A data center storage location allocation method, device, equipment, medium and product

Abstract

The application discloses a data center storage location allocation method, device, equipment, medium and product. It relates to the technical field of data center material storage. The method comprises the following steps: determining the minimum stock units to which each device in the data center belongs, and constructing a device correlation matrix according to the co-occurrence times between each minimum stock unit; for any candidate storage location allocation scheme, determining the total time consumption according to the time consumption of each storage location and the preset turnover frequency of each device; determining the stability according to the device weight distribution and the geometric parameters of the shelf; determining the separation penalty value according to the device and storage location allocation relationship, the distance between each storage location and the device correlation matrix; determining the time consumption difference between the time consumption of the first-in storage device and the time consumption of the last-in storage device in each minimum stock unit, and determining the first-in first-out violation penalty value according to the time consumption difference; generating an evaluation vector according to the total time consumption, the stability, the separation penalty value and the first-in first-out violation penalty value. The quality and engineering practicability of the storage location allocation scheme are improved.

Description

Technical Field

[0001] This invention relates to the field of data center material storage technology, and in particular to a data center storage location allocation method, device, equipment, medium and product. Background Technology

[0002] With the rapid expansion of the digital economy and computing power, the asset management of hyperscale data centers is accelerating its upgrade towards automated storage and retrieval systems (AS / RS). Data center spare parts warehouses differ from ordinary commodity warehouses; their IT equipment has unique attributes such as strong interrelationships, strict asset depreciation deadlines, and a high degree of diversity in specifications and models, placing higher demands on warehouse location allocation strategies.

[0003] Currently, automated storage and retrieval systems (AS / RS) primarily employ two technical approaches for allocating storage locations for incoming materials. The first approach is a heuristic rule-based method based on a single indicator, typically the ABC turnover rate classification method. This method categorizes materials into three classes (A, B, and C) based on their outbound frequency, allocating high-frequency materials to the area closest to the outbound exit. While simple to implement, this method only considers a single dimension and cannot adapt to the complex scenarios of multi-objective constraints in data center equipment. The second approach is a fixed-weight multi-objective weighted evaluation method. This method integrates multiple objectives such as efficiency and shelf stability, using manually set fixed weight coefficients to linearly weight and obtain a comprehensive score, which then drives the allocation decision. Although this method considers multiple objectives, it suffers from fundamental flaws such as the "curse of dimensions" and "subjective weight distortion."

[0004] However, existing technologies still have shortcomings in addressing the unique business attributes of data center IT equipment. For example, efficiency and stability objectives have drastically different physical meanings, and existing technologies use subjectively set weighting coefficients for linear addition. This cross-dimensional integration method fails to accurately reflect the physical conflicts within the system. Furthermore, data center IT equipment exhibits strong clustering characteristics when issuing work orders, but existing evaluation systems treat each device as an independent entity, failing to fully consider the interconnectedness of outbound processes. Additionally, the first-in, first-out (FIFO) rule in data center warehousing only applies to similar types of materials, while existing algorithms often generalize it to a global constraint across SKUs (Stock Keeping Units), deviating from actual business logic.

[0005] Therefore, there is an urgent need for a storage location allocation method that can overcome the above-mentioned shortcomings and better suit the characteristics of data center warehousing operations. Summary of the Invention

[0006] This invention provides a data center storage location allocation method, apparatus, equipment, medium, and product to solve the technical problems in existing storage location allocation methods, such as evaluation distortion caused by subjective weighting across dimensions, decreased outbound efficiency due to neglect of the inter-equipment correlation, and the disconnect between the first-in-first-out rule and the business logic of similar material management.

[0007] According to one aspect of the present invention, a data center storage location allocation method is provided, comprising: Determine the smallest inventory unit to which each device belongs within the data center, and construct a device correlation matrix based on the co-occurrence frequency among the smallest inventory units in historical outbound work orders; For any candidate storage location allocation scheme, the total time is determined based on the retrieval time of each storage location and the preset turnover frequency of each device; stability is determined based on the weight distribution of the devices and the geometric parameters of the shelves; separation penalty value is determined based on the allocation correspondence between devices and storage locations, the distance between each storage location, and the device correlation matrix; within each minimum inventory unit, the time difference between the retrieval time of the first-in-first-out device and the retrieval time of the later-in-first-out device is determined, and the first-in-first-out violation penalty value of the candidate storage location allocation scheme is determined based on the time difference. An evaluation vector is generated based on the total time consumption, stability, separation penalty value, and first-in-first-out violation penalty value to evaluate the candidate storage location allocation scheme.

[0008] According to another aspect of the present invention, a data center storage location allocation device is provided, comprising: The association matrix determination module is used to determine the smallest inventory unit to which each device belongs in the data center, and to construct a device association matrix based on the number of times each smallest inventory unit co-occurs in historical outbound work orders; The total time determination module is used to determine the total time for any candidate warehouse location allocation scheme based on the picking time of each warehouse location and the preset turnover frequency of each device. The stability determination module is used to determine stability based on the equipment weight distribution and shelf geometry parameters. The separation penalty determination module is used to determine the separation penalty value based on the allocation correspondence between equipment and storage location, the distance between each storage location, and the equipment correlation matrix. The violation penalty determination module is used to determine the time difference between the time taken by the first-entry equipment to retrieve goods and the time taken by the second-entry equipment to retrieve goods within each smallest inventory unit, and to determine the first-in-first-out violation penalty value of the candidate storage location allocation scheme based on the time difference. The evaluation vector generation module is used to generate an evaluation vector based on the total time consumption, stability, separation penalty value and first-in-first-out violation penalty value to evaluate the candidate storage location allocation scheme.

