An adaptive automatic bin packing method considering stability of goods

Abstract

The present application relates to the technical field of computer, and especially relates to a self-adaptive automatic loading method considering stability of goods, obtaining information of goods to be loaded and placement mode parameters, the information of goods including goods type, self weight, bearable weight and size, the placement mode parameters including placement pose; simple blocks are formed by simple grouping for goods of the same type, and mixed blocks are formed by mixed grouping for goods of different types; based on the simple blocks and the mixed blocks, an adaptive selection algorithm is used for evaluation, and placement of the goods in the loading box is determined as a flat placement mode or a ladder placement mode; a flat placement scheme or a ladder placement scheme is generated according to the placement mode of the goods. The adaptive selection algorithm is used for evaluation of the simple blocks and the mixed blocks in the loading box as flat placement or ladder placement, so that the safety of goods transportation and the space utilization rate are ensured, and the method is suitable for logistics loading problems containing various complex constraints. The method is suitable for loading of trucks, containers and airplane cargo holds.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to an adaptive automatic packing method that takes into account the stability of goods. Background Technology

[0002] Since the beginning of the 21st century, the booming global economy has injected strong momentum into the logistics industry. With the continuous expansion of enterprise production scale and the constant innovation of production technology, the logistics industry has ushered in unprecedented development opportunities. Against this backdrop, the key link in logistics transportation—cargo container selection and cargo packing optimization—has increasingly become the focus of industry attention. Scientific and reasonable cargo selection and packing schemes can not only maximize the utilization of cargo container space and improve transportation efficiency, but also effectively reduce the frequency of cargo handling and significantly reduce time costs. Therefore, in-depth research on the three-dimensional packing problem not only has important theoretical value, but also demonstrates great practical significance and application potential in improving logistics efficiency and reducing transportation costs.

[0003] In the logistics packing process, multiple orders of goods typically need to be placed into the same container and transported to different customers. The goal is to select the most suitable packing method and determine the placement and orientation of each item to maximize loading efficiency and minimize packing costs. Generally, to maximize space utilization, different types of goods are mixed and stacked. However, for automated packing equipment, the same type of goods need to be stacked continuously and in a concentrated manner, while meeting specific mixed-packing constraints. Existing technologies for handling 3D packing problems mainly focus on solving for space utilization, with relatively simple considerations of constraints. For example, CN113762899A discloses a 3D packing method for goods based on a hybrid algorithm. Besides loading order and weight support constraints, it does not simultaneously consider many key factors applicable to automated packing equipment, such as mixed-packing methods and anti-tipping supports. This limitation means that existing technologies may not show significant deviations when handling simple goods packing. However, when faced with logistics packing problems involving multiple complex constraints, existing technologies struggle to adapt effectively and cannot provide ideal solutions.

[0004] Furthermore, if goods are in various different positions during packing, it places higher demands on automated packing equipment. This not only means more investment in the equipment construction phase, but also frequent adjustments are required after the equipment is put into actual production. These additional costs and adjustments pose a significant potential obstacle to improving packing efficiency. Summary of the Invention

[0005] The purpose of this invention is to propose an adaptive automatic packing method that takes into account the stability of goods. An adaptive selection algorithm is used to evaluate whether simple blocks and mixed blocks are placed flat or in a stepped manner in the cargo box, so as to ensure the safety of cargo transportation and improve space utilization. It is applicable to logistics packing problems with multiple complex constraints.

[0006] To achieve this objective, the present invention adopts the following technical solution:

[0007] An adaptive automated packing method considering cargo stability includes the following steps:

[0008] S1. Obtain information on the goods to be packed and the placement parameters. The goods information includes the type of goods, its own weight, the weight it can bear, and its own dimensions. The placement parameters include the placement position.

[0009] S2. Simple blocks are formed by simply grouping the same type of goods into simple blocks, and mixed blocks are formed by mixing different types of goods into mixed blocks.

[0010] S3. Based on simple blocks and mixed blocks, an adaptive selection algorithm is used to evaluate and determine whether the goods are placed in the cargo box in a flat or stepped manner.

[0011] S4. Generate a flat placement plan or a tiered placement plan based on the placement method of the goods.

[0012] Furthermore, in step S2, the size of the block is defined by the structure of the cargo box: the loading depth of the cargo box is defined as the length of the box, the bottom width inside the cargo box is defined as the width of the box, and the height inside the cargo box is defined as the height of the box. When constructing the block, the length, width, and height of the block or cargo are parallel to the length, width, and height of the box.

