, which involves preprocessing the warehousing operation data to obtain a basic parameter set, may include: Step 121: Perform data cleaning on the warehousing operation data to obtain cleaned warehousing operation data; Step 122: Standardize the cleaned warehousing operation data to obtain standardized warehousing operation data; Step 123: Perform time and space alignment processing on the standardized warehousing operation data to obtain aligned warehousing operation data; Step 124: Perform feature extraction processing on the aligned warehousing operation data to obtain a basic parameter set.

[0028] In step 121 of this embodiment, the warehousing operation data is cleaned to obtain cleaned warehousing operation data. Specifically, this includes: First, screening the collected warehousing operation data for null and missing values; directly removing invalid data where the missing core fields (such as SKU, warehousing time, shelf life, and order fulfillment time) account for more than 30% of a single data entry; for incomplete data where the missing core fields account for less than 30%, interpolation is performed using the average value of the same dimension for the same batch and category of goods to complete the data; second, using... The criteria perform extreme value screening on numerical time-series data (such as warehouse temperature and humidity, equipment operating parameters, and storage duration data), and calculate the arithmetic mean of the same data dimension over the past 72 hours. and standard deviation It will exceed Extreme values ​​and noisy data within the interval are marked as outliers; finally, the isolated forest algorithm is used to classify data that meet the criteria. Hidden outliers that do not conform to business logic within a given range are identified and removed. The identification dimensions include: spatial conflicts between the coordinates of the storage location and the three-dimensional layout of the warehouse area, temporal conflicts where the time of goods entering the warehouse is later than the time of goods leaving the warehouse, and conflicts between the adaptability of cold chain temperature fluctuations and the storage requirements of goods. After the full outlier removal is completed, the cleaned warehouse operation data is obtained.

[0029] In step 122, the cleaned warehousing operation data is standardized to obtain standardized warehousing operation data. Specifically, this includes: performing a unified standardization conversion on warehousing operation data of different formats and units. Numerical data: The min-max normalization method is used to linearly map all numerical data to the interval [0,1]. The normalization formula is: ; in, X These are the original values. This is the historical minimum value for this data dimension. This is the historical maximum value for this data dimension, eliminating the influence of different units on subsequent calculations. Text-based data: Text-based data such as order type, product category, and equipment status are converted into structured numerical identifier data through a preset business mapping dictionary. For example, emergency orders are mapped to 1, regular retail orders to 0.6, replenishment and transfer orders to 0.4, and inventory transfer orders to 0.2. Event-based data: Event-based data such as equipment failure, temperature and humidity exceeding limits, and order changes are converted into structured data containing timestamps, event types, and impact levels. The impact levels are normalized to the [0,1] range. After standardizing the format and dimensions of all data, standardized warehouse operation data is obtained.

[0030] In step 123, the standardized warehousing operation data is subjected to time and space alignment processing to obtain aligned warehousing operation data, specifically including: Time dimension alignment: A unified UTC time axis is established with 100ms as the smallest time granularity. For high-frequency acquired device operation data (acquisition frequency 500ms / time), the mean downsampling method is used to synchronize it to a 100ms time granularity. For low-frequency acquired environmental sensor data (acquisition frequency 10s / time), the linear interpolation method is used to complete it to a 100ms time granularity, ensuring that the timestamps of all data are completely aligned and eliminating asynchronous deviations in the time dimension. Spatial Dimension Alignment: Based on the three-dimensional Cartesian coordinate system of the warehouse area, a unified spatial mapping relationship is established for storage locations, goods, sensor nodes, and handling equipment. All entity objects corresponding to the data are uniformly marked in the three-dimensional coordinate system. X , Y , Z The axis coordinate values ​​clarify the spatial location and adjacency relationship of each entity, ensuring that all data has spatial correlation. After completing the spatiotemporal alignment, the aligned warehousing operation data is obtained.

[0031] In step 124, feature extraction processing is performed on the aligned warehousing operation data to obtain a basic parameter set, specifically including: An improved Deep Belief Network (DBN) model is adopted to perform feature-level fusion and extraction on the spatiotemporally aligned full data. The model is a three-tiered structure: "input layer → first RBM (Restricted Boltzmann Machine) layer → second RBM layer → BP (Fully Linked) output layer". Each layer node is customized according to the feature dimensions of the warehouse data and business needs, with no redundancy design. Input layer: The number of nodes is consistent with the total number of feature dimensions of the full warehouse data after spatiotemporal alignment. It only receives and forwards data and does not participate in the calculation. Two RBM layers: The first RBM layer compresses the original data dimension to half and extracts shallow correlation features to complete the initial fusion of multi-source data; the second RBM layer further filters and fuses the data, outputting deep core features that are strongly correlated with warehouse scheduling and removing redundant information. BP output layer: The hidden layer nodes are matched with the second RBM layer, and the output layer nodes are fixed as the sum of the feature dimensions of the four basic parameters, which is specifically used to realize the business-oriented classification mapping of deep features.

