Intelligent order splitting method and system for furniture production line

Abstract

The present invention provides a furniture production line intelligent order distribution method and system, which relates to the field of intelligent control technology. The method comprises: for each cluster, calculating the average value of all orders in it, and setting the corresponding average value as a new cluster center, repeating the steps until a preset number of iterations is reached to obtain K clusters, and taking the K clusters as the final order classification results, wherein each cluster corresponds to a specific production line; in each category, sorting the orders according to the delivery date to obtain the sorting priority; determining the dynamic adjustment factors corresponding to each order according to the sorting priority; and intelligently allocating orders to each production line according to the orders of each category, the production capacity of the production line and the dynamic adjustment factor. The present invention reduces the time for production line conversion and adjustment, thereby improving production efficiency.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an intelligent order splitting method and system for a furniture production line. Background Art

[0002] The traditional method of order splitting in furniture production lines still relies on manual operations and relatively basic data processing. Although this method may have met the needs in the past, its efficiency and accuracy are gradually showing its shortcomings when facing today's large-scale and diversified orders.

[0003] Specifically, when dealing with massive amounts of order data, this method usually just mechanically distributes orders in the order in which they are received or follows certain fixed distribution principles. However, this approach ignores the actual carrying capacity of the production line, the urgency of the orders, and the correlation between different orders. As a result, the production line may face the problem of unbalanced load, which in turn affects the overall production efficiency. What's more serious is that due to the significant differences between orders, the production line may need to be adjusted frequently, which not only increases production costs, but also increases the possibility of errors. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent order splitting method and system for a furniture production line, so that the same production line can focus on producing similar types of furniture, reducing the time for production line conversion and adjustment, thereby improving production efficiency.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a method for intelligent order splitting of a furniture production line is provided, the method comprising:

[0007] Receive order information;

[0008] Determine the number of clusters K, and randomly determine K orders from the order information as the initial cluster centers; for each order, calculate its distance from each cluster center and assign it to the category where the corresponding cluster center is located; for each cluster, calculate the average value of all orders in it, and set the corresponding average value as the new cluster center, repeat the steps until the preset number of iterations is reached to obtain K clusters, and use the K clusters as the final order classification results, where each cluster corresponds to a specific production line;

[0009] In each category, the orders are sorted according to the delivery date to obtain the sorting priority;

[0010] Determine the dynamic adjustment factor corresponding to each order according to the sorting priority;

[0011] Orders are intelligently allocated to each production line based on each category of orders, the production capacity of the production line, and dynamic adjustment factors.

[0012] Furthermore, for each order, the distance between it and each cluster center is calculated, including:

[0013] For each order, according to Calculate its distance from each cluster center;

[0014] in, Represents order x and cluster center c k The distance between ij is an element of a weight matrix, indicating the importance of the interaction between the i-th feature and the j-th feature; σ is the bandwidth parameter of the Gaussian kernel function; x i and c k，i They are order x and cluster center c respectively k The value in the i-th dimension; n is the number of dimensions of orders and cluster centers; x j and c k，j They are order x and cluster center c respectively k The value in the jth dimension; i and j represent index variables, the value range of i is 1 to n, and the value range of j is 1 to n.

[0015] Furthermore, the element w of the weight matrix ij The calculation formula is:

[0016]

[0017] Among them, x kl and x kj are the values ​​of the k-th order on the l-th feature and the j-th feature respectively; and are the average values ​​of all orders on the i-th feature and the j-th feature respectively; is the average of all orders on the lth feature; m is the total number of orders; l is the feature index, which is used to traverse all features during the summation process; x ki is the value of the k-th order on the i-th feature; k is an index variable used to traverse all orders in the data set, and the value range of k is 1 to m.

[0018] Furthermore, in each category, the orders are sorted according to the delivery date to obtain the sorting priority, including:

[0019] For each category of orders, multiple sorting sequences are randomly generated as the initial population, and each individual represents a sorting method for the corresponding category of orders;

[0020] Define an evaluation function for evaluating the quality of each individual;

[0021] Calculate the fitness value of each individual according to the evaluation function, randomly determine the corresponding individual, and perform a crossover operation to generate new offspring;

[0022] The newly generated offspring are randomly mutated, and the selection, crossover and mutation operations are repeated until the preset number of iterations is reached to determine the final individual, and the final individual is used as the sorting result of the category order;

[0023] Based on the sorting results, each order is assigned a sorting priority.