[0009] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the data center storage location allocation method according to any embodiment of the present invention.

[0010] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the data center storage location allocation method according to any embodiment of the present invention.

[0011] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the data center storage location allocation method according to any embodiment of the present invention.

[0012] According to another aspect of the present invention, a computer program product is provided, comprising a computer program / instructions that, when executed by a processor, implement the data center storage location allocation method as described in any embodiment of the present invention.

[0013] This invention achieves objective quantitative evaluation of candidate warehouse location allocation schemes across multiple dimensions by isomorphically grouping equipment according to the smallest inventory unit, constructing a correlation matrix based on historical work orders, and generating a four-dimensional evaluation vector that includes total time consumption, stability, separation penalty value, and FIFO violation penalty value. It abandons the traditional method of subjective weighting across dimensions, avoiding the misleading optimization direction caused by weight distortion. Simultaneously, by introducing equipment matching correlation and grouping FIFO constraints, the evaluation results better align with the actual business needs of strong matching of data center IT equipment and prevention of expired similar materials, effectively improving the overall quality and engineering practicality of warehouse location allocation schemes.

[0014] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a data center storage location allocation method provided in an embodiment of the present invention; Figure 2 This is a flowchart of a data center storage location allocation method provided in an embodiment of the present invention; Figure 3This is a schematic diagram of the structure of a data center storage location allocation device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the data center storage location allocation method of this invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] Figure 1 This is a flowchart of a data center storage location allocation method provided by an embodiment of the present invention. This embodiment is applicable to the scenario of storage location allocation and optimization evaluation of IT equipment (including servers, network switches, storage arrays, etc.) in an automated three-dimensional warehouse of a data center. This method can be executed by a data center storage location allocation device, which can be implemented in hardware and / or software. This device can be configured in an electronic device with corresponding data processing capabilities. Figure 1 As shown, the method includes: S110. Determine the smallest inventory unit to which each device belongs in the data center, and construct a device correlation matrix based on the number of times each smallest inventory unit co-occurs in historical outbound work orders.

[0020] A Stock Keeping Unit (SKU) is a unique set of equipment determined by three dimensions: equipment type, brand, and specific model. For example, a 2U rack server V6 from a certain brand and a 2U rack server V5 from the same brand belong to different SKUs. Servers from different brands but with the same model also belong to different SKUs. SKU identification allows for the precise classification of highly diverse IT equipment in data centers, providing a foundation for subsequent first-in, first-out (FIFO) management based on similar equipment.

[0021] Optionally, an equipment correlation matrix is ​​constructed based on the co-occurrence frequency among the smallest inventory units in historical outbound work orders, including: determining the co-occurrence frequency of any pair of smallest inventory units in the same batch of historical outbound work orders, and the total historical outbound frequency of each smallest inventory unit; determining the symmetric correlation degree between each pair of smallest inventory units based on the co-occurrence frequency and the total historical outbound frequency; and constructing an equipment correlation matrix based on the symmetric correlation degree among all smallest inventory units.

[0022] The number of times each pair of SKUs co-occurs in the same batch of historical outbound work orders is counted, along with the total historical outbound count for each SKU. The symmetric correlation degree between each pair of SKUs is calculated based on the ratio of co-occurrence count to the total historical outbound count. Specifically, the symmetric correlation degree is obtained by averaging the unidirectional correlation degrees in both directions, with a value ranging from 0 to 1, where 1 indicates that the two SKUs are always shipped together, and 0 indicates that the two SKUs are never related. An equipment correlation degree matrix is ​​constructed based on the symmetric correlation degrees among all SKUs. This matrix quantifies the degree of compatibility between different devices. When historical outbound work order data is insufficient, business experts can manually initialize this equipment correlation degree matrix according to equipment compatibility rules.

[0023] For example, determine the minimum inventory unit With the smallest inventory unit Co-occurrence frequency in the same outbound batch ,as well as Total number of outbound shipments in history Calculate the first one-way correlation degree: Based on the number of co-occurrences and Total number of outbound shipments in history Calculate the second single-item correlation degree Determine the minimum inventory unit based on the first and second unidirectional correlation degrees. With the smallest inventory unit Symmetrical correlation between them: Construct an equipment correlation matrix based on the symmetric correlation between all the smallest inventory units.

[0024] By statistically analyzing the co-occurrence frequency of SKUs in historical outbound work orders and calculating bidirectional correlation, a matrix capable of quantifying the degree of compatibility between equipment was constructed. This transformed discrete historical business data into structured, computable parameters, providing an objective data foundation for correlation evaluation. Bidirectional symmetry processing ensured the mathematical symmetry of the correlation matrix, resulting in consistency and reversibility when subsequently calculating the separation penalty values ​​between equipment pairs. The equipment correlation matrix enabled the warehouse location allocation scheme to constrain strongly correlated equipment to nearby warehouse locations, thereby reducing the cross-regional round-trip travel of stacker cranes when performing complete outbound tasks and improving overall outbound efficiency.

[0025] S120. For any candidate warehouse location allocation scheme, determine the total time based on the picking time of each warehouse location and the preset turnover frequency of each device.