[0032] A hybrid training approach is adopted, combining unsupervised layer-by-layer pre-training with a small amount of supervised fine-tuning, to balance the ability to fuse massive heterogeneous data with adaptability to warehousing operations. Unsupervised layer-by-layer pre-training: Using spatiotemporally aligned unlabeled warehouse data as samples, the CD-3 contrastive divergence algorithm is used to train the two RBM layers sequentially. With the goal of minimizing reconstruction error, the model learns the spatiotemporal correlation and business coupling of warehouse data autonomously, and completes the deep fusion of multi-source heterogeneous data. Small-scale supervised fine-tuning: Select 5%-10% of manually labeled samples (labels are the feature values ​​of 4 basic parameters), with the goal of minimizing the mean square error between the model output and the manual labeling, fine-tune the weights of the BP layer through the gradient descent algorithm, and backpropagate the error to the RBM layer, so that the extracted features are precisely bound to the scheduling coefficient calculation requirements.

[0033] After the model is trained, the feature processing of real-time warehouse data is performed through unidirectional forward propagation, with no redundant steps, directly outputting a structured basic parameter set. Specific steps are as follows: Data input: The spatiotemporally aligned and standardized full-volume heterogeneous warehouse data is input into the model input layer according to the feature dimensions; Shallow fusion: Data passes through the first RBM layer, extracting shallow correlation features to complete the initial fusion of multi-source data and eliminate dimensional barriers; Deep filtering: The shallow features are processed through the second RBM layer to remove features that are irrelevant to scheduling, strengthen the core features related to timeliness, risk, performance, and resources, and output a deep fusion feature set; Business classification mapping: Deep features are input into the BP output layer and mapped to four types of features according to the needs of warehouse scheduling business: cargo timeliness, cargo damage risk, order fulfillment, and warehouse resources. Parameter set output: The four types of features are structurally integrated to output a standardized basic parameter set. This parameter set can be directly substituted into the subsequent scheduling coefficient calculation formula without additional feature transformation.

[0034] In an optional embodiment of the present invention, step 13, determining the cargo scheduling coefficient based on the basic parameter set, may include: Step 131: Determine the timeliness urgency coefficient based on the cargo timeliness parameters in the basic parameter set, wherein the cargo timeliness parameters include at least the cargo-related order identifier, remaining fulfillment time, total order fulfillment time, remaining shelf life, and total shelf life of the goods; Step 132: Determine the cargo damage risk coefficient based on the cargo damage risk parameters in the basic parameter set, wherein the cargo damage risk parameters include at least the environmental deviation risk value, the storage duration risk value, and the inherent cargo damage risk value. Step 133: Determine the order priority coefficient based on the order fulfillment parameters in the basic parameter set, wherein the order fulfillment parameters include at least the order type identifier; Step 134: Determine the operational resource adaptation coefficient based on the warehousing resource parameters in the basic parameter set, wherein the warehousing resource parameters include at least the location adaptation value and the equipment resource adaptation value.

[0035] In step 131 of this embodiment, the urgency coefficient is determined based on the cargo timeliness parameters in the basic parameter set, specifically including: Depending on whether the goods are associated with a valid fulfillment order, two calculation methods are used, and all calculation results are normalized to the [0,1] interval: (1) For goods with associated valid fulfillment orders: timeliness coefficient T The calculation formula is: ; in, This represents the remaining time until the order fulfillment deadline at the current moment. The total fulfillment time from order creation to fulfillment deadline; when the remaining fulfillment time... hour, T Forced to set to 1; synchronized update when the order fulfillment deadline is brought forward. and The value was recalculated. T value; (2) For inventory goods not associated with valid fulfillment orders: the formula for calculating the timeliness urgency coefficient is as follows: ; in, This represents the remaining shelf life of the goods at the current moment. This represents the total shelf life of the goods; when the remaining shelf life is lower than the preset safe shelf life threshold (the default setting is 10% of the total shelf life), T Forced setting to 1 ensures priority scheduling for goods nearing their expiration date.