[0013] When building blocks, only one column is constructed, and the pose of the same type of goods is kept consistent, and the type of goods in the same layer is kept consistent.

[0014] Furthermore, in step S2, the method for constructing a simple block includes:

[0015] Using a simple block as a wall block, the pose of the cargo remains unchanged when constructing the simple block. The length of the cargo is used as the length of the wall block to be formed. The maximum integer multiple that the cargo can fill in the width direction of the cargo box is used as the width of the wall block to be formed. Then, under the condition of satisfying the weight constraint, the blocks are stacked in integer multiples in the height direction to obtain wall blocks of different heights.

[0016] Methods for constructing mixed-use blocks include:

[0017] Mixed blocks are obtained by combining pre-constructed simple blocks. During the combination process, constraints are applied in the width direction to control the width difference between the upper and lower layers of goods within a set range, while also satisfying weight constraints.

[0018] Furthermore, in step S3, an adaptive selection algorithm is used to evaluate and determine whether the goods are placed flat or in a stepped manner within the cargo container, including:

[0019] The cargo's attribute trade-off parameters are defined based on its height and volume;

[0020] The stability parameters are calculated based on the block construction results: the stability parameters are approximately calculated by the lateral loading rate of the block, which is represented by the ratio between the length of the cargo block in the width direction of the cargo box and the width of the cargo box.

[0021] Distance trade-off parameters are derived by considering the distance between the first and last orders, the total transportation distance, and the number of orders.

[0022] The decision score is obtained by weighting and mapping the attribute trade-off parameters, stability parameters, and distance trade-off parameters.

[0023] Choose between a flat or tiered placement method based on the decision score.

[0024] Furthermore, the method for defining the attribute trade-off parameters of cargo based on cargo height and volume is as follows:

[0025] Let the height of the i-th item be h. i , i∈[0,n), volume is v i For i ∈ [0, n), the average height of all goods is Average volume is The formula for calculating the attribute tradeoff parameter α is as follows:

[0026] Among them, c min c represents the maximum number of chunk layers for the smallest cargo. step This indicates the maximum number of tiers for the last order;

[0027] The formula for calculating the stability parameter β is: Where, β ij J represents the lateral loading rate of cargo i in the j-th pose; i N represents the pose contained in cargo i; ij This indicates the total number of positions of the goods. For threshold;

[0028] Lateral Loading Rate β ij The calculation formula is: Where W represents the width of the cargo box; w ij Indicates the width of cargo i in the j-th pose; n ijThis represents the maximum loading quantity of cargo i in the j-th pose along the width direction of the cargo box;

[0029] The formula for calculating the distance tradeoff parameter γ is: Where m is the number of orders, L is the total shipping distance for the orders, and l is the distance to the destination of the first order;

[0030] The formula for calculating the decision score Z is: Z = α + β + 2γ, Z ∈ [0, 5]. When the decision score is greater than 2.5, the tiered placement method is selected, and when the decision score is less than or equal to 2.5, the flat placement method is selected.

[0031] Furthermore, in step S4, the method for generating the tiling scheme includes:

[0032] Using the cargo box as the loading space, and under the constraints of weight and loading sequence, a greedy strategy is adopted to select and load blocks with the goal of maximizing space utilization to obtain a preliminary loading plan.

[0033] Based on the preliminary loading plan, determine whether all the goods to be loaded can be loaded and obtain the remaining space in the length direction of the cargo box;

[0034] If there is remaining space along the length of the cargo box, adjust the loading space according to the size of the remaining space, reduce the loading height, and ensure that the cargo can be supported from the front and back after loading; re-obtain the placement plan until the plan meets the flat loading constraint;

[0035] If a tiling plan cannot be obtained within the set time, the current plan will be output and the remaining goods will be output.

[0036] Furthermore, the method for re-acquiring the placement scheme until the scheme satisfies the tiling loading constraint includes:

[0037] Let the new loading height be h1 and the original loading height be the upper limit h2. Under the constraint of the h1 height, we will use a greedy strategy to search again to obtain the placement solution.