[0024] Furthermore, the calculation formula of the evaluation function is:

[0025]

[0026] Where F(y) is the fitness value of individual y; b is the number of orders; d i represents the difference in days from the base date to the delivery date of each order; ∈ is a positive number; Q i is the importance weight of the i-th order.

[0027] Furthermore, the dynamic adjustment factors corresponding to each order are determined according to the sorting priority, including:

[0028]

[0029] Among them, DA i Represents the dynamic adjustment factor of the i-th order; N i Indicates the number of products contained in the i-th order; M i represents the number of steps in the production process of the i-th order; R i represents the number of types of raw materials required for the i-th order; θ is an adjustment parameter; α is an adjustment parameter for sorting priority; P i Indicates the sorting priority of the i-th order.

[0030] Furthermore, the order information includes furniture type, size, color, quantity, and delivery date.

[0031] In the second aspect, an intelligent order splitting system for a furniture production line includes:

[0032] An acquisition module is used to receive order information; determine the number of clusters K, and randomly determine K orders from the order information as initial cluster centers; for each order, calculate its distance from each cluster center, and assign it to the category where the corresponding cluster center is located; for each cluster, calculate the average value of all orders in it, and set the corresponding average value as the new cluster center, repeat the steps until the preset number of iterations is reached to obtain K clusters, and use the K clusters as the final order classification results, wherein each cluster corresponds to a specific production line;

[0033] The processing module is used to sort the orders in each category according to the delivery date to obtain the sorting priority; determine the dynamic adjustment factor corresponding to each order according to the sorting priority; and intelligently allocate the orders to each production line according to the orders of each category, the production capacity of the production line and the dynamic adjustment factor.

[0034] According to a third aspect, a computing device includes:

[0035] one or more processors;

[0036] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0037] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0038] The above scheme of the present invention includes at least the following beneficial effects.

[0039] Through cluster analysis, this method can intelligently classify orders so that similar or close orders are clustered together. In this way, the same production line can focus on producing similar types of furniture, reducing the time for production line conversion and adjustment, thereby improving production efficiency. Intelligently allocating orders to various production lines can dynamically adjust order allocation according to the actual production capacity of the production line to ensure balanced load on each production line. This helps avoid situations where some production lines are overloaded while others are idle, thereby optimizing the allocation of production resources.

[0040] By sorting orders by delivery date, this method can ensure that urgent orders are given priority and reduce the risk of delayed delivery. By setting dynamic adjustment factors, order allocation is more flexible and can be dynamically adjusted according to the real-time situation of the production line and order priority, further improving production efficiency and response speed. Traditional order allocation often relies on manual judgment and experience, which is prone to errors. This intelligent order allocation method automatically allocates orders through algorithms, reducing the possibility of human errors and improving the accuracy and efficiency of allocation. The intelligent order allocation system can provide real-time production data and reports to help management better monitor production progress and conduct decision analysis, thereby improving overall management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of an intelligent order splitting method for a furniture production line provided by an embodiment of the present invention.

[0042] Figure 2 It is a schematic diagram of an intelligent order splitting system for a furniture production line provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0044] like Figure 1 As shown, an embodiment of the present invention provides a method for intelligent order splitting of a furniture production line, the method comprising the following steps:

[0045] Step 11, receiving order information, the order information includes furniture type, size, color, quantity and delivery date;

[0046] Step 12, determine the number of clusters K, randomly determine K orders from the order information as the initial cluster centers; for each order, calculate its distance from each cluster center, and assign it to the category where the corresponding cluster center is located; for each cluster, calculate the average value of all orders in it, and set the corresponding average value as the new cluster center, repeat the steps until the preset number of iterations is reached to obtain K clusters, and use the K clusters as the final order classification results, where each cluster corresponds to a specific production line;

[0047] Step 13, in each category, sort the orders according to the delivery date to obtain the sorting priority;

[0048] Step 14, determining the dynamic adjustment factor corresponding to each order according to the sorting priority;

[0049] Step 15, intelligently allocate orders to each production line based on each category of orders, the production capacity of the production line, and the dynamic adjustment factor.