[0036] In step 132, the cargo damage risk coefficient is determined based on the cargo damage risk parameters in the basic parameter set, specifically including: Cargo damage risk coefficient R The weighted sum of environmental deviation risk value, storage duration risk value, and inherent cargo damage risk value is calculated using the following formula: ; in, , , The preset weighting coefficients satisfy... The default value is , , It can adaptively adjust according to the type of goods; The environmental deviation risk value is calculated using the following formula: ; in, This refers to the real-time ambient temperature of the warehouse area where the goods are currently located. The standard temperature for storing goods. The maximum allowable temperature fluctuation threshold for goods; when hour, Forced to set to 1; The formula for calculating the storage duration risk value is as follows: ; in, This refers to the cumulative duration of time the goods have been stored in the warehouse. The maximum safe storage time for goods; when hour, Forced to set to 1; This is the inherent risk value of the goods, pre-determined based on the goods' category, fragility, and value per kilogram, with a value range of [0, 1]. This includes easily damaged and high-value cold chain fresh goods. The default value is 0.8 to 1, which is typical for general cargo. The default value is between 0.2 and 0.5; The final calculated cargo damage risk coefficient R ranges from [0, 1]. When any sub-parameter exceeds the safety threshold, R is forcibly set to 1.

[0037] In step 133, the order priority coefficient is determined based on the order fulfillment parameters in the basic parameter set, specifically including: Based on the order type identifier, a fixed order priority coefficient is preset. O The value range is [0, 1]. The specific calibration rules are as follows: emergency supply orders and emergency outbound orders for goods nearing their expiration date. O =1; next-day delivery, same-day delivery, and other instant-fulfillment retail orders. O =0.8; Regular retail fulfillment orders, O =0.6; Cross-warehouse replenishment and transfer orders, O =0.4; Non-fulfillment operations such as inventory counting, transfer, and sorting within the warehouse. O =0.2; It also supports manual adjustment of order priority coefficients, with the adjustment range not exceeding the [0, 1] interval. After adjustment, the corresponding order's associated goods will be updated synchronously. O value.

[0038] In step 134, the operational resource adaptation coefficient is determined based on the warehousing resource parameters in the basic parameter set, specifically including: The operational resource adaptation coefficient A is the weighted sum of the storage location adaptation value and the equipment resource adaptation value, and the calculation formula is as follows: ; in, , The preset weighting coefficients satisfy... The default value is ; The formula for calculating the location adaptation value is as follows: ; in, This is the straight-line distance between the current location of the goods and the exit. This represents the straight-line distance from the farthest storage location in the warehouse area to the exit, with a value ranging from [0, 1]. The closer the storage location is to the exit, the better. The higher the value; The formula for calculating the device resource adaptation value is as follows: ; in, This represents the number of available idle handling equipment (stall cranes, AGVs, etc.) in the current warehouse area. The total number of handling equipment configured for the warehouse area, with a value range of [0, 1]. The more equipment available, the better. The higher the value, the better the final calculated job resource adaptation coefficient. A The value range is [0, 1].

[0039] In an optional embodiment of the present invention, step 14, determining the dynamic scheduling priority scoring data based on the cargo scheduling coefficient, may include: Step 141: Input the cargo scheduling coefficient into a preset priority scoring model for processing to obtain dynamic scheduling priority scoring data. The priority scoring model is as follows: ; in, This is for dynamic scheduling priority scoring data; This is the timeliness and urgency coefficient; This refers to the risk coefficient of cargo damage. This is the order fulfillment priority coefficient; The resource adaptation coefficient for the task; , , The preset weighting coefficients, and ; It is a non-linear risk amplification factor; To obtain the aforementioned timeliness coefficient and cargo damage risk coefficient The maximum value in.

[0040] In step 141 of this embodiment, the cargo scheduling coefficient is input into a preset priority scoring model for processing to obtain dynamic scheduling priority scoring data. Specifically, this includes: processing the calculated timeliness urgency coefficient... T Risk coefficient of cargo damage R Order fulfillment priority coefficient O Job resource adaptation coefficient A Substituting the values ​​into the preset priority scoring model, numerical calculations are performed to obtain dynamic scheduling priority scoring data with values ​​ranging from [0, 100]. P Among them, the weighting coefficient , , The default value is , , 5. The core emphasis is on the impact weight of cargo damage risk on priority, which can be adaptively adjusted according to warehousing business scenarios; the priority scoring model uses a non-linear risk amplification factor. Prioritize goods with high time sensitivity and high risk of damage to prevent them from being diluted and squeezed out by a large number of regular low-priority orders.

