Book box with variable partition plates and boxing method

Abstract

The invention provides a book box with variable partition plates and a boxing method. The method comprises the steps that a book list and an initial box body space are obtained; based on a branch and bound-greedy hybrid optimization algorithm, carrying out space state circulation processing to obtain an optimal space allocation scheme; based on the optimal space distribution scheme, re-planning the space layout, updating the optimal position of the partition plate, and obtaining the sizes of the four divided spaces; and on the basis of the sizes of the four divided spaces, book placement information is obtained, and books are placed in sequence. According to the method, the optimal space distribution scheme of the four partition spaces is obtained through the branch and limit-greedy hybrid optimization algorithm, so that the book placement direction and position are selected, and the residual space is minimum.

Description

Technical Field

[0001] This application belongs to the field of automated warehousing and logistics technology, specifically relating to a variable partition book box and a packing method. Background Technology

[0002] In the book warehousing and logistics process, boxing is a crucial step connecting book storage and transportation; its efficiency and space utilization directly impact warehousing costs and turnover efficiency. Currently, there are two main methods for implementing book boxing, both of which have significant limitations:

[0003] Traditional book packing relies on manual sorting, stacking, and placement, with the following core problems: 1. High labor intensity and low efficiency: Staff need to manually identify book sizes and adjust their placement, making each packing operation time-consuming and unable to meet the needs of large-scale, batch book processing; 2. Low space utilization: Manual placement lacks systematic planning, easily leading to wasted space inside the box due to book misalignment and excessive gaps, with an average space utilization rate of less than 60%; 3. Chaotic spine orientation: Manual operation makes it difficult to ensure that all books have the same spine orientation, requiring additional time for sorting during subsequent warehousing and inventory checks, increasing operating costs.

[0004] To address the efficiency issues of manual packing, robotic packing solutions have been introduced in some scenarios, but significant technical bottlenecks remain: Container structure flexibility is insufficient; most solutions use fixed partition containers, and once the space division is determined, it cannot be dynamically adjusted, making it difficult to adapt to scenarios with large differences in book size and dynamic batch changes. Stacking strategies are crude, generally employing simple heuristics such as "large to small" and "heavy to light," without considering the impact of the robotic arm's grasping posture on the subsequent book stacking direction—for example, if the robotic arm grasps a book with its width x thickness facing up and then places it directly, subsequent books may not fit tightly together, creating ineffective gaps. Algorithm optimization is unbalanced; existing algorithms struggle to balance "global optimum" and "real-time performance": pursuing global optimum using pure branch and bound, genetic algorithms, etc., results in excessive computation time, failing to meet the real-time requirements of automated operations; pursuing real-time performance using pure greedy algorithms easily leads to local optima, limiting the improvement in space utilization.

[0005] In summary, existing technologies cannot simultaneously meet the multi-dimensional requirements of "high space utilization, strong scene adaptability, sufficient real-time operation, and uniform spine orientation," and there is an urgent need for an automatic packing solution that integrates hardware structure innovation and algorithm optimization. Summary of the Invention

[0006] This application provides a variable-partition bookcase and a packing method to solve the above-mentioned technical problems.

[0007] To solve the above-mentioned technical problems, one technical solution adopted in this application is: a method for packing a variable-partition bookcase, comprising:

[0008] S1. Obtain the book list and initial box space;

[0009] S2. Based on the branch and bound-greedy hybrid optimization algorithm, perform space state cyclic processing to obtain the optimal space allocation scheme;

[0010] S3. Based on the optimal space allocation scheme, replan the space layout, update the optimal position of the partition and obtain the dimensions of the four divided spaces;

[0011] S4. Based on the dimensions of the four segmented spaces, obtain the book placement information and place the books in order.

[0012] Furthermore, the method in step S1 includes:

[0013] S11. Based on the book list, obtain the name and size of each book;

[0014] S12. Based on the initial position of the partition, obtain the initial size of the segmented space.

[0015] Furthermore, the method in step S2 includes:

[0016] S21. Based on the initial positions of the horizontal and vertical partitions, obtain the areas of the four subspaces;

[0017] S22. Based on the size of the books and space utilization, choose the orientation in which the books are placed;

[0018] S23. Construct a comprehensive cost function to maximize the space utilization of the four subspaces and obtain the optimal position range of the transverse and longitudinal partitions;

[0019] S24. Based on the optimal position range of the transverse and longitudinal partitions and the comprehensive loss function, determine the optimal position of the transverse and longitudinal partitions.