[0050] In an embodiment of the present invention, through cluster analysis, the method can intelligently classify orders so that similar or similar orders are clustered together. In this way, the same production line can focus on producing similar types of furniture, reducing the time for production line conversion and adjustment, thereby improving production efficiency. Intelligently allocating orders to each production line can dynamically adjust order allocation according to the actual production capacity of the production line to ensure load balance of each production line. This helps to avoid the situation where some production lines are overloaded while other production lines are idle, thereby optimizing the configuration of production resources. By sorting orders according to delivery dates, the method can ensure that urgent orders are given priority and reduce the risk of delayed delivery. By setting a dynamic adjustment factor, order allocation is made more flexible and can be dynamically adjusted according to the real-time situation of the production line and the order priority, further improving production efficiency and response speed. Traditional order allocation often relies on manual judgment and experience and is prone to errors. This intelligent order distribution method automatically allocates orders through algorithms, reduces the possibility of human errors, and improves the accuracy and efficiency of allocation. The intelligent order distribution system can provide real-time production data and reports to help management better monitor production progress and make decision analysis, thereby improving overall management efficiency.

[0051] In another preferred embodiment of the present invention, when specifically applied, it may include:

[0052] Clean and standardize order data to ensure data quality and consistency, extract key features of orders, such as size, complexity, etc., and convert them into numerical data; set the K value according to the number of production lines; randomly select K orders from the processed order data as initial cluster centers; use a loop structure to iterate, and in each iteration, calculate the distance from each order to each cluster center and classify it into the nearest cluster. For each cluster, calculate the average feature value of all orders inside and update the cluster center; check whether the number of iterations has been reached or whether the change in the cluster center is less than the preset threshold to determine whether to continue iterating; after the iteration is completed, output the order list of each cluster, and correspond each cluster to a specific production line based on the characteristics and capabilities of the production line; verify and optimize the classification results to ensure the accuracy and efficiency of the classification, and integrate the classification results with the production management system to automatically assign orders to the corresponding production lines.

[0053] In a preferred embodiment of the present invention, for each order, the distance between the order and each cluster center is calculated, including:

[0054] For each order, according to Calculate its distance from each cluster center;

[0055] in, Represents order x and cluster center c k The distance between ij is an element of a weight matrix, indicating the importance of the interaction between the i-th feature and the j-th feature; σ is the bandwidth parameter of the Gaussian kernel function; x i and c ki They are order x and cluster center c respectively k The value in the i-th dimension; n is the number of dimensions of orders and cluster centers; x j and c k，j They are order x and cluster center c respectively k The value in the jth dimension; i and j represent index variables, the value range of i is 1 to n, and the value range of j is 1 to n.

[0056] In the embodiment of the present invention, the distance formula is introduced by introducing the weight matrix w ij , which can take into account the interaction between different features. In traditional clustering algorithms, the relationship between features is often ignored, while this method can more comprehensively capture the complexity of order data, thereby improving the accuracy of clustering. Through the Gaussian kernel function, this formula can take into account the local similarity of features. σ (the bandwidth parameter of the Gaussian kernel function) controls the range of similarity, making it more likely that similar orders will be grouped into the same category during clustering. This helps to discover more subtle structures in the data set. The formula comprehensively evaluates the similarity between orders and cluster centers by considering the differences in all dimensions. This multi-dimensional fusion method makes the clustering results more robust and less susceptible to single feature outliers. By adjusting the weight matrix w ij and the bandwidth parameter σ of the Gaussian kernel function, this method can adapt to the characteristics of different data sets and provide a more flexible clustering solution. At the same time, the adjustment of these parameters also helps to understand which features play a key role in the clustering process. Since the interactivity and local similarity between features are taken into account, this method is expected to improve the quality of clustering, making orders within the same cluster more similar and orders between different clusters more different. In summary, using this improved distance formula for intelligent order splitting of furniture production lines can more accurately cluster similar orders together, thereby optimizing the production process, improving production efficiency, reducing production costs, and improving customer satisfaction.

