Efficient intelligent storage location distribution method
By constructing mathematical models and using particle swarm optimization algorithms, the problem of traditional warehouse layout not taking into account the frequencies of goods is solved, efficient intelligent warehouse location allocation is achieved, storage and access efficiency and dynamic adaptability are improved, and equipment energy consumption is reduced.
Patent Information
- Application Number
- CN202510393367.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-31
- Publication Date
- 2025-07-01
AI Technical Summary
The traditional automated warehouse cargo storage layout does not fully consider factors such as the frequency of goods entering and exiting the warehouse, resulting in an increase in cargo handling time, high equipment energy consumption, and lack of an adaptive adjustment mechanism, making it difficult to quickly respond to changes in cargo types and quantities.
The data acquisition module is used to comprehensively collect warehouse structural parameters and cargo raw data, normalize the data through the data preprocessing module, build a mathematical model to optimize the cargo storage layout, and use the particle swarm optimization algorithm execution module to find the optimal cargo storage layout solution through iterative operations.
It improves cargo storage and access efficiency, reduces equipment energy consumption, enhances dynamic adaptability, and can quickly find the optimal cargo storage layout plan to meet the requirements for flexibility and timeliness in efficient logistics operations.
Smart Images
Figure CN120235558A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of logistics warehousing, and particularly relates to an efficient intelligent storage location allocation method. Background Art
[0002] In traditional automated warehouse goods storage layouts, fixed partitioning or random storage modes are mostly adopted. Their designs are usually based on empirical rules or simple classification principles, such as dividing storage areas according to goods categories, specifications, or incoming time. Such layouts do not systematically integrate key parameters such as the frequency of goods in and out, weight and volume attributes, and handling path optimization, resulting in high-frequency in-and-out goods being mechanically allocated to storage locations far from the entrance / exit or the core area of equipment operation. This static space allocation method forces handling equipment (such as stackers, AGVs) to frequently perform long-distance round-trip transportation, significantly increasing the single operation cycle and the equipment empty running rate. In addition, there is a lack of collaborative optimization between storage strategies and equipment scheduling, and the goods access path cannot be dynamically planned based on real-time data, further exacerbating the imbalance between equipment energy consumption and operation efficiency, forming a vicious cycle of "inefficient handling of high-frequency goods - high-load operation of equipment - rising energy consumption costs".
[0003] The existing technologies have the following core defects: First, the efficiency of goods access and storage is low. The traditional automated warehouse goods storage layout does not fully consider factors such as the frequency of goods in and out, etc., making high-frequency in-and-out goods far from the entrance / exit, resulting in a significant increase in the goods handling time, reducing the overall access and storage efficiency, and affecting the progress of warehouse operations; Second, the dimension of energy consumption control is single. Due to the unreasonable goods layout, the equipment needs to run longer distances to handle goods, increasing energy consumption and operation costs, which is not conducive to energy conservation and emission reduction; Third, the lack of an adaptive adjustment mechanism. Facing the dynamic changes in the types and quantities of goods, the traditional warehouse layout is difficult to quickly respond and make effective adjustments, unable to meet the requirements of flexibility and timeliness in efficient logistics operations, and restricting the warehouse's ability to handle complex business scenarios. For example, when the demand for seasonal goods surges, the traditional warehouse needs manual intervention to re-plan the storage locations, which is time-consuming and has low accuracy, and cannot meet the requirements of modern logistics for real-time and flexible operation. These defects jointly restrict the comprehensive performance improvement of the warehousing system in terms of efficiency, cost, and sustainability, and there is an urgent need to achieve breakthroughs through technological innovation. Summary of the Invention
[0004] The problem to be solved by the present invention is to provide an efficient intelligent storage location allocation method, which can accurately evaluate and quickly find the optimal goods storage layout plan, improve the goods access and storage efficiency, reduce equipment energy consumption, enhance the dynamic adaptation ability, and meet the requirements of flexibility and timeliness in efficient logistics operations.
[0005] Solve a series of problems caused by the unreasonable goods layout in traditional automated warehouses.