[0041] Wherein, the nonlinear risk amplification factor The method for determining the value is as follows: the nonlinear risk amplification factor is compared with a preset critical threshold. When the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is greater than or equal to the critical threshold, the nonlinear risk amplification factor is set to a first value; when the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is less than the critical threshold, the nonlinear risk amplification factor is set to a second value; wherein, the first value is greater than the second value; the specific process includes: The preset critical threshold is 0.9, the preset first value is 2, and the preset second value is 1; Calculate the timeliness urgency coefficient of the current goods. T and cargo damage risk coefficient R The maximum value in, i.e. ; when At 9:00 AM, the goods were determined to be high-risk / time-sensitive urgent goods, and a non-linear risk amplification factor was applied. The priority score is forcibly set to 2, which exponentially amplifies the priority score and ensures that the scheduling priority of this cargo is absolutely the highest. when At 9:00, the goods were determined to be regular cargo, and the nonlinear risk amplification factor was applied. Set to 1 to maintain the linear calculation logic of the basic priority score; When goods are overdue or storage conditions are severely exceeded, or when other extreme abnormalities occur, T or R Forced to set to 1, at this time 9, Automatically set to 2, triggering the priority amplification mechanism.

[0042] In an optional embodiment of the present invention, step 15, determining a hierarchical scheduling strategy based on the dynamic scheduling priority scoring data, may include: Step 151: Divide the dynamic scheduling priority scoring data into levels according to at least two preset level thresholds to obtain scheduling level data; Step 152: Determine the scheduling execution rules based on the scheduling level data. The scheduling execution rules include resource allocation priority, job path selection strategy, and task interruption strategy.

[0043] In step 151 of this embodiment, the dynamic scheduling priority score data is classified into scheduling level data according to at least two preset classification thresholds. Specifically, this includes: setting four classification thresholds to divide the dynamic scheduling priority score data P into four scheduling levels, forming a four-level classification system of S / A / B / C. The specific classification rules are as follows: S-level dispatch level: P≥80 points, which is an emergency high-risk / high-timeliness level; Level A dispatching level: 60 points ≤ P < 80 points, which is a high-timeliness and high-value level; Level B dispatching: 30 minutes ≤ P < 60 minutes, which is the standard timeliness level; Level C scheduling: P < 30 points, which is a low-timeliness and low-priority level; Based on the above classification rules, corresponding scheduling level data is matched for each batch of goods / each operation task. For multiple batches of goods associated with the same order, the scheduling level corresponding to the highest priority score is taken as the unified scheduling level.

[0044] In step 152, scheduling execution rules are determined based on the scheduling level data. These rules include resource allocation priority, job path selection strategy, and task interruption strategy. Specifically, they include: matching differentiated scheduling execution rules to different scheduling levels to achieve refined hierarchical management. (1) Execution rules corresponding to S-level scheduling: Resource allocation priority: Adopting a resource exclusive strategy, priority is given to allocating the best-performing idle handling equipment in the warehouse area and the outbound port corresponding to the shortest operation path. If there is no idle equipment that can be directly called, the non-S-level operation task being executed can be interrupted to release equipment resources and prioritize S-level tasks. Operation route selection strategy: The shortest direct path strategy is adopted. The Dijkstra algorithm is used to plan the shortest conflict-free path from the storage location to the exit, avoiding all congested sections of the regular operation channels and giving priority to the right of way. Task interruption policy: It can only be interrupted by S-level tasks of the same level, and cannot be interrupted by A-level or lower-level tasks during execution; (2) Execution rules corresponding to Level A scheduling: Resource allocation priority: A priority scheduling strategy is adopted. During the execution intervals of S-level tasks, the best available equipment resources are allocated first, and B-level and above tasks that are currently being executed must not be interrupted. Job path selection strategy: Employ the optimal conflict-free path strategy, through... The algorithm plans the operation path, avoids congested sections, and ensures operation efficiency; Task interruption policy: Can be interrupted by S-level tasks, but cannot be interrupted by B-level or lower-level tasks during execution; (3) Execution rules corresponding to Level B scheduling: Resource allocation priority: A balanced scheduling strategy is adopted, which, in combination with the equipment load balance, allocates idle equipment resources in batches to ensure a balanced workload in the warehouse area. Operation path selection strategy: Adopt first-in-first-out batch operation path planning, and combine and schedule tasks in the same warehouse area and along the same path to reduce equipment idle running; Task interruption policy: Can be interrupted by S-level and A-level tasks, but cannot be interrupted by C-level tasks during execution; (4) Execution rules corresponding to Level C scheduling: Resource allocation priority: Adopt the idle time reuse strategy, and only use the remaining idle equipment resources to arrange operations during the off-peak period of warehouse orders (such as from 22:00 on the same day to 6:00 on the next day), and do not occupy the operation resources during peak hours; Job path selection strategy: Adopt the idle path reuse strategy, use the idle channels of regular jobs to plan the path, and do not occupy the main job channel; Task interruption strategy: It can be interrupted at any time by S-level, A-level, and B-level tasks, with priority given to ensuring the resource needs of higher-level tasks.