[0026] Obtain the physical structure parameters of the automated storage and retrieval system (AS / RS), including the number of rack rows, the number of layers, and the physical width and height of each storage location. Determine the physical coordinates of each storage location based on the number of rack rows, layers, and the physical width and height of each location. For example, the number of rack rows in the AS / RS... Number of floors And the physical width of each storage location in the shelving. With physical height The set of storage locations is denoted as Storage location physical coordinates Its value is determined by its column number, layer number, and the width and height of the storage location. , .

[0027] Obtain the motion parameters of the automated stacker crane, including its horizontal and vertical operating speeds. Determine the set of equipment in the racking system, including already stored equipment and equipment to be allocated in candidate storage location allocation schemes, and obtain the physical weight and expected turnover frequency of each piece of equipment. The expected turnover frequency can be predicted based on historical outbound data through time series analysis. Describe the allocation correspondence between equipment and storage locations using Boolean decision variables, such as Boolean decision variables... When the device Assigned to storage location hour ,otherwise And for any device satisfy Based on this, according to the stacker crane's motion characteristics, the physical coordinates of each storage location, the expected turnover frequency of each piece of equipment, and the allocation correspondence between equipment and storage locations, the weighted sum of the time consumption of all equipment is calculated as the total time consumption of the candidate storage location allocation scheme. The total time consumption is used to quantify the performance of the storage location allocation scheme in terms of inbound and outbound efficiency.

[0028] Optionally, the total time is determined based on the retrieval time of each storage location and the preset turnover frequency of each piece of equipment, including: determining the one-way retrieval time from the start and end points of the stacker crane to the storage location based on the horizontal and vertical running speeds of the stacker crane and the horizontal and vertical coordinates of the storage location allocated to each piece of equipment in the rack; wherein the one-way retrieval time is the maximum value of the horizontal and vertical time; determining the turnover time of the equipment based on the one-way retrieval time and the preset turnover frequency of the equipment; and determining the total time based on the turnover time of all equipment in the rack.

[0029] Optionally, for automated stacker cranes equipped with dual-axis linkage, their horizontal and vertical axes can operate simultaneously. The one-way travel time from the start and end points of the stacker crane to any storage location depends on the maximum of the horizontal and vertical travel times. Therefore, in this scenario, based on the horizontal and vertical operating speeds of the stacker crane, and the horizontal and vertical coordinates of the storage location assigned to each device, the horizontal and vertical travel times are calculated separately, and the maximum of the two is taken as the one-way retrieval time. This calculation method accurately maps the retrieval time characteristics of stacker cranes operating simultaneously with dual-axis linkage.

[0030] For single-axis material handling machines (such as purely horizontal conveyors or those with only vertical lifting capabilities), horizontal and vertical movements cannot occur simultaneously, and the total running time is the sum of the times spent in each direction. In such scenarios, Euclidean distance or Manhattan distance is used to calculate the one-way pickup time to suit different material handling machines. The one-way pickup time can be determined by taking the square root of the sum of the squares of the horizontal and vertical times (Euclidean distance) or by directly summing them (Manhattan distance).

[0031] Multiply the single-trip retrieval time of each device by 2 to obtain the time for the handling machine (stacking crane) to complete one round trip retrieval. Multiply the time for completing one round trip retrieval by the expected turnover frequency of the device within the evaluation period to obtain the turnover time of the device within the evaluation period. Sum the turnover times of all devices in the rack to obtain the total time of the candidate storage location allocation scheme, as shown in the formulas (1-1)-(1-2) below. This total time serves as the evaluation of the storage location allocation scheme in terms of inbound and outbound efficiency; the smaller the total time, the higher the inbound and outbound efficiency.

[0032] (1-1); (1-2); in, Indicates storage location Physical coordinates; The horizontal travel speed of the stacker crane The vertical running speed of the stacker crane; This represents the one-way pickup time for the equipment corresponding to storage location k. The preset turnover frequency of the equipment can be predicted through time series analysis based on historical outbound data; For Boolean decision variables, when the equipment Assigned to storage location hour ,otherwise And for any device satisfy . The total time taken for candidate warehouse location allocation scheme R (round trip pickup, one way × 2; unit: seconds / evaluation period).

[0033] By taking the maximum value between the horizontal and vertical time consumption to calculate the single-trip picking time, the mechanical motion characteristics of the automated stacker crane's dual-axis linkage and simultaneous operation are accurately mapped. This avoids the problem of artificially inflated time caused by the traditional method of simply adding the time consumption in each direction, making the efficiency assessment closer to the actual physical operation law of the equipment and providing an objective efficiency evaluation benchmark for warehouse location allocation schemes.

[0034] S130. Determine stability based on equipment weight distribution and rack geometry.

[0035] Based on the weight of all equipment in the shelving (including stored and unassigned equipment) and the spatial location of the assigned storage locations, combined with the shelving's geometric dimensions, a quantitative index characterizing the overall stability of the shelving is calculated. Stability comprehensively considers both the vertical center of gravity height and the horizontal center of gravity offset; a smaller value indicates better shelving stability, thus evaluating the performance of the storage location allocation scheme in terms of shelving mechanical safety.