[0057] In a preferred embodiment of the present invention, the element w of the weight matrixij The calculation formula is:

[0058]

[0059] Among them, x kl and x kj are the values ​​of the k-th order on the l-th feature and the j-th feature respectively; and are the average values ​​of all orders on the i-th feature and the j-th feature respectively; is the average of all orders on the lth feature; m is the total number of orders; l is the feature index, which is used to traverse all features during the summation process; x ki is the value of the k-th order on the i-th feature; k is an index variable used to traverse all orders in the data set, and the value range of k is 1 to m.

[0060] In an embodiment of the present invention, by calculating the correlation coefficient between features, the linear relationship between different features can be quantified, which helps to understand which features are interrelated, so as to better utilize these related information in the clustering process. It helps to improve the accuracy of clustering according to the correlation between features. If two features are highly correlated, then in the clustering process, the influence of these two features will be reasonably weighed to avoid clustering bias caused by repeated consideration of similar information. By considering the correlation between all features, the model can more comprehensively utilize the information in the data set, thereby enhancing robustness. By calculating the correlation between features, it is possible to identify which features are redundant or irrelevant.

[0061] In a preferred embodiment of the present invention, the above step 13, in each category, sorting the orders according to the delivery date to obtain the sorting priority, may include:

[0062] Step 131, for each category of orders, a plurality of sorting sequences are randomly generated as an initial population, and each individual represents a sorting method of the corresponding category of orders;

[0063] Step 132, defining an evaluation function for evaluating the quality of each individual;

[0064] Step 133, calculating the fitness value of each individual according to the evaluation function, randomly determining the corresponding individual, and performing a crossover operation to generate a new offspring;

[0065] Step 134, randomly mutate the newly generated offspring, repeat the selection, crossover and mutation operations until a preset number of iterations is reached to determine the final individual, and use the final individual as the sorting result of the category order;

[0066] Step 135, assigning a sorting priority to each order according to the sorting result.

[0067] In the embodiment of the present invention, the present invention is able to search for the optimal solution in a global scope, rather than being limited to the local optimum. This means that by sorting orders in this way, it is more likely to find a globally optimal sorting method, thereby managing the production and delivery of orders more effectively. The evaluation function can adapt to different business scenarios and needs; through the automated genetic algorithm process, the need for manual intervention can be greatly reduced, and the efficiency and accuracy of order sorting can be improved. In addition, since the algorithm can run automatically in the background, the order sorting can be updated in real time to adapt to the changing order situation. The genetic algorithm is insensitive to the selection of the initial population, so even if the initial sorting is not optimal, the algorithm can gradually approach the optimal solution through iteration. This robustness makes the method more reliable in practical applications. The method can be easily extended to the situation of handling a large number of orders. By increasing the population size and the number of iterations, more complex sorting problems can be handled while maintaining the effectiveness of the algorithm. The order sorting results determined by the genetic algorithm can assign a reasonable sorting priority to each order. This priority not only takes into account the delivery date, but may also take into account other business factors, thereby ensuring the smooth operation of the production line and maximizing customer satisfaction.

[0068] When applied specifically, the following steps are included:

[0069] Determine the size of the initial population, that is, how many different sorting sequences are to be generated as the starting point; for each category of orders, randomly generate a specified number of sorting sequences, each of which is a random arrangement of the category of orders, representing a possible delivery order; store these randomly generated sorting sequences as the initial population. The evaluation function is used to measure the quality of each sorting sequence. In the context of delivery date, the evaluation function may consider factors such as the urgency of the order delivery date and the time interval between orders. The evaluation function can calculate the sum of the difference between the delivery date of all orders in the sorting sequence and the current date. The smaller the difference, the higher the evaluation value. Use the evaluation function defined in step 132 to calculate the fitness value of each individual in the initial population; based on the fitness value, randomly select two individuals as parents for crossover operation, and the probability of individuals with high fitness being selected should be higher. Generate new offspring through crossover operations (such as single-point crossover, multi-point crossover, etc.). The crossover point can be randomly selected to ensure the diversity of the offspring. Randomly mutate the offspring with a certain mutation probability. Mutation can include operations such as exchanging the position of orders, inserting or deleting orders to increase the diversity of the population; adding the newly generated offspring to the population, and repeating the operations of selection, crossover and mutation until the preset number of iterations is reached. After the number of iterations is reached, the individual with the highest fitness value is selected as the final sorting result of the order of this category.