[0045] In an optional embodiment of the present invention, step 16, which involves allocating and scheduling goods according to the hierarchical scheduling strategy, may include: Step 161: Generate scheduling task instructions for each cargo according to the scheduling execution rules corresponding to different levels in the hierarchical scheduling strategy; Step 162: Send the scheduling task instruction to the warehouse control system to control the warehouse equipment to perform the corresponding operations; Step 163: Collect status data during the job execution process in real time and update the corresponding parameters in the basic parameter set.

[0046] In step 161 of this embodiment, scheduling task instructions for each cargo are generated according to the scheduling execution rules corresponding to different levels in the hierarchical scheduling strategy. Specifically, this includes: according to the hierarchical scheduling strategy, decomposing and generating standardized operation task instructions, each task instruction including at least: a unique task identifier, cargo SKU, cargo location coordinates, target location, operation type, execution equipment number, task priority level, task execution time limit, and path planning result; for S-level urgent tasks, marking the highest priority identifier in the task instruction and setting a mandatory execution time limit to ensure priority processing; for batch tasks of the same level and the same path, merging and packaging them to generate batch operation instructions to improve operation efficiency.

[0047] In step 162, the scheduling task instruction is sent to the warehouse control system to control the warehouse equipment to perform corresponding operations. Specifically, this includes: sending the scheduling task instruction to the WCS warehouse control system through a preset industrial communication interface; the WCS system, based on the instruction content, sends control instructions to the corresponding warehouse equipment such as stacker cranes, AGVs, and conveyors to drive the equipment to complete operations such as outbound, relocation, and handling of goods; for S-level emergency tasks, the WCS system prioritizes processing their instructions, suspends equipment actions for non-S-level tasks, and releases resources to ensure the execution of emergency tasks; during the operation execution, the WCS system transmits back the execution status, location, and operation progress data of the equipment in real time.

[0048] In step 163, real-time status data is collected during the operation execution process, and the corresponding parameters in the basic parameter set are updated. Specifically, this includes: collecting equipment operating status, cargo location, order fulfillment progress, and warehouse environment status data in real time during the operation execution process at a period of 100ms, and synchronously updating the cargo timeliness parameters, cargo damage risk parameters, and warehousing resource parameters in the basic parameter set; removing cargo from the scheduling task queue after the outbound operation is completed; and immediately updating the corresponding parameters, recalculating the dynamic scheduling priority score of the cargo, adjusting the hierarchical scheduling strategy, and realizing dynamic closed-loop control of the scheduling process when equipment failure, cargo abnormality, or order change occurs during the operation execution process.

[0049] A specific embodiment of the warehouse collaborative scheduling data processing method provided in this invention is as follows: This embodiment is applied to a cold chain warehousing center for a fresh food e-commerce company. The center is equipped with 8 stacker cranes and 20 AGV handling robots. The warehouse is divided into a low-temperature frozen area, a refrigerated area, and a normal temperature area, primarily storing imported cold chain meat, fruits, vegetables, and other fresh goods. Under the traditional first-in-first-out (FIFO) scheduling model, the monthly excess loss rate is approximately 1.2%, and severe congestion occurs during peak hours, resulting in low scheduling efficiency. The specific execution steps of this embodiment are as follows: Step 1, Acquire Warehouse Operation Data: Through the warehouse IoT gateway, connect to temperature and humidity sensors, WMS warehouse management system, WCS equipment control system, and order management system to collect warehouse operation data in real time, including: goods SKU, inbound time, shelf life, storage temperature and humidity, order fulfillment time, equipment operating status, storage location coordinates, and warehouse layout data. The collection cycle matches the data type, with a 500ms data collection cycle for equipment data and a 10s data collection cycle for environmental data. Order data is synchronized in real time.

[0050] Step 2: Preprocess the warehouse operation data to obtain the basic parameter set: First, clean the collected data, remove invalid data, and fill in missing values. The criteria and the isolated forest algorithm are used to remove outliers; secondly, the cleaned data is subjected to min-max normalization to map all numerical data to the [0, 1] interval and convert text-based order types into standardized numerical identifiers; then, the data is spatiotemporally aligned, with time synchronization completed at a time granularity of 100ms and spatial location alignment completed using the warehouse area's three-dimensional coordinate system; finally, core features are extracted through a deep belief network to obtain a basic parameter set containing cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters.