[0020] Furthermore, the method in step S21 includes:

[0021] Based on formulas (1)-(4), the dimensions of the four partitioned spaces are obtained; where formulas (1)-(4) are:

[0022] (1);

[0023] (2);

[0024] (3);

[0025] (4);

[0026] in, and These are the initial coordinate positions of the transverse and longitudinal partitions; Represents the length of the box; Represents the width of the box; This represents the base area of ​​the k-th space.

[0027] Furthermore, the method in step S22 includes:

[0028] Based on formula (5), the space utilization rate is obtained; where formula (5) is:

[0029] (5);

[0030] in, This indicates the current space utilization rate of the container; Indicates a collection of books; The corresponding length of the book and the corresponding projection size of the spine on the bottom surface of the box; The width corresponds to the book's width, and the thickness corresponds to the page orientation. Corresponding to the height of the book, corresponding to the vertical dimension; This represents the total area of ​​the book box.

[0031] Furthermore, the method in step S23 includes:

[0032] Based on formula (6), a comprehensive cost function is constructed; the formula is:

[0033] (6);

[0034] in, This indicates the actual area occupied by the container. This represents the total distance the partition moves from its current position to a candidate position; As the weight of the partition energy consumption; The fragmentation penalty reflects the degree to which the remaining space is divided into invalid small pieces; For fragmentation penalty weights;

[0035] Based on formula (7), the optimal position interval is obtained; where formula (7) is:

[0036] (7);

[0037] in, This indicates the coordinate positions of the horizontal and vertical partitions.

[0038] Furthermore, the method in step S24 includes:

[0039] Based on formula (8), the partition cost function is obtained; where formula (8) is:

[0040] (8);

[0041] in, The total area of ​​the k-th subspace; The area already utilized in the Kth subspace; These are the weighting coefficients.

[0042] Furthermore, the method in step S24 includes:

[0043] Based on formula (8), the comprehensive loss function is obtained; wherein, formula (8) is:

[0044] (8);

[0045] in, The total area of ​​the k-th subspace; The area already utilized in the Kth subspace; These are the weighting coefficients.

[0046] Another technical solution adopted in this application is: a variable partition bookcase, comprising:

[0047] The box is used to store books of different sizes, and it contains multiple movable partitions.

[0048] The latching track is located inside the box on the left, right, front, and rear sides;

[0049] The longitudinal partition is connected to the buckle track on both sides by buckles, and the longitudinal partition slides back and forth along the buckle track;

[0050] The transverse partition is connected to the buckle track on both sides by buckles and inserted into the slots in the longitudinal partition. The longitudinal partition slides back and forth along the buckle track.

[0051] The beneficial effects of this application are: This application obtains the optimal space allocation scheme of four partitioned spaces through a branch and bound-greedy hybrid optimization algorithm, thereby selecting the book placement direction and position to minimize the remaining space. Attached Figure Description

[0052] Figure 1 This is a schematic flowchart of an embodiment of the packing method for the variable partition bookcase of this application;

[0053] Figure 2 yes Figure 1 A flowchart illustrating step S2 of an embodiment;

[0054] Figure 3 This is a structural schematic diagram of an embodiment of the variable partition bookcase of this application. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments.

[0056] Numerous specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways than those described herein, and therefore the invention is not limited to the specific embodiments disclosed in the following specification.

[0057] See Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the packing method for the variable partition bookcase of this application; the method includes:

[0058] S1. Obtain the list of books and the initial box space.

[0059] Specifically, the method in step S1 includes:

[0060] S11. Based on the book list, obtain the name and size of each book.

[0061] Specifically, the book information in the book list is obtained, which includes attributes such as book length Li, width Wi, thickness Hi, and unique number (id) to distinguish different books and record placement information, and the base area is fixed at "Li*Hi" (strictly following the "placement direction constraint" of the algorithm).

[0062] S12. Based on the initial position of the partition, obtain the initial size of the segmented space.

[0063] Specifically, it obtains the dimensions of the internal space of the enclosure, including the x-coordinate and y-coordinate of the lower left corner of the space, the width (w), and the height (h), providing three main functions.

[0064] The total number of books to be packed is N; the size of a single book is the length of each book i. ,width i. Thickness ;

[0065] Box dimensions: Length =800mm, width =600mm, height =300mm;

[0066] Initial partition position: the division position along the length direction Segmentation positions along the width direction .