[0070] Step 135, assign a sorting priority to each order according to the final sorting result determined in step 134. For example, the order ranked first has the highest priority, and so on. The sorting priority information of each order is stored or output to the relevant production management system.

[0071] In a preferred embodiment of the present invention, the calculation formula of the evaluation function is:

[0072]

[0073] Where F(y) is the fitness value of individual y; b is the number of orders; d i represents the difference in days from the base date to the delivery date of each order; ∈ is a positive number; Q i is the importance weight of the i-th order.

[0074] In the embodiment of the present invention, by setting the importance weight Qi of each order, the evaluation function can flexibly adapt to different business scenarios. For example, certain high-value orders can be given higher weights, so that they can obtain higher priority in sorting. i+1 -d iThe +∈ part reflects the urgency of order delivery. When the delivery dates of two adjacent orders are very close, this value will increase, which means that these orders need to be given more attention when sorting to avoid delivery delays. ∈ , the evaluation function avoids the division by zero error that occurs when two orders have the same delivery date, increasing the robustness and stability of the function. The evaluation function takes into account the optimality of all orders, not just a single order. This helps find a globally optimal sorting solution, rather than a local optimum.

[0075] In a preferred embodiment of the present invention, the dynamic adjustment factors corresponding to each order are determined according to the sorting priority, including:

[0076]

[0077] Among them, DA i Represents the dynamic adjustment factor of the i-th order; N i Indicates the number of products contained in the i-th order; M i represents the number of steps in the production process of the i-th order; R i represents the number of types of raw materials required for the i-th order; θ is an adjustment parameter; α is an adjustment parameter for sorting priority; P i Indicates the sorting priority of the i-th order.

[0078] In the embodiment of the present invention, by introducing a dynamic adjustment factor, each order can be managed more finely. i It reflects the complexity and resource requirements of each order in the production process. The θ and α parameters in the formula provide flexibility, and the calculation method of the dynamic adjustment factor can be adjusted according to actual conditions. For example, when the production pressure is high, the dynamic adjustment factor of all orders can be increased or decreased by adjusting θ to adapt to production needs. i 、M i and R i They represent the number of products, the number of production process steps, and the number of raw material types of the order, respectively. By incorporating them into the calculation of the dynamic adjustment factor, the actual production difficulty of each order can be more accurately reflected. i The dynamic adjustment factor can reflect the urgency and importance of the order. Orders with higher priority will receive higher dynamic adjustment factors.

[0079] In another preferred embodiment of the present invention, the above step 15 may include:

[0080] Summarize order information for each category, including order quantity, product type, size, delivery date, and dynamic adjustment factor; obtain production capacity data for each production line, such as the number of products that can be processed per hour / day, and the characteristics of the production line (product types and size ranges suitable for production, etc.); formulate an allocation strategy by comprehensively considering the order sorting priority, dynamic adjustment factors, and production capacity of the production line; orders with high priority and large dynamic adjustment factors should be given priority to be allocated to production lines with matching production capacity.

[0081] Execute the smart allocation algorithm:

[0082] Initialize an empty allocation plan list, sort the orders of each category according to the sorting priority and dynamic adjustment factor, traverse the sorted order list, and for each order:

[0083] Check the current load and remaining production capacity of each production line; select the most suitable production line based on the production capacity of the production line, the matching degree of product type and the dynamic adjustment factor of the order; allocate the order to the selected production line, update the load information of the production line, and record the allocation result in the allocation plan list.