[0051] Step 3, calculate cargo scheduling coefficient and dynamic scheduling priority score data: For the two batches of core cargo in this embodiment, the calculations are performed separately: (1) Target Goods 1: Imported frozen beef, total shelf life 90 days, stored for 65 days, associated with next-day fulfillment orders, remaining fulfillment time 12 hours, total order fulfillment time 48 hours, storage temperature -18℃, meets standard temperature requirements, storage location 100m from the outbound gate, the farthest storage location in the warehouse area 500m from the outbound gate, currently 8 available stacker cranes, total number of equipment 8; Calculated: The urgency coefficient T = 1 - (12 / 48) = 0.75; The risk coefficient for cargo damage is R = 0.4 × 0.1 + 0.35 × (65 / 90) + 0.25 × 0.8 = 0.492; Order priority coefficient O = 0.8; The job resource matching coefficient A = 0.5 × (1 - 100 / 500) + 0.5 × (8 / 8) = 0.9; The basic weighted sum = 0.35 × 0.75 + 0.4 × 0.492 + 0.25 × 0.8 = 0.6593; max(T,R)=0.75<0.9, ; The final dynamic scheduling priority score P = 100 × 0.6593 × (1 + 1 × 0.75) × 0.9 ≈ 104, taking the upper limit of 100 points, is classified as S-level scheduling level; (2) Target Goods 2: Regular domestic leafy vegetables, total shelf life 30 days, already stored for 2 days, associated with a regular retail order, remaining fulfillment time 40 hours, total order fulfillment time 48 hours, storage temperature meets requirements, storage location 300m from the outbound exit; Calculations show: The urgency coefficient T = 1 - (40 / 48) ≈ 0.167; The risk coefficient for cargo damage is R≈0.12; Order priority coefficient O = 0.6; The job resource adaptation coefficient A≈0.7; The basic weighted sum is approximately 0.256; max(T,R) = 0.167 < 0.9. The final dynamic scheduling priority score P≈25.4 points, classifying it as a C-level scheduling level; At the same time, 120 regular orders in the warehouse area were prioritized and graded, including 12 A-level tasks, 98 B-level tasks, and 10 C-level tasks.

[0052] Step 4: Determine the hierarchical scheduling strategy based on the hierarchical results and execute the scheduling job: For target cargo 1 (Level S), a resource monopoly strategy is implemented, allocating the optimal stacker crane, planning the shortest operation path, interrupting the ongoing Level C sorting task, and prioritizing the outbound operation of target cargo 1; for Level A tasks, priority scheduling is given during the execution intervals of Level S tasks; for Level B tasks, batch scheduling is performed according to the first-in-first-out rule; for Level C tasks, execution is scheduled during off-peak hours after 22:00 on the same day; during the operation execution process, equipment and cargo status data are collected in real time, and parameters and scheduling strategies are dynamically updated.

[0053] Step 5, scheduling result verification and strategy optimization: In this embodiment, target cargo 1 is fulfilled on time with no risk of overdue delivery. After applying this method, the monthly overdue loss rate of the warehousing center is reduced to 0.08%, which is 93% lower than the traditional model. The throughput during peak hours is increased by 35%, the equipment idle rate is reduced by 28%, and the scheduling congestion problem is completely solved. After the daily operation is completed, based on the scheduling execution data of the day, the weight coefficients and grading thresholds of the priority scoring model are iteratively optimized using a reinforcement learning algorithm to achieve continuous self-optimization of the scheduling strategy.

[0054] This invention proposes a warehouse collaborative scheduling data processing method that, through a dynamic priority scoring model with a nonlinear risk amplification factor, overturns the traditional static first-in-first-out (FIFO) rule of warehouse scheduling based on inbound time. It establishes a dynamic hierarchical scheduling mechanism deeply coupled with real-time cargo status, order demand, and warehouse resources. Through nonlinear amplification design, it locks in the priority scheduling rights of high-time-sensitivity and high-risk-for-damage goods at the scheduling rule level, fundamentally solving the industry problem of high-risk goods being diluted and squeezed out by regular orders, leading to overstocking and damage. Simultaneously, through full-process preprocessing of multi-source heterogeneous data, it ensures the accuracy of scheduling decisions. Through hierarchical and differentiated scheduling strategies, it achieves precise allocation of warehouse resources, significantly improving overall warehouse scheduling efficiency and resource utilization. This method is adaptable to all types of warehousing scenarios, including cold chain fresh produce, high-value perishable goods, and general e-commerce, exhibiting strong scenario adaptability and robustness. It can reduce warehousing operating costs and damage risks, and improve the overall service quality of the logistics supply chain.