[0067] Enter the list of books to be packed (including the contents of each book). The inputs to the algorithm are id (the number of books N and the size of a single book) and an initial space list (four variable rectangular spaces inside the book box, and the initial partition positions). and The default settings are 200mm and 300mm (i.e.) =200mm =300mm), which is the "initial partition position" of the algorithm input); initialize the optimal bin packing scheme to be empty, build a priority queue (to achieve a max heap with "negative utilization", and prioritize the processing of states with high utilization), and define the initial state as "utilization 0.0, empty bin packing scheme, initial space list, complete book list".

[0068] S2. Based on the branch and bound-greedy hybrid optimization algorithm, perform spatial state cyclic processing to obtain the optimal space allocation scheme.

[0069] Specifically, the state with the highest utilization rate is popped from the priority queue. If the remaining books to be packed are empty, then the current solution is the optimal solution, and the loop terminates. If the utilization rate of the current state is less than 90% of the optimal solution found, then the state is pruned (to reduce invalid calculations and improve real-time performance). Otherwise, the current book to be packed is retrieved, and all available spaces are traversed. Two rotation orientations, 0° and 90°, are tried (only the placement orientation of Li and Hi is adjusted, while the base area remains Li*Hi). The Space class method is used to determine whether the book fits the space (verifying that max(Li,Hi)≤max(w,h) and min(Li,Hi)≤min(w,h)).

[0070] See Figure 2 The method of step S2 includes:

[0071] S21. Based on the initial positions of the horizontal and vertical partitions, obtain the areas of the four subspaces.

[0072] Specifically, under the constraints of the box-shaped space, there is a collection of books:

[0073] ;

[0074] in, These are the book's length, width, and thickness, respectively.

[0075] The geometric parameters of the box are defined as follows: ;

[0076] The initial coordinates of the partition on the bottom surface are ( , The box is divided into four rectangular subspaces:

[0077] (1);

[0078] (2);

[0079] (3);

[0080] (4);

[0081] The bottom area of ​​the box is: ;

[0082] S22. Based on the size of the books and space utilization, choose the orientation in which the books are placed.

[0083] Specifically, to measure packing efficiency, this invention defines space utilization rate:

[0084] (5);

[0085] in, Let be the set of books already placed. The bin packing problem in this application is essentially a two-dimensional rectangle filling problem, with the geometric objective of maximizing the placement. The ratio of the total base area of ​​the books to the base area of ​​the bin is considered. Since the bin height is fixed and the partitions only affect the planar division, the thickness dimension does not participate in the objective function at this stage. This space utilization is monotonic (η does not decrease with increasing placement) and bounded (η≤1), meeting the requirements of the optimization problem for feasible region and convergence.

[0086] The packing process is modeled as a state transition system:

[0087] ;

[0088] in This is a collection of books to be placed. This is the set of available space.

[0089] After each placement operation, the system state evolves as follows:

[0090] ;

[0091] in To select books, The placement angle.

[0092] To avoid exponential search, a hybrid strategy of "branch and bound + greedy local optimization" is adopted:

[0093] Branch and Bound Level: Set the current optimal upper bound. If the potential maximum utilization rate of a certain state is estimated satisfy: If the condition is met, then the branch is pruned and discarded. This condition guarantees that the search space monotonically shrinks.

[0094] Greedy local layer: For each subspace Select the best matching book:

[0095] ;

[0096] It also attempts two rotation directions. This strategy uses heuristic selection of local optima to accelerate global convergence. The fusion of these two strategies ensures the convergence and computational feasibility of the algorithm, where the branch and bound layer guarantees a monotonically decreasing upper bound on the solution, and the greedy layer guarantees fast local approximation.

[0097] S23. Construct a comprehensive cost function to maximize the space utilization of the four subspaces and obtain the optimal position range of the transverse and longitudinal partitions.

[0098] Specifically, based on formula (6), a comprehensive cost function is constructed; the formula is:

[0099] (6);

[0100] in, This indicates the actual area occupied by the container. This represents the total distance the partition moves from its current position to a candidate position; As the weight of the partition energy consumption; The fragmentation penalty reflects the degree to which the remaining space is divided into invalid small pieces; For fragmentation penalty weights;

[0101] The overall cost function contains two mutually constraining directions:

[0102] f1: Maximize space utilization; f2: Minimize energy consumption and fragmentation.