[0006] To solve the above technical problems, the technical solution adopted by the present invention is as follows: An efficient and intelligent storage location allocation method, characterized by comprising the following steps: (1) Using a data acquisition module to comprehensively collect the warehouse structure parameters and the original data of the goods: The data acquisition module includes a first interface docked with the order management system and a second interface for measuring or reading data for collecting the warehouse structure parameters. Through the first interface, various types of sales data of the goods are captured in real time from the order management system, and the frequency of goods in and out of the warehouse is counted; through the second interface, the warehouse structure parameters are read from the warehouse design drawings, and these original data are collected and summarized and transmitted to the data preprocessing module for preprocessing to make these original data suitable for subsequent operations; (2) Using a data preprocessing module to perform normalization processing on the collected original data: The data preprocessing module includes a data conversion algorithm module and a data verification module; the data conversion algorithm module is used to convert data with different dimensions into a form that can be uniformly operated, and the data verification module is used to perform accuracy verification on the processed data, and then the processed data is transmitted to the mathematical model construction module; (3) Using a mathematical model construction module to construct a mathematical model for the optimization objective and constraints: The mathematical model construction module includes an objective function construction unit, a constraint condition setting unit, and a model calculation unit; the objective function construction unit constructs a function with the shortest average access time of the goods as the objective, and the constraint condition setting unit sets constraint conditions according to that each storage location can only store one type of goods, the volume and weight limits of the goods, and the model calculation unit substitutes the data from the data preprocessing module into the mathematical model; (4) Using a particle swarm optimization algorithm execution module to find the optimal goods storage layout plan through iterative operations: (4-1) Initializing the particle swarm algorithm; (4-2) Calculating the fitness value of the particles; (4-3) Updating the particle velocity and position; (4-4) Controlling the number of iterations of the particles to determine the optimal goods storage layout plan; (5) According to the optimal goods storage layout plan obtained in step (4), using a layout adjustment execution module to adjust the goods storage positions and monitor the operation effect.
[0007] In the preferred solution, in the step (1), the original data of the goods includes the types of goods, the volume of the goods, the weight of the goods, and the frequency of goods in and out of the warehouse; the warehouse structure parameters include the warehouse area, the number of shelves, the number of columns of the shelves, the number of aisles, the size of the storage location, the maximum volume of the warehouse storage location, and the maximum weight that the warehouse can bear.
[0008] In a further preferred embodiment, in step (2), the data conversion algorithm module divides the volume of the goods by the maximum volume of the storage location in the warehouse and divides the weight of the goods by the maximum weight that the warehouse can bear, so as to convert data with different dimensions into a form that can be uniformly calculated, and normalize the collected data such as the volume and weight of the goods, so as to make the data meet the requirements of subsequent mathematical models and algorithm operations, and provide high-quality data support for the subsequent process.
[0009] In a further preferred embodiment, in step (3), the constraint conditions include that each storage location can only store one kind of goods, the volume of the goods does not exceed the volume of the storage location, and the weight of the goods does not exceed the bearing capacity of the shelf. When constructing the mathematical model, these constraint conditions need to be considered to ensure that the mathematical model can accurately evaluate the rationality and practicability of the goods storage layout plan.
[0010] In a further preferred embodiment, in step (3), the formula is used as the mathematical model, and the average access time of the goods is calculated through the formula to provide a quantitative basis for the particle swarm optimization algorithm execution module to evaluate the layout plan; in the formula, , represents the average access time of the goods; represents the inbound and outbound frequency of the goods ; represents the access time of the th kind of goods at a certain storage location; represents the sum of multiplying the inbound and outbound frequency of each kind of goods by the time to access the goods from the storage location and then performing accumulation, , is the number of goods types; , represents the sum of the inbound and outbound frequencies of all goods. In step (3) above, the average access time of the goods is calculated through the formula to obtain the value, which can comprehensively reflect the average time-consuming situation of the warehouse to access goods under the current storage layout, and provide a quantitative basis for evaluating the advantages and disadvantages of different goods storage layout plans.
[0011] In a further preferred embodiment, the particle swarm optimization algorithm execution module in step (4) includes a particle initialization unit, a fitness calculation unit, a velocity / position update unit, and an iteration control unit; Step (4-1): The particle initialization unit randomly generates particles and initializes the starting positions and velocities of each particle according to the cargo and storage location information involved in the mathematical model in step (3). Each particle represents a cargo storage layout plan, and the distribution of cargo in each storage location is represented by a position vector. Step (4-2): The fitness calculation unit calculates the fitness value of the particle according to the mathematical model. Step (4-3): The velocity / position update unit updates the particle velocity and position through calculation based on the historical best position of the particle itself and the global best position. Step (4-4): The iteration control unit controls the number of iterations of the particle until the preset number of iterations or the fitness value convergence condition is met, finds the position vector corresponding to the optimal particle, which is the optimal cargo storage layout plan, and then transfers the optimal cargo storage layout plan to the layout adjustment execution module.