[0055] like Figure 2As shown, this embodiment of the invention also provides a warehouse collaborative scheduling data processing device 20, comprising: Module 21 is used to acquire warehouse operation data; Processing module 22 is used to preprocess the warehousing operation data to obtain a basic parameter set, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters; determine cargo scheduling coefficients based on the basic parameter set, which include timeliness urgency coefficients, cargo damage risk coefficients, order priority coefficients, and operation resource adaptation coefficients; determine dynamic scheduling priority scoring data based on the cargo scheduling coefficients; determine a hierarchical scheduling strategy based on the dynamic scheduling priority scoring data, which includes at least two levels of scheduling strategy; and allocate and schedule cargo according to the hierarchical scheduling strategy.

[0056] Optionally, processing module 22 is specifically used for: The warehousing operation data is cleaned to obtain cleaned warehousing operation data; The cleaned warehousing operation data is standardized to obtain standardized warehousing operation data; The standardized warehousing operation data is then subjected to time and space alignment processing to obtain aligned warehousing operation data; The aligned warehousing operation data is processed by feature extraction to obtain a basic parameter set.

[0057] Optionally, the processing module 22 is also specifically used for: Based on the cargo timeliness parameters in the aforementioned basic parameter set, a timeliness urgency coefficient is determined, wherein the cargo timeliness parameters include at least the cargo-related order identifier, remaining fulfillment time, total order fulfillment time, remaining shelf life, and total shelf life of the goods; Based on the cargo damage risk parameters in the aforementioned basic parameter set, a cargo damage risk coefficient is determined, wherein the cargo damage risk parameters include at least the environmental deviation risk value, the storage duration risk value, and the inherent cargo damage risk value. Based on the order fulfillment parameters in the basic parameter set, an order priority coefficient is determined, wherein the order fulfillment parameters include at least an order type identifier; Based on the warehousing resource parameters in the basic parameter set, the operational resource adaptation coefficient is determined, wherein the warehousing resource parameters include at least the storage location adaptation value and the equipment resource adaptation value.

[0058] Optionally, the processing module 22 is also specifically used for: The cargo scheduling coefficients are input into a preset priority scoring model for processing to obtain dynamic scheduling priority scoring data. The priority scoring model is as follows: ; in, This is for dynamic scheduling priority scoring data; This is the timeliness and urgency coefficient; This refers to the risk coefficient of cargo damage. This is the order fulfillment priority coefficient; The resource adaptation coefficient for the task; , , The preset weighting coefficients, and ; It is a non-linear risk amplification factor; To obtain the aforementioned timeliness coefficient and cargo damage risk coefficient The maximum value in.

[0059] Optionally, the nonlinear risk amplification factor can be determined as follows: The nonlinear risk amplification factor is compared with a preset critical threshold. When the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is greater than or equal to the critical threshold, the nonlinear risk amplification factor is set to a first value; when the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is less than the critical threshold, the nonlinear risk amplification factor is set to a second value. The first value is greater than the second value.

[0060] Optionally, the processing module 22 is also specifically used for: The dynamic scheduling priority scoring data is divided into levels according to at least two preset grading thresholds to obtain scheduling level data; Based on the scheduling level data, scheduling execution rules are determined, including resource allocation priority, job path selection strategy, and task interruption strategy.

[0061] Optionally, the processing module 22 is also specifically used for: Based on the scheduling execution rules corresponding to different levels in the hierarchical scheduling strategy, scheduling task instructions for each cargo are generated. The scheduling task instructions are sent to the warehouse control system to control the warehouse equipment to perform corresponding operations; The status data during the job execution process is collected in real time, and the corresponding parameters in the basic parameter set are updated accordingly.

[0062] It should be noted that this device is a device corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0063] like Figure 3As shown, this embodiment of the invention also provides a computing device 30, including a processor 31, a memory 32, and a program or instructions stored in the memory 32 and executable on the processor 31. When the program or instructions are executed by the processor 31, they implement the various processes of the above-described warehouse collaborative scheduling data processing method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here. It should be noted that the computing device in this embodiment of the invention includes the above-described mobile electronic devices and non-mobile electronic devices.

[0064] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0065] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0066] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0067] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0068] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0069] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0070] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve by using their basic programming skills after reading the description of the present invention.

[0071] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code for implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps for performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0072] The above are preferred embodiments 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 warehouse collaborative scheduling data processing method, characterized in that, include: Obtain warehouse operation data; The warehousing operation data is preprocessed to obtain a set of basic parameters, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters. Based on the aforementioned set of basic parameters, a cargo scheduling coefficient is determined, which includes a timeliness urgency coefficient, a cargo damage risk coefficient, an order priority coefficient, and an operational resource adaptation coefficient. Based on the cargo scheduling coefficient, determine the dynamic scheduling priority score data; Based on the dynamic scheduling priority scoring data, a hierarchical scheduling strategy is determined, which includes at least two levels of scheduling strategy. According to the hierarchical scheduling strategy, the goods are allocated and scheduled.