[0103] Because there is a trade-off between the two, this invention uses a weighted sum method to form a comprehensive objective: Expanding this gives us the above equation, where... These correspond to the normalized weights of the energy consumption and fragmentation terms, respectively. This function describes the balance between two conflicting objectives in a real-world packing system: when When the size is smaller, the system tends to pursue higher packing density; when With increased weights, the system tends to reduce movement and fragmentation.

[0104] The core idea of ​​fragmentation penalty originates from the "placeability determination" in spatial geometry. Each subspace unit is defined. Minimum accommodable book area :

[0105] like or If so, the space is unavailable.

[0106] therefore: This formula can be interpreted as the "percentage of ineffective space." Physically, it means "the proportion of empty space where no more books can be placed after packing." In actual packaging production, the smaller this value, the more compact the packing and the higher the shock resistance during transport.

[0107] Based on formula (7), the optimal position interval is obtained; where formula (7) is:

[0108] (7);

[0109] The candidate partition set (x', y') indicates that the partition position can be adjusted within a certain step range. , This represents the maximum movable step length of the partition drive system, in millimeters.

[0110] S24. Based on the optimal position range of the transverse and longitudinal partitions and the partition cost function, determine the optimal position of the transverse and longitudinal partitions.

[0111] Specifically, the movement of partitions and the division of space directly affect packing efficiency and mechanical energy consumption.

[0112] Therefore, this application comprehensively considers three aspects when constructing the integrated loss function: spatial equilibrium, energy consumption of partition movement, and geometric fragmentation. The formula is:

[0113] (8);

[0114] in, The total area of ​​the k-th subspace; The area already utilized in the Kth subspace; These are the weighting coefficients.

[0115] The logic behind the formation of this formula is as follows:

[0116] Spatial equilibrium term The goal of partition adjustment is to achieve the most balanced loading across all zones. If one subspace is highly saturated while another is almost empty, packing efficiency decreases. Therefore, let the "unused area difference" of each subspace be... Weight This reflects the importance of the area (e.g., spaces near the edge of the enclosure have higher priority). This term originates from the principle of spatial equilibrium, corresponding to "error energy minimization" in control theory.

[0117] The partition movement cost item The partition is driven by a servo motor during the packing process. The greater the movement distance, the higher the energy consumption and the longer the execution time. Since the movement trajectory of the partition is a planar linear motion, the power consumption is approximately proportional to the displacement.

[0118] In conclusion, This represents the overall cost of the system after adjusting the baffle. Its physical meaning is "the comprehensive loss function under the constraints of energy consumption and spatial balance". By minimizing J, the algorithm automatically finds the optimal position of the baffle.

[0119] S3. Based on the optimal space allocation scheme, replan the space layout, update the optimal position of the partition and obtain the dimensions of the four divided spaces.

[0120] Specifically, the optimal space allocation scheme includes constraints on partition position, space capacity, book allocation, and placement orientation. The horizontal and vertical partitions are moved according to their optimal positions.

[0121] S4. Based on the dimensions of the four segmented spaces, obtain the book placement information and place the books in order.

[0122] See Figure 3 , Figure 2 This is a structural schematic diagram of an embodiment of the variable partition bookcase of this application. The bookcase includes: a box body 1, which is used to store books of different sizes, wherein the box body 1 contains multiple movable partitions; the box body 1 is constructed of aluminum profiles with a rectangular cross-section, and the aluminum profiles are fixedly connected by angle brackets and bolts to form a rectangular frame with a length × width × height of 800mm × 600mm × 300mm; waterproof plates can be selectively installed on the bottom and sides of the frame to prevent the books from getting damp.

[0123] The housing 1 includes: a row of rails 2, a longitudinal partition 3, and a transverse partition 4.

[0124] The snap-lock track 2 is located inside the box 1 on the left, right, front, and rear sides. Multiple snap-lock tracks 2 are evenly spaced along the inner wall of the basic box 1 in both the length (X-axis) and width (Y-axis) directions. The snap-lock track 2 is a U-shaped, bi-directional open guide rail structure with evenly distributed slots (5mm spacing) on ​​the inner side. The snaps used to connect the snap-lock track 2 and the partitions are made of elastic plastic and are shaped to match the slots of the snap-lock track 2. They can be inserted into any slot of the snap-lock track 2 by pressing, or removed to change position.

[0125] The book box is equipped with one vertical baffle (along the X-axis, corresponding to x) and one horizontal baffle (along the Y-axis, corresponding to y), with a travel range of 0-400mm for both baffles. The two baffles divide the internal space into four variable rectangular spaces, providing a physical basis for the algorithm to dynamically adjust the space size and adapt to the packing of books of different sizes.