[0012] In the above step (4-1), each particle represents a cargo storage layout plan, the distribution of cargo in each storage location is represented by a position vector, and the starting position and velocity of the particle are initialized. The velocity of the particle determines the direction and degree of layout change. For example, the position vector [1, 3, 5, 2, 4] represents that the first type of cargo is stored in the first storage location, the second type of cargo is stored in the third storage location, the third type of cargo is stored in the fifth storage location, the fourth type of cargo is stored in the second storage location, and the fifth type of cargo is stored in the fourth storage location.
[0013] In a further preferred solution, in the above step (4-2), when calculating the fitness value, it is necessary to substitute the cargo storage layout plan corresponding to each particle into the mathematical model according to the established mathematical model to calculate the fitness value of each particle; the smaller the fitness value of the particle, the better the cargo storage layout plan.
[0014] In a further preferred solution, in the above step (4-3), when updating the particle position and velocity, it is necessary to use the velocity update formula based on the historical best position of the particle itself and the global best position of the entire particle swarm and the position update formula . By continuously updating the particle velocity and position according to these two formulas, the particle swarm continuously iterates and optimizes in the search space, gradually approaching the optimal cargo storage layout plan, so as to find the optimal cargo storage layout plan for the automated storage and retrieval system. In the velocity update formula: is the current velocity of particle in dimension , reflecting the current movement trend of particle in this dimension . is called the inertia weight, which serves to balance the global search and local search capabilities of the particle; a larger value helps the particle to conduct a more extensive global exploration in the search space, while a smaller value enables the particle to focus more on the local area for fine search; and are learning factors, mainly affects the degree to which the particle learns from its own historical best position while determines the intensity of the particle's learning towards the global best position ; by adjusting these two factors, the degree of dependence of the particle on its own experience and the group experience during the search process can be controlled; and are random numbers between 0 and 1, introducing randomness to the velocity update, so that the particle will not get trapped in a local optimal solution during the search process and can explore the search space more comprehensively; is the historical best position of the particle in dimension , that is, the best position the particle has reached during the previous search process; is the current position of the particle in dimension ; This term prompts the particle to move towards its own historical best position, reflecting the particle's learning from its own experience; This term drives the particle to move towards the global best position, reflecting the particle's learning from the best experience in the group; In the position update formula: represents the new position of the particle in dimension ; represents the current position of the particle in dimension ; represents the updated velocity of the particle in dimension .
[0015] In the above position update formula, the new position of the particle in dimension is determined by its current position Plus the updated speed Obtained
[0016] In a preferred embodiment, in step (5), the layout adjustment execution module includes an instruction sending unit, a handling sequence planning unit, and an operation monitoring unit; the instruction sending unit sends an instruction to adjust the storage location of goods to the automated equipment through the warehouse management system; the handling sequence planning unit reasonably arranges the handling sequence of goods to avoid equipment collision; the operation monitoring unit collects the goods access efficiency and equipment energy consumption of the adjusted warehouse, feeds back the goods access efficiency and equipment energy consumption of the warehouse to the particle swarm optimization algorithm execution module in step (4), and transfers the new warehouse operation data to the data collection module in step (1) for continuous optimization.
[0017] In a further preferred embodiment, in step (5), an instruction to adjust the storage location of goods is sent to the automated equipment through the warehouse management system, and the storage location of goods is adjusted according to the optimal goods storage layout plan. During the adjustment process, the goods that are far from the entrance and exit and have a low frequency of inbound and outbound are first transported to the new location, and then the positions of other goods are gradually adjusted. By this setting, equipment conflict and collision can be avoided, ensuring the efficiency and safety of the adjustment process. The above-mentioned automated equipment can be a stacker, a conveyor, etc.
[0018] In a more preferred embodiment, in step (5), after the layout adjustment is completed, the operation monitoring unit monitors and evaluates the operation of the warehouse, and compares the goods access efficiency and equipment energy consumption before and after the adjustment; if the operation effect of the goods storage location is not ideal, the algorithm parameters are further adjusted or re-optimized to continuously improve the efficiency of the goods storage layout of the automated stereoscopic warehouse.