[0084] Verify the generated allocation plan to ensure that the delivery requirements of all orders are met and the load of each production line is within the acceptable range; if necessary, fine-tune the allocation plan, such as optimizing overall efficiency by exchanging the allocation of some orders; send the final allocation plan to each production line to start production; monitor the production progress in real time to ensure that it proceeds as planned, and adjust the allocation plan in time to deal with emergencies.

[0085] For example, suppose a furniture manufacturing company has three production lines (A, B, C), each with different production capacity and specializing in producing different types and sizes of furniture. Now the company receives three types of orders (X, Y, Z), each containing products with different quantities, sizes, and delivery dates.

[0086] Order X: 50 large sofas, delivery date is 15 days later, dynamic adjustment factor is 1.8.

[0087] Order Y: 100 medium-sized desks, delivery date is 10 days later, and the dynamic adjustment factor is 1.5.

[0088] Order Z: 200 small chairs, delivery date 20 days later, dynamic adjustment factor 1.2.

[0089] Production line A: specializes in producing large furniture, with a daily production capacity of 20 pieces.

[0090] Production line B: suitable for the production of medium-sized furniture, with a daily production capacity of 30 pieces.

[0091] Production line C: can produce all types of furniture, but the efficiency is slightly lower, with a daily production capacity of 25 pieces.

[0092] According to the sorting priority (taking into account the delivery date and dynamic adjustment factor), order Y (the most urgent and with a higher adjustment factor) is processed first, followed by order X, and finally order Z; order Y is assigned to production line B, which takes about 3.3 days to complete (100 pieces / 30 pieces / day, rounded up); order X is assigned to production line A, which takes 2.5 days to complete (50 pieces / 20 pieces / day, rounded up); order Z is assigned to production line C, which takes 8 days to complete (200 pieces / 25 pieces / day); the allocation plan is verified to ensure that the production line load is reasonable and all orders can be completed before the delivery date, and the allocation plan is notified to each production line to start production.

[0093] like Figure 2 As shown, an embodiment of the present invention further provides an intelligent order distribution system 20 for a furniture production line, comprising:

[0094] The acquisition module 21 is used to receive order information; determine the number of clusters K, and randomly determine K orders from the order information as initial cluster centers; for each order, calculate its distance from each cluster center, and assign it to the category where the corresponding cluster center is located; for each cluster, calculate the average value of all orders in it, and set the corresponding average value as the new cluster center, repeat the steps until a preset number of iterations is reached to obtain K clusters, and use the K clusters as the final order classification results, wherein each cluster corresponds to a specific production line;

[0095] The processing module 22 is used to sort the orders in each category according to the delivery date to obtain the sorting priority; determine the dynamic adjustment factor corresponding to each order according to the sorting priority; and intelligently allocate the orders to each production line according to the orders of each category, the production capacity of the production line and the dynamic adjustment factor.

[0096] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0097] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0098] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

Claims

1. A method for intelligent order splitting in a furniture production line, characterized in that: The method comprises the following steps: Receive order information; Determine the number of clusters K, and randomly determine K orders from the order information as the initial cluster centers; for each order, calculate its distance from each cluster center and assign it to the category where the corresponding cluster center is located; for each cluster, calculate the average value of all orders in it, and set the corresponding average value as the new cluster center, repeat the steps until the preset number of iterations is reached to obtain K clusters, and use the K clusters as the final order classification results, where each cluster corresponds to a specific production line; In each category, the orders are sorted according to the delivery date to obtain the sorting priority; Determine the dynamic adjustment factor corresponding to each order according to the sorting priority; Intelligently distribute orders to each production line based on each category of orders, production line capacity, and dynamic adjustment factors; for each order, calculate its distance to each cluster center, including: For each order, according to Calculate its distance from each cluster center; in, Indicates order With cluster center The distance between is an element of a weight matrix, indicating the importance of the interaction between the i-th feature and the j-th feature; is the bandwidth parameter of the Gaussian kernel function; Orders With cluster center The value in the i-th dimension; n is the number of dimensions of orders and cluster centers; Orders With cluster center In the The values ​​in the dimensions; i and j represent index variables, the value range of i is 1 to n, and the value range of j is 1 to n; the elements of the weight matrix The calculation formula is: ； in, The kth order is features and the value of the j-th feature; are the average values ​​of all orders on the i-th feature and the j-th feature respectively; All orders in The average value of features; m is the total number of orders; is the feature index, used to traverse all features during the summation process; is the value of the k-th order on the i-th feature; k is an index variable used to traverse all orders in the data set, and the value range of k is 1 to m; the dynamic adjustment factors corresponding to each order are determined according to the sorting priority, including: ； in, represents the dynamic adjustment factor of the i-th order; Indicates the number of products contained in the i-th order; represents the number of steps in the production process of the i-th order; represents the number of types of raw materials required for the i-th order; is a tuning parameter; is the adjustment parameter of sorting priority; Indicates the sorting priority of the i-th order.