[0038] If the obtained placement scheme can achieve flat placement of goods, the result is output. If there is still remaining space for placing goods, h1 and h2 are updated again and a greedy strategy is used for searching. If the goods cannot be placed in the end or there is less than one layer left, they are stacked directly. Otherwise, h1 and h2 are updated, with (h1+h2) / 2 as h1, and the upper limit height h2 remains unchanged, until a placement scheme is obtained or there is no more search time.

[0039] Furthermore, in step S4, the method for generating the staircase placement scheme includes:

[0040] Goods in adjacent columns must meet a height difference constraint, where the height difference must be less than or equal to the height of the smaller good. Furthermore, when placing items in a hierarchy, the basic constraint that later items should be placed on top must still be met. Based on this, the good with the highest space utilization is selected for placement; as follows:

[0041] First, place the goods with the highest space utilization in the last order in a staircase. Use the wall block with the highest space utilization as the pose and the number of layers as the maximum number of layers m of the current staircase. Start placing the goods on the first layer and make them touch the door of the cargo box. Then, place them on the second or third layer until the maximum number of layers is reached. If the goods cannot be placed, add them to the remaining goods in the order.

[0042] After the tiered arrangement is completed, the remaining goods are laid out flat based on the height of the cargo box.

[0043] If the space outside the steps cannot be fully loaded, i.e. the height difference between the steps and the space is too large, then the steps should be adjusted: reduce the maximum number of steps (m) and then place and load the remaining goods on the steps.

[0044] The technical solution provided by this invention may include the following beneficial effects:

[0045] This invention adds restrictions on two support modes: flat or stepped, to ensure the safety of cargo transportation. This invention is aimed at automated packing equipment and, for machine packing problems involving multiple constraints such as anti-tilting constraints and mixed-packing constraints, proposes a heuristic algorithm based on multi-level progressive tree search. This algorithm can quickly solve for the optimal machine packing scheme that satisfies all constraints within a specified time limit.

[0046] This invention also constrains the placement posture of goods, requiring each type of goods to be placed in only one posture. Finally, it introduces a mixed-packing constraint that different types of goods in the same column can only be stacked vertically, but not on the same layer, striving to maximize space utilization while meeting the operational requirements of automated packing equipment. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating an embodiment of the adaptive automatic packing method that considers cargo stability according to one embodiment of the present invention.

[0048] Figure 2 It is a simple block ( Figure 2 a) and mixed blocks ( Figure 2 Schematic diagram of b);

[0049] Figure 3 This is a diagram illustrating the flat placement method;

[0050] Figure 4 This is a flowchart for generating a tile layout scheme;

[0051] Figure 5 This is a schematic diagram of the tiered placement method;

[0052] Figure 6 This is a flowchart for generating a staircase placement scheme. Detailed Implementation

[0053] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0054] The following is combined with Figures 1 to 6 This invention describes an adaptive automatic packing method that takes into account the stability of goods.

[0055] Reference Figure 1 and Figure 2 An adaptive automatic packing method considering cargo stability according to an embodiment of the present invention includes the following steps:

[0056] S1. Obtain information on the goods to be packed and the placement parameters. The goods information includes the type of goods, its own weight, the weight it can bear, and its own dimensions. The placement parameters include the placement position.

[0057] S2. Simple blocks are formed by simply grouping the same type of goods into simple blocks, and mixed blocks are formed by mixing different types of goods into mixed blocks.

[0058] S3. Based on simple blocks and mixed blocks, an adaptive selection algorithm is used to evaluate and determine whether the goods are placed in the cargo box in a flat or stepped manner.

[0059] S4. Generate a flat placement plan or a tiered placement plan based on the placement method of the goods.

[0060] The purpose of this invention is to fit as many regular boxes of different specifications as possible into a single cargo box. The goods within the box must meet various constraints, including anti-tipping and mixed-packing constraints, and the placement and orientation of all goods within the box must be determined. Due to the anti-tipping constraint, adjustments need to be made based on specific requirements when obtaining the placement plan. This can be achieved by adjusting the cargo box's height to lay the goods at similar heights to provide support between them; or by stacking the cargo as full as possible inside the box and then arranging the outermost rows in a stepped fashion to prevent the ends of the goods from tipping over due to inertia during transport.