[0019] Compared with the prior art, the present invention has the following advantages: (1) The present invention adopts data collection and preprocessing technology, docks with the order management system and adopts a unique normalization method to provide accurately adapted data for subsequent operations; (2) The present invention constructs a mathematical model with the goal of the shortest average access time and comprehensively considers various actual constraints. This mathematical model provides a quantitative basis for optimizing the goods storage layout. Compared with the traditional model, it can better fit the actual situation of the warehouse and accurately evaluate the layout plan; (3) The present invention innovatively applies the particle swarm optimization algorithm to the optimization of the storage layout of the automated warehouse. Each particle corresponds to a goods storage layout plan. By continuously iteratively updating the particle position and speed, the particle with the optimal objective function value is found, and the optimal goods storage layout plan is quickly found, breaking through the efficiency bottleneck of the traditional search algorithm; (4) The present invention realizes collaborative control by using a warehouse management system, seamlessly docks with automated equipment and plans the handling sequence, completes the adjustment of the storage locations of goods, and continuously optimizes the solution through monitoring and evaluation of actual operation data, so as to achieve efficient optimization of the storage layout of goods in an automated warehouse and ensure the efficient and safe adjustment of the layout; (5) The present invention establishes a continuous monitoring and optimization innovation mechanism, flexibly adjusts algorithm parameters or restarts the process according to the monitoring data to improve the layout efficiency, and makes up for the deficiency of the existing technology in lacking a mechanism to flexibly respond to business changes; (6) Improve the efficiency of goods storage and retrieval: By scientifically planning the storage locations of goods, fully considering factors such as the inbound and outbound frequencies of goods, the present invention places goods with frequent inbound and outbound in favorable positions close to the entrances and exits, thus significantly shortening the handling time of goods and remarkably improving the overall storage and retrieval efficiency of warehouse goods; (7) Reduce equipment energy consumption: Starting from optimizing the goods storage layout, the present invention reasonably arranges the goods storage points, reduces the running distance of automated equipment during the process of handling goods, and thus reduces the energy consumption of the equipment, achieves the goal of energy conservation and emission reduction, and reduces the warehousing operation cost; (8) Enhance dynamic adaptability: The present invention constructs an effective mechanism that can automatically and efficiently adjust the storage layout according to the real-time changing data of the types and quantities of goods, enhances the flexibility of the warehouse to cope with complex business scenarios, meets the strict requirements of efficient logistics operation for timeliness and adaptability, and improves the overall operation efficiency of the automated stereoscopic warehouse. Brief Description of the Drawings
[0020] Figure 1 is the flowchart of optimizing the storage layout of goods in the automated warehouse of the specific embodiment of the present invention; Figure 2 is the flowchart of the particle swarm optimization algorithm of the specific embodiment of the present invention; Figure 3 is the schematic diagram of the layout before optimization of the specific embodiment of the present invention; Figure 4 is the schematic diagram of initializing the particle swarm algorithm of the specific embodiment of the present invention; Figure 5 is the schematic diagram of calculating the fitness value of the specific embodiment of the present invention; Figure 6 is the schematic diagram of iteratively updating particles of the specific embodiment of the present invention; Figure 7 is the schematic diagram of determining the optimal layout of the specific embodiment of the present invention. Detailed Embodiments
[0021] The present invention will be specifically described below in conjunction with the drawings and specific embodiments.
[0022] Such as Figure 1-2As shown in the figure, the efficient intelligent storage location allocation method in this embodiment includes the following steps: (1) Use the data acquisition module to comprehensively collect the warehouse structure parameters and the original data of the goods; (2) Use the data preprocessing module to perform normalization processing on the collected original data; (3) Use the mathematical model construction module to construct a mathematical model for the optimization objective and constraints; (4) Use the particle swarm optimization algorithm execution module to find the optimal goods storage layout plan through iterative operations; (5) According to the optimal goods storage layout plan obtained in step (4), use the layout adjustment execution module to adjust the goods storage location and monitor the operation effect.
[0023] Next, take a specific example to implement steps (1)-(5) of the efficient intelligent storage location allocation method. The example is: find the optimal goods storage layout plan for 10 kinds of goods in a scenario with 20 storage locations to find the shortest average access time of the goods .
[0024] Step (1): Data acquisition a - Goods data acquisition: Obtained through the warehouse management system interface. The monthly average inbound and outbound frequencies of 10 kinds of goods are as shown in Table 1 below. Assume that the volume of each good is ≤1m³ and the weight of each good is ≤1000kg (meeting the storage location constraints).