2. The intelligent order splitting method for a furniture production line according to claim 1 is characterized in that: Within each category, orders are sorted by delivery date to obtain sorting priorities, including: For each category of orders, multiple sorting sequences are randomly generated as the initial population, and each individual represents a sorting method for the corresponding category of orders; Define an evaluation function for evaluating the quality of each individual; Calculate the fitness value of each individual according to the evaluation function, randomly determine the corresponding individual, and perform a crossover operation to generate new offspring; The newly generated offspring are randomly mutated, and the selection, crossover and mutation operations are repeated until the preset number of iterations is reached to determine the final individual, and the final individual is used as the sorting result of the category order; Based on the sorting results, each order is assigned a sorting priority.

3. The intelligent order splitting method for furniture production line according to claim 2 is characterized in that: The calculation formula of the evaluation function is: ； in, Is an individual The fitness value of is the order quantity; Indicates the difference in days from the base date to the delivery date for each order; is a positive number; is the importance weight of the i-th order.

4. The intelligent order splitting method for a furniture production line according to claim 1 is characterized in that: Order information includes furniture type, size, color, quantity, and delivery date.

5. An intelligent order splitting system for a furniture production line, characterized in that: include: Acquisition module, used to receive order information; Determine the number of clusters K, and randomly select K orders from the order information as the initial cluster centers; For each order, calculate its distance from each cluster center and assign it to the category where the corresponding cluster center is located; for each cluster, calculate the average value of all orders in it, and set the corresponding average value as the new cluster center. Repeat the steps until the preset number of iterations is reached to obtain K clusters, and use the K clusters as the final order classification results, where each cluster corresponds to a specific production line; The processing module is used to sort the orders in each category according to the delivery date to obtain the sorting priority; determine the dynamic adjustment factor corresponding to each order according to the sorting priority; intelligently allocate the orders to each production line according to the orders of each category, the production capacity of the production line and the dynamic adjustment factor; for each order, calculate its distance from each cluster center, including: For each order, according to Calculate its distance from each cluster center; in, Indicates order With cluster center The distance between is an element of a weight matrix, indicating the importance of the interaction between the i-th feature and the j-th feature; is the bandwidth parameter of the Gaussian kernel function; Orders With cluster center The value in the i-th dimension; n is the number of dimensions of orders and cluster centers; Orders With cluster center In the The values ​​in the dimensions; i and j represent index variables, the value range of i is 1 to n, and the value range of j is 1 to n; the elements of the weight matrix The calculation formula is: ； in, The kth order is features and the value of the j-th feature; are the average values ​​of all orders on the i-th feature and the j-th feature respectively; All orders in The average value of features; m is the total number of orders; is the feature index, used to traverse all features during the summation process; is the value of the k-th order on the i-th feature; k is an index variable used to traverse all orders in the data set, and the value range of k is 1 to m; the dynamic adjustment factors corresponding to each order are determined according to the sorting priority, including: ； in, represents the dynamic adjustment factor of the i-th order; Indicates the number of products contained in the i-th order; represents the number of steps in the production process of the i-th order; represents the number of types of raw materials required for the i-th order; is a tuning parameter; is the adjustment parameter of sorting priority; Indicates the sorting priority of the i-th order.

6. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which implements the method according to any one of claims 1 to 4 when executed by a processor.

Images