[0061] When generating a flat placement scheme, three loading scenarios are possible: the ideal scenario where the cargo box is perfectly loaded; the scenario where the cargo box is full but not yet fully loaded; and the scenario where the cargo is loaded but there is still space remaining. When there is still space remaining after loading, the loading scheme needs to be adjusted to ensure the stability of the cargo during transportation. If the remaining space is small, rearranging the cargo boxes by adjusting their height would increase the number of unloaded goods due to limited space, contradicting the goal of maximizing cargo loading. Therefore, in this case, the present invention adopts a more flexible approach: a tiered placement method. First, goods that are later in the loading sequence are selected and placed in a tiered manner, ensuring the tier height is within an acceptable range. In this scheme, the average height of the goods is used as the preferred tier height and the upper limit of the tier. After satisfying the tiered placement requirement, the remaining goods are then laid flat based on the cargo box height.

[0062] Specifically, based on anti-tipping restrictions, it is first necessary to ensure that the space between the cargo and the sides of the cargo compartment is not too large. Secondly, the loading height of the cargo should be limited, and it should be placed flat or in a stepped manner as much as possible. Flat placement means that the height of each row of cargo from the innermost to the outermost side of the cargo box is similar and fills the entire cargo box in the depth direction. Stepped placement means that the cargo inside the cargo box fills the entire cargo box as much as possible, while the cargo near the outer side of the cargo box is placed in a stepped shape to provide support for the cargo at the end of the loading process.

[0063] See Figure 1 After inputting parameters, the goods are first grouped into blocks, and then an adaptive selection algorithm is used to evaluate the goods, choosing between flat or stepped placement to generate a placement plan. In the step of obtaining the placement plan, the algorithm calculates and compares the space utilization rate of different blocks placed into the cubic space, selecting the optimal placement method. After each placement, the remaining space information in the cargo box is updated in a timely manner, providing accurate data support for the next round of goods placement comparison and decision-making. The algorithm determines whether to enter a new loop or terminate based on whether the goods are fully loaded and whether there is still available space after placement. For the flat placement plan, the placement height of the cargo box is adjusted to ensure support between each item; for the stepped placement plan, the end steps are obtained first before placing the remaining cargo boxes.

[0064] Specifically, in step S2, the size of the block is defined by the structure of the cargo box: the loading depth of the cargo box is defined as the length of the box, the bottom width inside the cargo box is defined as the width of the box, and the height inside the cargo box is defined as the height of the box. When constructing the block, the length, width, and height of the block or cargo are parallel to the length, width, and height of the box.

[0065] When building blocks, only one column is constructed, and the pose of the same type of goods is kept consistent, and the type of goods in the same layer is kept consistent.

[0066] Methods for constructing simple blocks include:

[0067] Using a simple block as a wall block, the pose of the cargo remains unchanged when constructing the simple block. The length of the cargo is used as the length of the wall block to be formed. The maximum integer multiple that the cargo can fill in the width direction of the cargo box is used as the width of the wall block to be formed. Then, under the condition of satisfying the weight constraint, the blocks are stacked in integer multiples in the height direction to obtain wall blocks of different heights. Figure 2 'a' is a schematic diagram of a simple block of goods. Goods of the same type are constructed by placing them in the same position. The final block contains at least one layer of complete goods. The weight constraint mentioned above means that the upper layer of goods will not exceed the maximum weight that the lower layer of goods can bear.

[0068] Methods for constructing mixed-use blocks include:

[0069] Mixed blocks are obtained by combining pre-constructed simple blocks. During the combination process, constraints are applied in the width direction to control the width difference between the upper and lower layers of goods within a set range (the width difference between the upper and lower layers does not exceed one-third of the width of the upper layer to prevent tilting), while also meeting weight constraints. Figure 2 b is a schematic diagram of a mixed block, which is obtained by stacking two types of simple blocks one on top of the other, consisting of a two-layer block and a four-layer block. By assembling blocks, simple blocks and mixed blocks at different heights can be obtained, facilitating subsequent searches.

[0070] Existing block construction techniques categorize blocks into two types: blocks composed of the same type of goods and blocks composed of different types of goods. In the algorithm, the block construction method is generally combined based on the quantity of goods; when the quantity of goods is small, blocks are constructed using the same type of goods, and vice versa. The above block construction methods allow for diverse placement of goods, making them unsuitable for meeting the mixed-loading constraints of automated packing equipment. Furthermore, they do not consider whether insufficient support will cause goods to tip over after placement, making them unsuitable for loading large quantities of lightweight goods.