[0025] Table 1: Goods (1-10) 1 2 3 4 5 6 7 8 9 10 Frequency Fi (times / month) 120 82 150 102 94 111 73 136 68 144 b - Warehouse structure parameter acquisition: Assume the warehouse is rectangular, 11 meters long and 9 meters wide. The storage locations are 1-meter * 1-meter squares, with a total of 20 storage locations, arranged in a 4-row and 5-column layout. The speed of the handling equipment is 1.5 meters per second, as shown in Figure 3 the schematic diagram of the layout of 10 kinds of goods before optimization.
[0026] Step (2): Data preprocessing Normalization processing (example: The volume of the good is 0.8m³ (range 0 - 1m³) and the weight is 600kg (range 0 - 1000kg). After normalization, both are converted into numerical values in the 0 - 1 interval to avoid the unreasonable dominant position of the weight in the model calculation due to the unit difference between "kg" and "m³".) Here, since the volume and weight of the goods both meet the storage location constraints, only the inbound and outbound frequencies (already counted) are directly used without additional normalization.
[0027] Step 3: Construct a mathematical model a - Objective function: With the shortest average access time of the goods as the objective, the formula is , where , 。
[0028] b - Constraints: Bin volume constraint: cargo volume ≤ 1 m³; Shelf load - bearing constraint: cargo weight ≤ 1000 kg; Uniqueness constraint: each bin stores only one type of cargo.
[0029] Merge the pre - optimization layout diagram of the 10 types of cargo shown Figure 3 into Table 1 above to obtain Table 2 below.
[0030] Table 2: Goods (1-10) 1 2 3 4 5 6 7 8 9 10 Frequency Fi (times / month) 120 82 150 102 94 111 73 136 68 144 Storage location coordinates (x, y) （2，4） （2，3） （5，4） （4，2） （2，2） （1，3） （5，3） （3，3） （3，4） （1，4） Calculate the access time of each type of cargo , : Cargo 1: seconds; Cargo 2: seconds; Cargo 3: seconds; Cargo 4: seconds; Cargo 5: seconds; Cargo 6: seconds; Cargo 7: seconds; Cargo 8: seconds; Cargo 9: seconds; Cargo 10: seconds; Calculate the numerator of the objective function :
[0031] Calculate the denominator of the objective function :
[0032] Calculate the value of the objective function seconds.
[0033] Step (4): (4 - 1) Particle swarm algorithm initialization a - Initialize relevant concepts: In the particle swarm algorithm, each particle represents a possible cargo - bin allocation scheme. Each particle is an array of length n (here n = 10, i.e., the number of types of cargo), and each element in the array represents the bin number where the corresponding cargo is stored. We need to randomly generate a certain number (e.g., m = 50) of particles to form the initial particle swarm.
[0034] b - Initialization steps and calculation process, as Figure 4 shown: Determine the number of particles m: Select m = 30 particles; Generate a single particle: For each particle, randomly assign 10 types of goods to 20 storage locations, ensuring that each storage location is assigned only one type of good, i.e., the storage location numbers do not repeat; Generate a particle swarm: Repeat the process of generating a single particle m times to obtain an initial particle swarm containing m particles.
[0035] Step (4): (4-2) Calculate the fitness value of the particle As Figure 5 shown, traverse each particle. For each storage location number in the particle, calculate the access time according to the storage location coordinates as the fitness value; the smaller the fitness value, the better the goods-storage location allocation scheme corresponding to the particle.
[0036] Step (4): (4-3) Update the particle velocity and position Set the inertia weight , learning factor , maximum number of iterations 50 times, and update according to the following formula Velocity update formula:
[0037] Position update formula:
[0038] Initialize the historical optimal position: At the beginning of the algorithm, the initial historical optimal position of each particle is its initial position, and the corresponding fitness value is the initial fitness value calculated in Step (4): (4-2).
[0039] For example, if the initial position of particle 1 is [8, 13, 3, 14, 4, 0, 11, 2, 15, 16] and its fitness value is 3.54, then in the initial stage, the historical optimal position of particle 1 is [8, 13, 3, 14, 4, 0, 11, 2, 15, 16], and the corresponding optimal fitness value is 3.54. In subsequent iteration processes, if particle 1 finds a new position with a fitness value less than 3.54, its historical optimal position and fitness value will be updated.