[0036] Optionally, stability is determined based on equipment weight distribution and shelf geometry parameters, including: determining a normalized vertical center of gravity value based on the weight of each piece of equipment in the shelf, the vertical coordinate of the storage location occupied by each piece of equipment, and the maximum height of the shelf; determining a normalized horizontal offset value based on the weight of each piece of equipment in the shelf, the offset distance of the storage location allocated to each piece of equipment relative to the horizontal center line of the shelf, and the maximum half-width of the shelf; determining vertical and horizontal weights based on the maximum height and maximum half-width of the shelf; and determining stability by weighted summation of the normalized vertical center of gravity value and the normalized horizontal offset value based on the vertical and horizontal weights.

[0037] Determine the maximum height and maximum width of the shelving. Let the total number of shelves be L, and the height of each shelf be h. Then the maximum height of the shelving... =L×h; Let the total number of columns of the shelf be C, and the width of each column be w, then the maximum width of the shelf is... =C×w, the maximum half width of the shelf is / 2.

[0038] The normalized vertical center of gravity (COP) value measures the relative position of the actual physical center of gravity of the shelving unit when it is fully loaded in the vertical direction. For example, the COP is normalized to the range of 0 to 1 by summing the products of the weight of all equipment and the vertical coordinate of the storage location assigned to each equipment, and then dividing by the product of the maximum height of the shelving unit and the total weight of all equipment. This is illustrated in formula (2-1). A COP value closer to 0 indicates a lower center of gravity and greater vertical stability of the shelving unit.

[0039] The normalized horizontal offset value is used to measure the degree of deviation of the actual center of gravity of the rack relative to the horizontal center line of the rack when it is fully loaded. For example, the product of the weight of all equipment and the offset distance of its assigned storage location relative to the horizontal center line of the rack (i.e., the absolute value of the difference between the horizontal coordinate of the storage location and the horizontal center line) is summed and divided by the product of the maximum half width of the rack and the total weight of all equipment, thereby normalizing the horizontal offset to the range of 0 to 1. The specific formula is shown in formula (2-2) below. The closer the normalized horizontal offset value is to 0, the more centered the center of gravity is, and the more stable the rack is in the horizontal direction.

[0040] Since racks with different aspect ratios exhibit varying sensitivities to vertical and horizontal stability, this embodiment employs a weighting method objectively derived from the rack's geometric parameters to determine vertical and horizontal weights without manual intervention. For example, the vertical weight equals the rack's maximum height divided by the sum of its maximum height and maximum half-width, while the horizontal weight equals 1 minus the vertical weight. This is illustrated in formulas (2-3). For tall and narrow racks, a larger vertical weight is assigned, resulting in a higher evaluation weight for vertical stability; conversely, for wide and short racks, a larger horizontal weight is assigned, resulting in a higher evaluation weight for horizontal stability. This adaptive weighting mechanism allows this embodiment to be applied to different types of automated warehouses without requiring parameter recalibration for each type of rack.

[0041] Based on the vertical and horizontal weights, the normalized vertical centroid value and the normalized horizontal offset value are weighted and summed to obtain the stability of the candidate storage location allocation scheme R, as shown in formula (2-4) below. The stability is a dimensionless pure numerical value, ranging from 0 to 1. The smaller the value, the better the overall stability of the shelving.

[0042] (2-1); (2-2); (2-3); (2-4); in, Indicates the weight of device i; The vertical coordinates of the storage location assigned to the equipment; For Boolean decision variables, when the equipment Assigned to storage location hour ,otherwise And for any device satisfy ; This is the maximum height of the shelf; This is the maximum width of the shelf.

[0043] By calculating the normalized vertical centroid value and the normalized horizontal offset value separately, and objectively deriving the weights of the vertical and horizontal directions based on the height-to-width ratio of the shelving itself, a two-way comprehensive evaluation of shelving stability is achieved. This mechanism does not require manual setting of weight parameters and can adapt to different shelving types with different height-to-width ratios: tall and narrow shelving automatically emphasizes vertical stability, while wide and short shelving automatically emphasizes horizontal stability, thereby improving the cross-scenario versatility and objectivity of the evaluation method.

[0044] S140. Based on the allocation and correspondence between equipment and storage locations, determine the separation penalty value according to the distance between each storage location and the equipment correlation matrix.

[0045] The separation penalty value quantifies the degree of dispersion of related equipment within the shelving space. Data center IT equipment exhibits strong clustering characteristics during outbound shipments; for example, servers and network switches are often retrieved simultaneously. If these related devices are stored in geographically dispersed locations, stacker cranes must frequently make long-distance round trips across areas when performing complete outbound shipment tasks, significantly reducing outbound efficiency. The separation penalty value is measured by calculating the spatial distance between related devices. A smaller separation penalty value indicates a more concentrated distribution of related devices, which is more conducive to improving the efficiency of complete outbound operations.

[0046] S150. Determine the time difference between the time taken by the first-entry equipment to retrieve goods and the time taken by the subsequent-entry equipment within each smallest inventory unit, and determine the first-in-first-out violation penalty value of the candidate storage location allocation scheme based on the time difference.

[0047] The FIFO (First-In, First-Out) violation penalty value measures the degree to which a candidate storage location allocation scheme violates the FIFO management rule. In data center warehousing, the FIFO rule only applies to similar equipment within the same smallest unit of inventory, aiming to prevent some equipment from exceeding its depreciation period due to prolonged storage. If older equipment that entered the warehouse earlier is assigned to a remote storage location with longer retrieval times, while newer equipment that entered later is assigned to a nearby storage location with shorter retrieval times, it will make it more difficult to retrieve the older equipment in a timely manner, violating the FIFO logic. The FIFO violation penalty value is measured by summing the differences in retrieval times for all such violations. A smaller FIFO violation penalty value indicates that the candidate storage location allocation scheme better complies with FIFO management requirements and is more conducive to reducing the risk of equipment exceeding its depreciation period.