[0071] In this invention, to meet mixed-packing constraints, the concept of a wall is used to construct blocks. Only one column is constructed during block construction, and the orientation of goods of the same type is kept consistent, as are the types of goods on the same layer, facilitating the debugging of automated packing equipment. This solution provides two types of wall blocks: simple blocks composed of the same type of goods, and mixed blocks composed of two different types of goods. When constructing blocks, the sides of the block or goods parallel to the length, width, and height of the container correspond to the length, width, and height of the block, respectively. When constructing simple blocks, the orientation of the goods remains unchanged; the length of the goods is used as the length of the wall block to be formed, and the width of the wall block is the maximum integer multiple that the goods can fill in the width direction of the container. This improves space utilization while reducing gaps between the blocks and the sides of the container, preventing lateral tilting. Finally, while meeting weight constraints, blocks are stacked in integer multiples in the height direction to obtain wall blocks of different heights.

[0072] Regarding cargo placement, considering that all six faces of the cargo are different and that the cargo can be placed upright or upside down, theoretically there are 12 possible placement positions. However, comprehensively considering all 12 positions would significantly increase the computational load, and some placement methods may not conform to practical specifications. Therefore, in this invention, the number and specific types of cargo positions are defined by the user based on the characteristics of the specific cargo, thereby ensuring that the cargo placement positions meet practical needs while avoiding damage to the cargo during placement and transportation. The parameter definition is that when inputting the i-th placement position of the cargo, the longer side corresponding to the face that can serve as the bottom of the cargo is defined as l. i The shorter side is w i The remaining edge is h i This serves as the corresponding data for the placement posture i. When constructing the mixed-pack block, it is achieved by combining pre-constructed simple blocks. During the combination process, constraints are applied in the width direction to ensure that the width difference between the upper and lower goods is controlled within a certain range, while strictly adhering to the load-bearing weight constraints of the goods, thereby preventing problems such as tipping or crushing of the goods during transportation.

[0073] The mixed loading constraint in this invention specifically means that the stacking of different types of goods is only allowed to occur on different layers of goods in the same column, with goods on the inside of the carriage at the bottom and goods on the outside at the top, so as to facilitate the stacking of goods by automated packing equipment.

[0074] The flat and stepped placement schemes are suitable for different types of goods. The flat placement scheme is more suitable for situations where goods are of similar height and numerous. When adjusting the height of the cargo boxes, it can, to some extent, avoid situations where the number of loaded goods is significantly reduced or where a feasible solution cannot be found due to excessive height differences. The stepped placement scheme is suitable for situations where goods are larger and relatively heavier. With larger goods, fewer columns need to be adjusted, the algorithm is faster, and heavier goods have greater inertia and are more stable, not necessarily requiring complete front and rear support to prevent displacement. Since goods produced by the same manufacturer have similar weights, and volume is proportional to weight, the influence of weight is considered by weighing the volume of the goods. Furthermore, the order quantity factor must be considered. Although multiple target locations for the same task order are relatively close to other task locations, tasks with a large number of orders, after partial disassembly, will result in unsupported loading. When the goods are not tightly packed together, the higher the stack, the less stable the center of gravity, and the more prone it is to tipping over during transportation. Therefore, it is necessary to choose flat placement to limit the height, and to weigh the unsupported transportation distance and the lateral loading of the blocks. Thus, in one embodiment of the present invention, the adaptive selection algorithm makes a selection based on information such as the height and volume of the goods, the lateral loading rate of the blocks, and the order quantity.

[0075] Specifically, in step S3, an adaptive selection algorithm is used to evaluate and determine whether the goods are placed flat or in a stepped manner within the cargo container, including:

[0076] The cargo's attribute trade-off parameters are defined based on its height and volume;

[0077] The stability parameters are calculated based on the block construction results: the stability parameters are approximately calculated by the lateral loading rate of the block, which is represented by the ratio between the length of the cargo block in the width direction of the cargo box and the width of the cargo box.

[0078] Distance trade-off parameters are derived by considering the distance between the first and last orders, the total transportation distance, and the number of orders.

[0079] The decision score is obtained by weighting and mapping the attribute trade-off parameters, stability parameters, and distance trade-off parameters.

[0080] Choose between a flat or tiered placement method based on the decision score.