[0040] Initialize the global optimal position: Select the smallest one from the initial fitness values of all particles, and the corresponding particle position is the initial global optimal position. As Figure 6 shown, among the fitness values of the 30 particles given, the fitness value of particle 18 is the smallest, which is 2.80. Then the initial global optimal position is the position of particle 18 [3, 2, 8, 11, 18, 15, 12, 5, 10, 17], and the initial global optimal fitness value is 2.80.
[0041] During the iterative update process of the particle swarm algorithm, each particle adjusts its velocity and position based on its own historical best position and the global best position.
[0042] Step (4): (4-4) Control the number of iterations of the particles to determine the optimal cargo storage layout plan After 50 iterations of update, the particle swarm algorithm converges to the global optimal fitness value of 2.80, and the corresponding optimal position is: [3, 2, 8, 11, 18, 15, 12, 5, 10, 17], and mark the optimal positions of these 10 cargos on Figure 7 the corresponding storage locations in, and according to Figure 7 as shown, fill in the corresponding storage location distributions of the 10 storage locations in Table 3 below.
[0043] Table 3: Goods (1-10) 1 2 3 4 5 6 7 8 9 10 Frequency Fi (times / month) 120 82 150 102 94 111 73 136 68 144 Storage location coordinates (x, y) （4，4） （3，4） （4，3） （2，2） （4，1） （1，1） （3，2） （1，3） （1，2） （3，1） Step (4): (4-5) After optimization, verify the fitness value of the particles Calculate the access time of each cargo after optimization , : Cargo 1: seconds; Cargo 2: seconds; Cargo 3: seconds; Cargo 4: seconds; Cargo 5: seconds; Cargo 6: seconds; Cargo 7: seconds; Cargo 8: seconds; Cargo 9: seconds; Cargo 10: seconds Calculate the numerator according to the objective function
[0044] Calculate the denominator according to the objective function
[0045] Fitness after optimization , which is consistent with the iterative result, verifying that the obtained optimal solution is correct.
[0046] Step (5): (5-1) Execute the layout adjustment When the layout adjustment is executed, plan the handling path, prioritize the handling of high-frequency goods, and coordinate the operation of multiple devices with the help of the warehouse management system to avoid conflicts; at the same time, enter the optimized correspondence between storage locations and goods into the warehouse management system, and update the equipment scheduling strategy (such as adjusting task priorities and navigation rules) to ensure the efficient implementation of the layout adjustment.
[0047] Step (5): (5-2) Effect detection Compared with the average access time of goods in the initial layout ( ))), after the layout adjustment, the average access time of goods is reduced from 3.68 seconds to 2.80 seconds, and the average access time of goods is reduced by about 24% ( ), which significantly improves the handling efficiency of high-frequency goods and also significantly improves the warehouse operation efficiency; the system has dynamic adaptability, can quickly respond to frequency changes and maintain a high storage location utilization rate, verifying the effectiveness of the algorithm and the engineering practicability.
[0048] According to Figure 3 、 Figure 7 and the data in Table 1, calculate the total running distance of goods in the initial layout and the total running distance of goods after the layout adjustment: for example, for goods 3 with a frequency of 150 times per month, it is in storage location 4 (5,4) in the initial layout and in storage location 8 (4,3) after the layout adjustment; The total running distance of goods 3 in the initial layout is (5 + 4) * 150 = 1350; The total running distance of goods 3 after the layout adjustment is (4 + 3) * 150 = 1050; The reduction ratio of the total running distance of goods = (the total running distance of goods in the initial layout - the total running distance of goods after the layout adjustment) / the total running distance of goods in the initial layout = (1350 - 1050) / 1350 = 0.22. It can be seen that after the first layout optimization, the total running distance of some goods is reduced compared with the total running distance of goods in the initial layout; with the continuous optimization of subsequent particles, dynamic path planning can be carried out for the operation of goods, reducing the running distance of automated equipment during the process of handling goods, and thus reducing the energy consumption of the equipment.
[0049] The present invention is applicable to scenarios such as warehousing logistics centers, intelligent manufacturing factories, cross-border e-commerce bonded warehouses, third-party logistics, emergency material reserve warehouses, and automated stereoscopic warehouses. By optimizing storage location allocation and dynamic path planning, it effectively improves access efficiency, equipment utilization rate, and space density, and is particularly suitable for complex environments such as high-frequency demand fluctuations, multi-device collaboration, and high-density storage.