[0048] S160. Based on the total time consumption, stability, separation penalty value, and first-in-first-out violation penalty value, generate an evaluation vector to evaluate the candidate storage location allocation scheme.

[0049] The total time, stability, separation penalty value, and FIFO violation penalty value are combined into a four-dimensional evaluation vector in a preset order. The total time reflects the candidate storage location allocation scheme's performance in terms of inbound / outbound efficiency; stability reflects its performance in terms of shelf mechanical safety; separation penalty value reflects its performance in terms of the spatial aggregation of related equipment; and FIFO violation penalty value reflects its performance in terms of asset management compliance. The four dimensions are independent of each other, retaining their original dimensions and physical meaning, and are not subject to cross-dimensional weighted fusion.

[0050] This evaluation vector serves as a quantitative benchmark describing the objective performance of a warehouse location allocation scheme across multiple physical dimensions, and is output to the decision-making module of the warehouse control system. Using a multi-objective optimization algorithm (such as the Pareto non-dominated sorting genetic algorithm), different candidate warehouse location allocation schemes can be compared and screened based on this evaluation vector to find the optimal scheme that performs well across all dimensions, ultimately outputting an executable warehouse location allocation command. By eliminating subjective weighting, this provides the warehouse control system with an objective, comprehensive, and comparable basis for scheme evaluation.

[0051] This invention achieves objective quantitative evaluation of candidate warehouse location allocation schemes across multiple dimensions by isomorphically grouping equipment according to the smallest inventory unit, constructing a correlation matrix based on historical work orders, and generating a four-dimensional evaluation vector that includes total time consumption, stability, separation penalty value, and FIFO violation penalty value. It abandons the traditional method of subjective weighting across dimensions, avoiding the misleading optimization direction caused by weight distortion. Simultaneously, by introducing equipment matching correlation and grouping FIFO constraints, the evaluation results better align with the actual business needs of strong matching of data center IT equipment and prevention of expired similar materials, effectively improving the overall quality and engineering practicality of warehouse location allocation schemes.

[0052] Figure 2 This is a flowchart of a data center storage location allocation method provided by an embodiment of the present invention. This embodiment is an optimization and improvement based on the above embodiment. Figure 2 As shown, the method includes: S210. Determine the smallest inventory unit to which each device belongs in the data center, and construct a device correlation matrix based on the number of times each smallest inventory unit co-occurs in historical outbound work orders.

[0053] S220. For any candidate warehouse location allocation scheme, determine the total time based on the picking time of each warehouse location and the preset turnover frequency of each device.

[0054] S230. Determine stability based on equipment weight distribution and rack geometry.

[0055] S240. Based on the equipment correlation matrix, determine the equipment correlation between any pair of equipment in the shelf; based on the horizontal and vertical coordinates of the two storage locations assigned to the equipment pair, determine the physical separation distance between the two storage locations.

[0056] The equipment correlation degree between two devices belonging to the same smallest inventory unit is 1.

[0057] S250. The product of the equipment correlation degree and the physical separation distance is used as the separation value of the equipment pair.

[0058] S260. Sum the separation values ​​of all equipment pairs in the shelf to obtain the separation penalty value.

[0059] The equipment correlation matrix is ​​constructed based on the co-occurrence frequency of each minimum inventory unit in historical outbound work orders. The element values ​​reflect the probability of matching outbound shipments between the minimum inventory units to which two corresponding equipment belong. For any two pieces of equipment, the minimum inventory unit to which the equipment belongs is determined, and the correlation value corresponding to the minimum inventory units of the two pieces of equipment is read from the equipment correlation matrix as the equipment correlation degree. Specifically, when two pieces of equipment belong to the same minimum inventory unit, the equipment correlation degree between them is set to 1. This is because similar equipment of the same model and specification is considered completely interchangeable in business and has the strongest substitution or matching relationship. Simultaneously, based on the horizontal and vertical coordinates of the two storage locations allocated to the equipment in the candidate storage location allocation scheme, the Manhattan physical distance between the two storage locations is calculated, which is the sum of the absolute values ​​of the horizontal coordinate differences and the vertical coordinate differences, and is taken as the physical separation distance.

[0060] Multiplying the device correlation degree by the physical separation distance yields the separation value for the device pair. The separation value comprehensively reflects the combined impact of the device pair on both the strength of business correlation and the degree of spatial separation: the stronger the business correlation and the greater the spatial distance, the larger the separation value.

[0061] By iterating through all possible equipment pairs on the shelf and summing the separation values ​​of all equipment pairs, the separation penalty value of the candidate storage location allocation scheme is obtained. This is illustrated in formulas (3-1) and (3-2) below. The separation penalty value quantifies the overall spatial dispersion of related equipment throughout the entire warehouse and is an important indicator for evaluating the performance of the storage location allocation scheme in terms of complete set outbound operations.

[0062] (3-1); (3-2); in, Indicates storage location Physical coordinates Indicates storage location Physical coordinates; This represents the physical separation distance between the two assigned storage locations k and l; This indicates the degree of device association between devices i and j; , For Boolean decision variables, when the equipment Assigned to storage location hour ,otherwise And for any device satisfy .