[0081] Furthermore, the method for defining the attribute trade-off parameters of cargo based on cargo height and volume is as follows:

[0082] Let the height of the i-th item be h. i , i∈[0,n), volume is vi For i ∈ [0, n), the average height of all goods is Average volume is The formula for calculating the attribute tradeoff parameter α is as follows: Among them, c min c represents the maximum number of chunk layers for the smallest cargo. step This indicates the maximum number of tiers in the last order. The first part of the attribute tradeoff parameter calculation formula obtains the height difference parameter of the goods through the height difference. The smaller the height difference, the closer the result is to 0, which tends to choose flat placement. The larger the difference, the closer it is to 1. Since most goods have similar volumes, the second part of the attribute tradeoff parameter calculation formula obtains the volume parameter by comparing the maximum number of tiers in the last order with the maximum number of block layers in the smallest goods. The larger this value is, the closer it is to 1, indicating that the repeated calculation time used for tiered placement is shorter, and the more inclined it is to choose tiered placement.

[0083] The formula for calculating the stability parameter β is: Where, β ij J represents the lateral loading rate of cargo i in the j-th pose; i N represents the pose contained in cargo i; ij The total number of poses of the goods is represented by θ, which is the threshold.

[0084] Lateral Loading Rate β ij The calculation formula is: Where W represents the width of the cargo box; w ij Indicates the width of cargo i in the j-th pose; n ij This represents the maximum loading quantity of cargo i in the j-th pose along the width direction of the cargo box; this value is obtained from the block construction results. The smaller the β value, the closer the loading situation is to the minimum loading rate, and the more the scheme tends to choose flat placement. Since the lateral loading rate can exceed 80%, the threshold θ is taken as 0.8 in this invention.

[0085] The formula for calculating the distance tradeoff parameter γ is: Where m is the number of orders, L is the total transportation distance of the orders, and l is the distance to the destination of the first order; the loading scheme is selected by comparing the proportion of the transportation distance before the first order is unloaded. The smaller the proportion of the distance before the first order is unloaded, the more inclined to use flat placement.

[0086] The formula for calculating the decision score Z is: Z = α + β + 2γ, Z ∈ [0, 5]. When the decision score is greater than 2.5, the tiered placement method is selected, and when the decision score is less than or equal to 2.5, the flat placement method is selected.

[0087] Reference Figure 4 In one embodiment of the present invention, step S4, the method for generating a flat placement scheme includes:

[0088] Using the cargo box as the loading space, and under the constraints of weight and loading sequence, a greedy strategy is adopted to select and load blocks with the goal of maximizing space utilization to obtain a preliminary loading plan.

[0089] Based on the preliminary loading plan, determine whether all the goods to be loaded can be loaded and obtain the remaining space in the length direction of the cargo box;

[0090] If there is remaining space along the length of the cargo box, adjust the loading space according to the size of the remaining space, reduce the loading height, and ensure that the cargo can be supported from the front and back after loading; re-obtain the placement plan until the plan meets the flat loading constraint;

[0091] If a tiling plan cannot be obtained within the set time, the current plan will be output along with the remaining goods, allowing the user to make adjustments.

[0092] Specifically, when laying out the blocks, a greedy strategy is used to obtain the initial placement scheme that maximizes space utilization while satisfying weight and loading sequence constraints. Different blocks are selected and placed into the remaining space according to the order sequence, and their space utilization is compared to choose the optimal block. During placement, situations where blocks are not complete need to be handled. That is, when placing mixed-load columns, if the remaining quantity of goods inside is insufficient to complete the column, goods can be removed from the top layer of the column to fill the remaining layer, allowing placement to continue until all goods are placed or there is insufficient space. This adjustment of removing items from the top and adding them to the bottom ensures that all goods are loaded without significantly altering the upper layers.

[0093] When generating a flat placement scheme, after iteratively selecting blocks for the goods, the resulting initial loading method maximizes the utilization of the container's internal space. In this case, three loading scenarios are possible: an ideal situation where the container is fully loaded with goods; a situation where the container is full but the goods are not yet fully loaded; and a situation where the goods are fully loaded but there is still remaining space. In the second scenario, maximum space utilization is achieved, and the amount of remaining goods is relatively small; therefore, it is marked as a feasible loading scheme in this invention.