[0050] The present invention optimizes the warehousing layout through intelligent algorithms, significantly improving efficiency and space utilization. It is applicable to fields such as e-commerce, manufacturing, and cross-border logistics, especially suitable for high-density storage and high-frequency operation scenarios. It can effectively solve industry pain points, with strong market demand. It is expected to save more than 15% of costs annually, have a short investment payback period, and have significant technological innovation advantages, presenting broad commercialization prospects.
[0051] If the following problems are encountered during the implementation process, the following solutions can be adopted to solve the problems: (1) Incomplete or abnormal data collection: By adding a data verification mechanism, filling missing values with the mean, marking and manually reviewing outliers; synchronizing data with the order system and warehouse sensors in real time to ensure data integrity; (2) Equipment handling conflicts: By developing a dynamic path planning algorithm, monitoring the equipment location in real time and adjusting the handling order; adopting time window management to avoid overlapping of equipment paths.
Claims
1. An efficient and intelligent storage location allocation method, characterized in that The steps include: (1) The data acquisition module is used to comprehensively collect warehouse structural parameters and raw data of goods: the data acquisition module includes a first interface for connecting to the order management system and a second interface for collecting measurements or reading data of warehouse structural parameters. The first interface is used to capture the sales data of various types of goods from the order management system in real time and to count the frequency of goods entering and leaving the warehouse; the second interface is used to read warehouse structural parameters from the warehouse design drawings, collect and summarize these raw data, and transmit them to the data preprocessing module for preprocessing, so that these raw data are suitable for subsequent operations; (2) Using the data preprocessing module to normalize the collected raw data: the data preprocessing module includes a data conversion algorithm module and a data verification module; The data conversion algorithm module is used to convert data of different dimensions into a form that can be uniformly operated. The data verification module is used to verify the accuracy of the processed data, and then the processed data is transmitted to the mathematical model construction module. (3) Using the mathematical model construction module to construct a mathematical model for the optimization objective and constraint conditions: The mathematical model construction module includes an objective function construction unit, a constraint condition setting unit, and a model calculation unit; the objective function construction unit constructs a function with the shortest average storage and retrieval time of goods as the objective; the constraint condition setting unit sets the constraint conditions according to the fact that each cargo space can only store one type of goods and the volume and weight of the goods are limited; the model calculation unit substitutes the data from the data preprocessing module into the mathematical model; (4) The particle swarm optimization algorithm execution module is used to find the optimal cargo storage layout solution through iterative calculation: (4-1) Particle swarm algorithm initialization; (4-2) Calculate the fitness value of the particle; (4-3) Update particle velocity and position; (4-4) Control the number of particle iterations to determine the optimal cargo storage layout plan; (5) Based on the optimal cargo storage layout plan obtained in step (4), a layout adjustment execution module is used to adjust the cargo storage location and monitor the operation effect.
2. The efficient and intelligent storage location allocation method according to claim 1, characterized in that: In step (1), the original data of the goods include the type of goods, the volume of goods, the weight of goods and the frequency of goods entering and leaving the warehouse; the warehouse structure parameters include the warehouse area, the number of shelf layers, the number of shelf rows, the number of aisles, the size of the cargo space, the maximum cargo space volume of the warehouse and the maximum weight that the warehouse can bear.
3. The efficient and intelligent storage location allocation method according to claim 2, characterized in that: In step (2), the data conversion algorithm module is used to convert data of different dimensions into a form that can be uniformly calculated by dividing the cargo volume by the maximum cargo space volume of the warehouse and dividing the cargo weight by the maximum weight that the warehouse can bear. The collected cargo volume and weight data are normalized to achieve the effect of making the data conform to the requirements of subsequent mathematical models and algorithm operations, thereby providing high-quality data support for subsequent processes.
4. The efficient and intelligent storage location allocation method according to claim 3, characterized in that: In step (3), the formula As a mathematical model, the formula Calculate the average storage and retrieval time of goods , which provides a quantitative basis for evaluating the layout scheme for the particle swarm optimization algorithm execution module; where, Indicates the average storage and retrieval time of goods; Indicates goods The frequency of inbound and outbound storage; Indicates The storage and retrieval time of a certain type of goods at a certain storage location; Indicates that each type of goods Frequency of inbound and outbound and retrieve the goods from the storage location Time After multiplication, accumulation is performed. , is the number of cargo types; It represents the sum of the in-and-out frequencies of all goods.