[0063] By calculating the separation penalty value by multiplying the equipment correlation degree by the physical distance between storage locations and then summing the results, it is clarified that the equipment correlation degree within the same smallest storage unit is set to 1, ensuring that similar equipment is preferentially clustered in space. The Manhattan distance, which is continuously differentiable, is used as the penalty gradient, providing a smooth search direction for the upper-level optimization algorithm. By accumulating all equipment pairs, the degree of dispersion of related equipment in the overall warehouse space is comprehensively quantified, effectively guiding the allocation scheme to place strongly matching equipment in nearby storage locations, reducing the cross-regional round trips when stacker cranes are dispatching complete sets of equipment.

[0064] S270. Determine the time difference between the time taken by the first-entry equipment to retrieve goods and the time taken by the subsequent-entry equipment within each smallest inventory unit, and determine the first-in-first-out violation penalty value of the candidate storage location allocation scheme based on the time difference.

[0065] Optionally, within each minimum inventory unit, determine the time difference between the time taken by the first-entry device to retrieve goods and the time taken by the second-entry device to retrieve goods. Based on this time difference, determine the first-in-first-out (FIFO) violation penalty value for the candidate storage location allocation scheme. This includes: sorting all devices belonging to the same minimum inventory unit on the shelf according to their entry timestamps; traversing the sorted device sequence; if the retrieval time of the storage location allocated to the first-entry device is greater than the retrieval time of the storage location allocated to the second-entry device, calculating the time difference between the two devices; and summing the time differences among all devices on the shelf to obtain the FIFO violation penalty value for the candidate storage location allocation scheme.

[0066] The equipment belonging to the same smallest unit of inventory on the shelf is sorted according to its entry timestamp. The entry timestamp of the stored equipment is the time recorded when it was actually entered into the warehouse, and the entry timestamp of the equipment to be allocated is its planned entry time or current scheduling time. This determines the order of the equipment within the same smallest unit of inventory, with the equipment that entered earlier entering the warehouse first and the equipment that entered later entering the warehouse last.

[0067] Traverse the sorted sequence of devices and compare the retrieval time of the storage locations assigned to the devices that entered the warehouse earlier with that of the devices that entered the warehouse later. If the retrieval time of the storage location assigned to the device that entered the warehouse earlier is greater than that of the storage location assigned to the device that entered the warehouse later, that is, the older device was assigned to a storage location that is slower than the newer device, then calculate the time difference between the devices that entered the warehouse earlier and the devices that entered the warehouse later.

[0068] The first-in, first-out (FIFO) violation penalty value for the candidate location allocation scheme is obtained by summing the time differences between the locations allocated to the earlier-entering equipment and the locations allocated to the later-entering equipment, all within the same minimum inventory unit. Optionally, all devices belonging to the same minimum inventory unit in the shelf are sorted according to their inbound timestamps. Then, adjacent device pairs are compared sequentially in the sorted order. If the retrieval time of the device that entered the warehouse earlier is greater than that of the device that entered the warehouse later, the time difference between the two is accumulated. Using the adjacent comparison method can optimize the computational complexity to a linear relationship with the number of devices in the same minimum inventory unit, avoiding the quadratic computational overhead of traversing all device pairs.

[0069] The first-in-first-out violation penalty value adopts a continuous time difference rather than a discrete violation judgment, which makes the evaluation vector respond to the scheme perturbation with continuous and smooth characteristics. This provides effective gradient information for the upper-level multi-objective optimization algorithm and helps to accelerate the convergence of the optimization process.

[0070] By sorting items by their arrival timestamp within the same minimum inventory unit and accumulating the time difference when the retrieval time of the earlier-arriving device is greater than that of the later-arriving device, the first-in, first-out (FIFO) rule is limited to the same minimum inventory unit. This avoids business logic errors caused by global queuing across minimum inventory units in traditional methods, aligning with the actual needs of preventing overdue management of similar devices in data centers. The use of continuous time difference values ​​instead of discrete Boolean judgments ensures that the evaluation vector exhibits smooth and differentiable response to scheme disturbances, providing effective gradient information for upper-level multi-objective optimization algorithms. Furthermore, by employing a linear scanning and adjacent comparison method, the computational complexity is controlled to be linear with the number of grouped devices, maintaining high computational performance even in large-scale warehousing scenarios.

[0071] S280. Based on the total time consumption, stability, separation penalty value, and first-in-first-out violation penalty value, generate an evaluation vector to assess the candidate storage location allocation scheme.

[0072] The total time, stability, separation penalty value, and first-in-first-out violation penalty value are combined into a four-dimensional evaluation vector according to a preset order: Dimensions of each dimension: (seconds / evaluation period) (dimensionless, ), (millimeters) (seconds). The four dimensions are independent of each other, and there is no cross-dimensional weighting, thus avoiding the evaluation distortion problem caused by subjective setting of weight coefficients in traditional methods. This ensures that the evaluation results truly reflect the physical constraints of the warehousing system and provides an objective and comparable basis for selecting solutions for upper-level optimization algorithms.