[0094] If there is a large remaining space, the cargo may tip over during transportation due to the excess space and the cargo's inertia. To avoid this, this algorithm evaluates the initial loading plan, adjusts the cargo container's loading height, and recalculates the loading plan to search for a better loading scheme that provides support between the cargo. Specifically, the method for re-obtaining the placement plan until it satisfies the flat loading constraint includes:

[0095] Let the new loading height be h1 and the original loading height be the upper limit h2. Under the constraint of the h1 height, we will use a greedy strategy to search again to obtain the placement solution.

[0096] If the obtained placement scheme can achieve flat placement of goods, the result is output. If there is still remaining space for placing goods, h1 and h2 are updated again and a greedy strategy is used for searching. If the goods cannot be placed in the end or there is less than one layer left, they are stacked directly. Otherwise, h1 and h2 are updated, with (h1+h2) / 2 as h1, and the upper limit height h2 remains unchanged, until a placement scheme is obtained or there is no more search time.

[0097] like Figure 3 As shown, although there is still considerable remaining space at the top of the cargo box, it will be discarded to meet the anti-tipping constraint. The optimal loading height of the cargo box needs to be obtained through iteration.

[0098] Reference Figure 6 In one embodiment of the present invention, step S4, the method for generating a staircase placement scheme, includes:

[0099] Goods in adjacent columns must meet a height difference constraint, where the height difference must be less than or equal to the height of the smaller good. Furthermore, when placing items in a hierarchy, the basic constraint that later items should be placed on top must still be met. Based on this, the good with the highest space utilization is selected for placement; as follows:

[0100] First, place the goods with the highest space utilization in the last order in a staircase. Use the wall block with the highest space utilization as the pose and the number of layers as the maximum number of layers m of the current staircase. Start placing the goods on the first layer and make them touch the door of the cargo box. Then, place them on the second or third layer until the maximum number of layers is reached. If the goods cannot be placed, add them to the remaining goods in the order.

[0101] After the tiered arrangement is completed, the remaining goods are laid out flat based on the height of the cargo box.

[0102] If the space outside the steps cannot be fully loaded, i.e. the height difference between the steps and the space is too large, then the steps should be adjusted: reduce the maximum number of steps (m) and then place and load the remaining goods on the steps.

[0103] By placing the goods in the last few rows in a manner different from laying them flat, a mutually supporting structure is formed between the goods and between the goods and the containers, effectively improving the overall stability of the cargo and preventing it from tipping over due to shaking or bumping during transportation. Figure 5As shown, the last column of goods laid flat is far from the cargo door and cannot support the goods in the inner columns. Directly adjusting the height of the cargo box may reduce the amount of goods loaded. Therefore, the last few columns were adjusted to be placed in a stepped manner to provide support.

[0104] Other configurations and operations of an adaptive automatic packing method considering cargo stability according to an embodiment of the present invention are known to those skilled in the art and will not be described in detail here.

[0105] In the description of this specification, references to terms such as "embodiment," "example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0106] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. An adaptive auto-boxing method considering stability of goods, characterized in that, The method comprises the following steps: S1, obtaining information of goods to be packed and placement mode parameters, the information of goods including the type of goods, the weight of the goods, the loadable weight of the goods and the size of the goods, and the placement mode parameters including a placement pose; S2, simple blocks are formed for the same type of goods, and mixed blocks are formed for different types of goods; S3, based on the simple blocks and the mixed blocks, an adaptive selection algorithm is used for evaluation to determine whether the placement of the goods in the container is in a flat placement mode or a stepped placement mode; S4, a flat placement scheme or a stepped placement mode is generated according to the placement mode of the goods; In the step S3, the adaptive selection algorithm is used for evaluation to determine whether the placement of the goods in the container is in a flat placement mode or a stepped placement mode, comprising: Defining the attribute trade-off parameter of the goods according to the height and volume of the goods; Calculating the stability parameter according to the results of block construction: the stability parameter is approximately calculated through the transverse loading rate of the block, and the transverse loading rate is represented by the ratio between the length of the goods block in the width direction of the container and the width of the container; Distance trade-off is performed according to the distance between the first order and the last order, the total transportation distance and the number of orders to obtain a distance trade-off parameter; The attribute trade-off parameter, the stability parameter and the distance trade-off parameter are weighted and mapped to obtain a decision score; The flat placement mode or the stepped placement mode is selected according to the decision score; The method for defining the attribute trade-off parameter of the goods according to the height and volume of the goods is: The height of the first cargo is , the volume is , the average height of all cargoes is , and the average volume is ; the calculation formula of the attribute trade-off parameter is as follows: wherein, represents the maximum number of steps of the smallest goods, represents the maximum number of steps of the last order; Stability parameter The calculation formula is: wherein, denotes the goods The first transverse loading rate of the pose; denotes the goods contains the pose; denotes the total number of poses of the goods, is a threshold value; Transverse loading rate The calculation formula is: wherein W represents the width of the container; represents the width of the cargo in the first position; represents the maximum loading number of the cargo in the first position in the width direction of the container; Distance trade-off parameter The calculation formula is: Wherein, m is the order number, L is the total distance of order transportation, is the distance from the first order target location Decision score The calculation formula is: , When the decision score is greater than 2.5, the ladder placement mode is selected, and when the decision score is less than or equal to 2.5, the tile placement mode is selected.