5. The efficient and intelligent storage location allocation method according to claim 2, characterized in that: The particle swarm optimization algorithm execution module in step (4) includes a particle initialization unit, a fitness calculation unit, a speed / position update unit and an iteration control unit; Step (4-1) The particle initialization unit randomly generates particles and initializes the starting position and speed of each particle according to the cargo and cargo location information involved in the mathematical model of step (3). Each particle represents a cargo storage layout plan, and the distribution of the cargo in each cargo location is represented by a position vector; Step (4-2) The fitness calculation unit calculates the fitness value of the particle according to the mathematical model; Step (4-3) The speed / position updating unit updates the particle speed and position by calculation according to the particle's own historical optimal position and the global optimal position; In step (4-4), the iterative control unit controls the number of iterations of the particles until the preset number of iterations or fitness value convergence conditions are met, and the position vector corresponding to the optimal particle is found, which is the optimal cargo storage layout plan, and then the optimal cargo storage layout plan is passed to the layout adjustment execution module.
6. The efficient and intelligent storage location allocation method according to claim 5, characterized in that: In the step (4-2), when calculating the fitness value, it is necessary to substitute the cargo storage layout plan corresponding to each particle into the mathematical model according to the established mathematical model. The fitness value of each particle is calculated; the smaller the fitness value of the particle, the better the cargo storage layout plan.
7. The efficient and intelligent storage location allocation method according to claim 5, characterized in that: In step (4-3), when updating the particle position and speed, it is necessary to use the speed update formula based on the particle's own historical optimal position and the global optimal position of the entire particle group And the position update formula , by continuously updating the speed and position of particles according to these two formulas, the particle swarm continuously iterates and optimizes in the search space, gradually approaching the optimal cargo storage layout plan, thereby finding the optimal cargo storage layout plan for the automated warehouse; In the velocity update formula: It is a particle In Dimension The current velocity of the particle In this dimension on current sports trends; It is called inertia weight, which is used to balance the global and local search capabilities of particles; larger A value of helps particles to conduct a wider global exploration in the search space, while a smaller The value can make the particles focus more on the local area and conduct a detailed search; and is the learning factor, Mainly affects the particle to its own historical optimal position The degree of learning, This determines the particle's global optimal position The intensity of learning; By adjusting these two factors, it is possible to control the degree to which particles rely on their own experience and group experience during the search process; and is a random number between 0 and 1, which introduces randomness to the speed update, so that the particle will not fall into the local optimal solution during the search process and can explore the search space more comprehensively; It is a particle In Dimension The historical optimal position on is the optimal position that the particle has reached in the previous search process; It is a particle In Dimension Current location on ; This item prompts the particle to move to its historical optimal position, reflecting the particle's learning from its own experience; This term pushes the particle toward the global optimal position, reflecting the particle's learning of the optimal experience in the group; In the position update formula: Represents particles In Dimension New position on Represents particles In Dimension Current location on ; Represents particles In Dimension The updated speed.
8. The efficient and intelligent storage location allocation method according to claim 1 or 2, characterized in that: In step (5), the layout adjustment execution module includes an instruction sending unit, a handling sequence planning unit and an operation monitoring unit; the instruction sending unit sends instructions for adjusting the storage position of goods to the automated equipment through the warehouse management system; the handling sequence planning unit reasonably arranges the cargo handling sequence to avoid equipment conflicts and collisions; the operation monitoring unit collects and adjusts the cargo storage and retrieval efficiency and equipment energy consumption of the warehouse, feeds back the warehouse cargo storage and retrieval efficiency and equipment energy consumption to the particle swarm optimization algorithm execution module in step (4), and passes the new warehouse operation data to the data acquisition module in step (1) for continuous optimization.
9. The efficient and intelligent storage location allocation method according to claim 8, characterized in that: In step (5), the warehouse management system sends an instruction to the automated equipment to adjust the storage location of the goods, and the storage location of the goods is adjusted according to the optimal cargo storage layout plan. During the adjustment process, the goods that are far away from the entrance and exit and have a low frequency of entering and leaving the warehouse are first moved to the new location, and then the positions of other goods are gradually adjusted.
10. The efficient and intelligent storage location allocation method according to claim 9, characterized in that: In step (5), after the layout adjustment is completed, the operation monitoring unit monitors and evaluates the warehouse operation status, and compares the cargo storage and retrieval efficiency and equipment energy consumption before and after the adjustment; if the operation effect of the cargo storage location is not ideal, the algorithm parameters are further adjusted or re-optimized to continuously improve the efficiency of the automated warehouse cargo storage layout.