2. The warehouse collaborative scheduling data processing method according to claim 1, characterized in that, The warehousing operation data is preprocessed to obtain a set of basic parameters, including: The warehousing operation data is cleaned to obtain cleaned warehousing operation data; The cleaned warehousing operation data is standardized to obtain standardized warehousing operation data; The standardized warehousing operation data is then subjected to time and space alignment processing to obtain aligned warehousing operation data; The aligned warehousing operation data is processed by feature extraction to obtain a basic parameter set.

3. The warehouse collaborative scheduling data processing method according to claim 1, characterized in that, Based on the aforementioned set of basic parameters, the cargo scheduling coefficients are determined, including: Based on the cargo timeliness parameters in the aforementioned basic parameter set, a timeliness urgency coefficient is determined, wherein the cargo timeliness parameters include at least the cargo-related order identifier, remaining fulfillment time, total order fulfillment time, remaining shelf life, and total shelf life of the goods; Based on the cargo damage risk parameters in the aforementioned basic parameter set, a cargo damage risk coefficient is determined, wherein the cargo damage risk parameters include at least the environmental deviation risk value, the storage duration risk value, and the inherent cargo damage risk value. Based on the order fulfillment parameters in the basic parameter set, an order priority coefficient is determined, wherein the order fulfillment parameters include at least an order type identifier; Based on the warehousing resource parameters in the basic parameter set, the operational resource adaptation coefficient is determined, wherein the warehousing resource parameters include at least the storage location adaptation value and the equipment resource adaptation value.

4. The warehouse collaborative scheduling data processing method according to claim 1, characterized in that, Based on the aforementioned cargo scheduling coefficient, dynamic scheduling priority scoring data is determined, including: The cargo scheduling coefficients are input into a preset priority scoring model for processing to obtain dynamic scheduling priority scoring data. The priority scoring model is as follows: ； in, This is for dynamic scheduling priority scoring data; This is the timeliness and urgency coefficient; This refers to the risk coefficient of cargo damage. This is the order fulfillment priority coefficient; The resource adaptation coefficient for the task; , , The preset weighting coefficients, and ; It is a non-linear risk amplification factor; To obtain the aforementioned timeliness coefficient and cargo damage risk coefficient The maximum value in.

5. The warehouse collaborative scheduling data processing method according to claim 4, characterized in that, The nonlinear risk amplification factor is determined as follows: The nonlinear risk amplification factor is compared with a preset critical threshold. When the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is greater than or equal to the critical threshold, the nonlinear risk amplification factor is set to a first value. When the maximum value of the timeliness urgency coefficient and the cargo damage risk coefficient is less than the critical threshold, the nonlinear risk amplification factor is set to a second value; The first value is greater than the second value.

6. The warehouse collaborative scheduling data processing method according to claim 1, characterized in that, Based on the dynamic scheduling priority scoring data, a hierarchical scheduling strategy is determined, including: The dynamic scheduling priority scoring data is divided into levels according to at least two preset grading thresholds to obtain scheduling level data; Based on the scheduling level data, scheduling execution rules are determined, including resource allocation priority, job path selection strategy, and task interruption strategy.

7. The warehouse collaborative scheduling data processing method according to claim 1, characterized in that, According to the hierarchical scheduling strategy, the cargo allocation and scheduling operation includes: Based on the scheduling execution rules corresponding to different levels in the hierarchical scheduling strategy, scheduling task instructions for each cargo are generated. The scheduling task instructions are sent to the warehouse control system to control the warehouse equipment to perform corresponding operations; The status data during the job execution process is collected in real time, and the corresponding parameters in the basic parameter set are updated accordingly.

8. A warehouse collaborative scheduling data processing device, characterized in that, include: The acquisition module is used to acquire warehouse operation data; The processing module is used to preprocess the warehousing operation data to obtain a basic parameter set, which includes cargo timeliness parameters, cargo damage risk parameters, order fulfillment parameters, and warehousing resource parameters; based on the basic parameter set, determine cargo scheduling coefficients, which include timeliness urgency coefficients, cargo damage risk coefficients, order priority coefficients, and operation resource adaptation coefficients; based on the cargo scheduling coefficients, determine dynamic scheduling priority scoring data; based on the dynamic scheduling priority scoring data, determine a hierarchical scheduling strategy, which includes at least two levels of scheduling strategy; and allocate and schedule cargo according to the hierarchical scheduling strategy.

9. A computing device, characterized in that, include: A processor, a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images