2. The method of claim 1, wherein, In the step S2, the size of the block is defined according to the structure of the container: the loading depth of the container is defined as the length of the box, the width of the bottom in the container is defined as the width of the box, and the height in the container is defined as the height of the box. When constructing the block, the length, width and height of the block or the goods are parallel to the length, width and height of the box; When constructing the block, only one column is constructed, and the poses of the same type of goods are kept consistent, and the types of goods in the same layer are kept consistent.

3. The method of claim 2, wherein, In the step S2, the method for constructing the simple block comprises: The simple block is used as a kind of wall block. When constructing the simple block, the pose of the goods is kept unchanged, the length of the goods is used as the length of the wall block to be formed, and the maximum integer multiple of the width of the goods that can be filled in the width direction of the container is used as the width of the wall block to be formed. Then, under the condition of satisfying the weight constraint, the integer multiple of the height is stacked in the height direction to obtain wall blocks of different heights; The method for constructing the mixed block comprises: The mixed block is obtained by combining the constructed simple blocks. In the combination process, a constraint is applied in the width direction to control the width difference between the upper and lower layers of goods within a set range, while satisfying the weight constraint.

4. The method according to any one of claims 1 to 3, characterized in that, In the step S4, the method for generating the flat placement scheme comprises: The container is used as the loading space. Under the condition of satisfying the weight and loading sequence constraints, a greedy strategy with the maximum space utilization rate as the target is used to select and load the blocks to obtain a preliminary loading scheme; Based on the preliminary loading scheme, it is determined whether the to-be-loaded goods can be completely loaded, and the remaining space in the length direction of the container is obtained; If there is a remaining space in the length direction of the container, the loading space is adjusted according to the size of the remaining space, the loadable height is reduced, and the goods can be loaded to realize front and rear support after loading; the placement scheme is reacquired to satisfy the flat loading constraint; If a flat scheme cannot be obtained within a set time, the current scheme is output and the remaining goods are output.

5. The method of claim 4, wherein, The method for reacquiring the placement scheme to satisfy the flat loading constraint includes: Let the new loading height be The original loading height is the upper limit. ,exist Under high constraints, a greedy strategy is used again to search for placement solutions; If the obtained placement scheme can realize the goods to meet the flat placement, the result output is performed, and if there is still remaining space for goods placement, the update is performed again With And search is performed by using the greedy strategy; if the goods cannot be placed finally or only insufficient layers are left, direct stacking is performed, otherwise, update is performed With To As Upper limit height Until the placement scheme is obtained or there is no search time.

6. The method of claim 4, wherein, In step S4, the method for generating the step placement scheme includes: The goods of adjacent columns need to satisfy the height difference constraint, the height difference needs to be less than or equal to the height of the smaller goods, and the basic constraint that the goods in the rear sequence are on the top still needs to be satisfied when performing hierarchical placement, and on this basis, the goods with the largest space utilization are selected for placement; as follows: First, the goods with the largest space utilization in the last order are placed in steps, the wall block with the largest space utilization is used as the pose, the maximum number of layers m of the current step is used as the number of layers, the placement starts from one layer, the goods are placed against the container door, and then two layers, three layers are placed, until the maximum number of layers is reached, if the goods cannot be placed, the remaining goods in the order are added; After step placement, the remaining goods are placed in flat mode based on the height of the container; If the space other than the step placement cannot be fully loaded, that is, the height difference between the steps is too large, the step goods are adjusted: the maximum number of layers m of the step is reduced, and then the step placement and the remaining goods are loaded.

Images