[0126] The longitudinal partition 3 is connected to the latching track 2 on both sides via latches, and the longitudinal partition 3 slides back and forth along the latching track 2. The transverse partition 4 is connected to the latching track 2 on both sides via latches and is inserted into the slots in the longitudinal partition 3, and the longitudinal partition 3 slides back and forth along the latching track 2. The sliding of the partitions is adjusted by external mechanical arms or adjusting levers according to the calculated optimal space allocation scheme, thereby maximizing the book capacity and minimizing the remaining space.

[0127] Horizontal partition 3 and vertical partition 4: both are thin plastic sheets (thickness 2mm, strength ≥5MPa), and the edges of the partitions are provided with buckles that cooperate with the buckles; by connecting the buckles of the horizontal partition 3 and the vertical partition 4 with the buckles at different positions, the interior of the basic box 1 can be divided into four independent rectangular spaces, and the space size can be dynamically changed according to the position of the buckles.

[0128] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for packing a variable-partition bookcase, characterized in that, Includes the following steps: S1. Obtain the book list and initial box space; S2. Based on the branch and bound-greedy hybrid optimization algorithm, perform space state cyclic processing to obtain the optimal space allocation scheme; S3. Based on the optimal space allocation scheme, replan the space layout, update the optimal position of the partition and obtain the dimensions of the four divided spaces; S4. Based on the dimensions of the four segmented spaces, obtain the book placement information and place the books in sequence.

2. The method according to claim 1, characterized in that, The method of step S1 includes: S11. Based on the book list, obtain the name and size of each book; S12. Based on the initial positions of the transverse and longitudinal partitions, obtain the initial dimensions of the segmented space.

3. The method according to claim 2, characterized in that, The method of step S2 includes: S21. Based on the initial positions of the transverse partition and the longitudinal partition, obtain the area of ​​the four subspaces; S22. Based on the size of the book and the space utilization rate, select the orientation in which the book is placed; S23. Construct a comprehensive cost function to maximize the space utilization of the four subspaces and obtain the optimal position range of the transverse partition and the longitudinal partition; S24. Based on the optimal position range of the transverse partition and the longitudinal partition and the comprehensive loss function, determine the optimal position of the transverse partition and the longitudinal partition.

4. The method according to claim 3, characterized in that, The method of step S21 includes: Based on formulas (1)-(4), the dimensions of the four segmented spaces are obtained; wherein, formulas (1)-(4) are: （1）； （2）； （3）； （4）； in, and The initial coordinate positions of the transverse and longitudinal partitions; Represents the length of the box; Represents the width of the box; This represents the base area of ​​the k-th space.

5. The method according to claim 2, characterized in that, The method of step S22 includes: The space utilization rate is obtained based on formula (5); wherein, formula (5) is: （5）； in, This indicates the current space utilization rate of the container; Indicates a collection of books; The corresponding length of the book and the corresponding projection size of the spine on the bottom surface of the box; The width corresponds to the book's width, and the thickness corresponds to the page orientation. Corresponding to the height of the book, corresponding to the vertical dimension; This represents the total area of ​​the book box.

6. The method according to claim 5, characterized in that, The method of step S23 includes: Based on formula (6), the comprehensive cost function is constructed; the formula is: （6）； in, This indicates the actual area occupied by the container. This represents the total distance the partition moves from its current position to a candidate position; As the weight of the partition energy consumption; The fragmentation penalty reflects the degree to which the remaining space is divided into invalid small pieces; For fragmentation penalty weights; Based on formula (7), the optimal position interval is obtained; wherein, formula (7) is: （7）； in, This indicates the coordinate positions of the transverse partition and the longitudinal partition.

7. The method according to claim 6, characterized in that, The method of step S24 includes: Based on formula (8), the comprehensive loss function is obtained; wherein, formula (8) is: （8）； in, The total area of ​​the k-th subspace; This represents the area already utilized in the Kth subspace. These are the weighting coefficients.

8. A type of variable-partition bookcase, characterized in that, include: A box for storing books of different sizes, wherein the box contains multiple movable partitions; The buckle track is located inside the box on the left, right, front, and rear sides; A longitudinal partition is connected to the buckle track on both sides by buckles, wherein the longitudinal partition slides left and right along the buckle track; A transverse partition has two sides connected to the buckle track via buckles and inserted into the slots of the longitudinal partition, wherein the longitudinal partition slides back and forth along the buckle track.

Images