[0073] This invention constructs a device correlation matrix based on historical outbound work orders and uses the sum of the products of device correlation and physical distance between storage locations as a separation penalty value. This achieves an objective quantitative evaluation of the spatial aggregation degree of candidate storage location allocation schemes. By utilizing the correlation matrix to reflect the actual business characteristics of strongly coupled outbound data center IT equipment, the allocation scheme guides the placement of related devices with high co-occurrence probability in nearby storage locations. This reduces the cross-regional round-trip travel of stacker cranes when performing complete outbound tasks, effectively improving the overall outbound efficiency of the warehousing system.

[0074] Figure 3 This is a schematic diagram of the structure of a data center storage location allocation device provided in an embodiment of the present invention. Figure 3 As shown, the device includes: The association matrix determination module 310 is used to determine the smallest inventory unit to which each device belongs in the data center, and to construct the device association matrix based on the number of times each smallest inventory unit co-occurs in the historical outbound work orders; The total time determination module 320 is used to determine the total time for any candidate warehouse location allocation scheme based on the picking time of each warehouse location and the preset turnover frequency of each device. Stability determination module 330 is used to determine stability based on equipment weight distribution and shelf geometry parameters; The separation penalty determination module 340 is used to determine the separation penalty value based on the allocation correspondence between equipment and storage location, the distance between each storage location, and the equipment correlation matrix. The violation penalty determination module 350 is used to determine the time difference between the time taken by the first-entry equipment to retrieve goods and the time taken by the subsequent-entry equipment within each smallest inventory unit, and to determine the first-in-first-out violation penalty value of the candidate storage location allocation scheme based on the time difference. The evaluation vector generation module 360 ​​is used to generate an evaluation vector based on the total time consumption, stability, separation penalty value and first-in-first-out violation penalty value to evaluate the candidate storage location allocation scheme.

[0075] The data center storage location allocation device provided in this embodiment of the invention can execute the data center storage location allocation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0076] Optionally, the separation penalty determination module is specifically used to determine the equipment correlation degree between any pair of equipment in the shelf according to the equipment correlation degree matrix; determine the physical separation distance between the two storage locations assigned to the equipment pair according to the horizontal and vertical coordinates; take the product of the equipment correlation degree and the physical separation distance as the separation value of the equipment pair; and sum the separation values ​​of all equipment pairs in the shelf to obtain the separation penalty value.

[0077] Optionally, the stability determination module is specifically used to determine the normalized vertical center of gravity value based on the weight of each piece of equipment in the shelf, the vertical coordinate of the storage location occupied by each piece of equipment, and the maximum height of the shelf; determine the normalized horizontal offset value based on the weight of each piece of equipment in the shelf, the offset distance of the storage location allocated to each piece of equipment relative to the horizontal center line of the shelf, and the maximum half width of the shelf; determine the vertical weight and horizontal weight based on the maximum height and maximum half width of the shelf; and determine the stability by weighted summation of the normalized vertical center of gravity value and the normalized horizontal offset value based on the vertical weight and horizontal weight.

[0078] Optionally, the violation penalty determination module is specifically used to sort all equipment belonging to the same smallest storage unit in the shelf according to the entry timestamp; traverse the sorted equipment sequence, and if the retrieval time of the storage location allocated to the equipment that entered first is greater than the retrieval time of the storage location allocated to the equipment that entered later, calculate the time difference between the equipment that entered first and the equipment that entered later; and sum up the time differences between all equipment in the shelf to obtain the first-in-first-out violation penalty value of the candidate storage location allocation scheme.

[0079] Optionally, the total time determination module is specifically used to determine the one-way retrieval time from the start and end points of the stacker crane to the storage location based on the horizontal and vertical running speeds of the stacker crane and the horizontal and vertical coordinates of the storage location allocated to each device in the rack; wherein the one-way retrieval time is the maximum value of the horizontal and vertical running times; the turnaround time of the device is determined based on the one-way retrieval time and the preset turnover frequency of the device; and the total time is determined based on the turnaround time of all devices in the rack.

[0080] Optionally, the correlation matrix determination module is specifically used to determine the number of times any pair of minimum inventory units co-occurs in the same batch of historical outbound work orders, and the total number of historical outbounds for each minimum inventory unit; determine the symmetric correlation degree between each pair of minimum inventory units based on the co-occurrence count and the total number of historical outbounds; and construct an equipment correlation degree matrix based on the symmetric correlation degree between all minimum inventory units.

[0081] The data center storage location allocation device further described can also execute the data center storage location allocation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0082] According to embodiments of the present invention, the present invention also provides an electronic device, a readable storage medium, and a computer program product.

[0083] Figure 4 A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0084] like Figure 4As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory 42 or a random access memory 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 42 or loaded from storage unit 48 into the random access memory 43. The random access memory 43 may also store various programs and data required for the operation of the electronic device 40. The processor 41, read-only memory 42, and random access memory 43 are interconnected via a bus 44. An input / output interface 45 is also connected to the bus 44.

[0085] Multiple components in electronic device 40 are connected to input / output interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0086] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as data center storage location allocation methods.

[0087] In some embodiments, the data center storage location allocation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via read-only memory 42 and / or communication unit 49. When the computer program is loaded into random access memory 43 and executed by processor 41, one or more steps of the data center storage location allocation method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the data center storage location allocation method by any other suitable means (e.g., by means of firmware).

[0088] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), systems-on-a-chip (SoCs), payload programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0089] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0090] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0091] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube, liquid crystal display, or monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0092] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0093] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product within the cloud computing service system to address the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.

[0094] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0095] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A data center storage location allocation method, characterized in that, The method includes: Determine the smallest inventory unit to which each device belongs within the data center, and construct a device correlation matrix based on the co-occurrence frequency among the smallest inventory units in historical outbound work orders; For any candidate storage location allocation scheme, the total time is determined based on the retrieval time of each storage location and the preset turnover frequency of each device; stability is determined based on the weight distribution of the devices and the geometric parameters of the shelves; separation penalty value is determined based on the allocation correspondence between devices and storage locations, the distance between each storage location, and the device correlation matrix; within each minimum inventory unit, the time difference between the retrieval time of the first-in-first-out device and the retrieval time of the later-in-first-out device is determined, and the first-in-first-out violation penalty value of the candidate storage location allocation scheme is determined based on the time difference. An evaluation vector is generated based on the total time consumption, stability, separation penalty value, and first-in-first-out violation penalty value to evaluate the candidate storage location allocation scheme.

2. The method according to claim 1, characterized in that, The step of determining the separation penalty value based on the allocation correspondence between equipment and storage locations, the distance between each storage location, and the equipment correlation matrix includes: Based on the equipment correlation matrix, determine the equipment correlation between any pair of equipment in the shelf; The physical separation distance between the two storage locations is determined based on the horizontal and vertical coordinates of the two storage locations assigned by the device. The product of the device correlation degree and the physical separation distance is used as the separation value of the device pair; The separation penalty value is obtained by summing the separation values ​​of all equipment pairs in the shelf.

3. The method according to claim 1, characterized in that, The determination of stability based on equipment weight distribution and rack geometry includes: The normalized vertical center of gravity value is determined based on the weight of each piece of equipment in the shelf, the vertical coordinate of the storage space occupied by each piece of equipment, and the maximum height of the shelf. The normalized horizontal offset value is determined based on the weight of each piece of equipment in the shelving, the offset distance of the storage location allocated to each piece of equipment relative to the horizontal center line of the shelving, and the maximum half width of the shelving. Determine the vertical and horizontal weights based on the maximum height and maximum half width of the shelf; Stability is determined by weighting and summing the normalized vertical centroid value and the normalized horizontal offset value according to the vertical and horizontal weights.

4. The method according to claim 1, characterized in that, The step of determining the time difference between the time taken by the first receiving equipment and the time taken by the subsequent receiving equipment within each minimum inventory unit, and determining the first-in-first-out (FIFO) violation penalty value for the candidate storage location allocation scheme based on the time difference, includes: Sort all equipment belonging to the same smallest storage unit in the shelf according to the inbound timestamp; Traverse the sorted sequence of equipment. If the retrieval time of the storage location allocated to the equipment that enters the warehouse first is greater than the retrieval time of the storage location allocated to the equipment that enters the warehouse later, calculate the time difference between the equipment that enters the warehouse first and the equipment that enters the warehouse later. The time difference between all devices in the shelf is summed to obtain the first-in-first-out violation penalty value of the candidate storage location allocation scheme.

5. The method according to claim 1, characterized in that, The determination of the total time based on the retrieval time of each storage location and the preset turnover frequency of each piece of equipment includes: Based on the horizontal and vertical running speeds of the stacker crane and the horizontal and vertical coordinates of the storage location assigned to each device in the rack, the one-way retrieval time from the start and end points of the stacker crane to the storage location is determined; wherein, the one-way retrieval time is the maximum value of the horizontal and vertical retrieval times. The turnover time of the equipment is determined based on the single-trip pickup time and the preset turnover frequency of the equipment, and the total time is determined based on the turnover time of all equipment in the shelf.

6. The method according to claim 1, characterized in that, The construction of the equipment correlation matrix based on the co-occurrence frequency among the smallest inventory units in historical outbound work orders includes: Determine the number of times any pair of minimum inventory units co-occurs in the same batch of historical outbound work orders, and the total number of historical outbounds for each minimum inventory unit; Based on the co-occurrence frequency and the total historical outbound frequency, determine the degree of symmetry between each pair of minimum inventory units; Construct a device correlation matrix based on the symmetric correlation between all the smallest inventory units.

7. A data center storage location allocation device, characterized in that, The device includes: The association matrix determination module is used to determine the smallest inventory unit to which each device belongs in the data center, and to construct a device association matrix based on the number of times each smallest inventory unit co-occurs in historical outbound work orders; The total time determination module is used to determine the total time for any candidate warehouse location allocation scheme based on the picking time of each warehouse location and the preset turnover frequency of each device. The stability determination module is used to determine stability based on the equipment weight distribution and shelf geometry parameters. The separation penalty determination module is used to determine the separation penalty value based on the allocation correspondence between equipment and storage location, the distance between each storage location, and the equipment correlation matrix. The violation penalty determination module is used to determine the time difference between the time taken by the first-entry equipment to retrieve goods and the time taken by the second-entry equipment to retrieve goods within each smallest inventory unit, and to determine the first-in-first-out violation penalty value of the candidate storage location allocation scheme based on the time difference. The evaluation vector generation module is used to generate an evaluation vector based on the total time consumption, stability, separation penalty value and first-in-first-out violation penalty value to evaluate the candidate storage location allocation scheme.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the data center storage location allocation method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the data center storage location allocation method according to any one of claims 1-6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the data center storage location allocation method according to any one of claims 1-6.

Images