Cable production scheduling modeling method

Abstract

The invention relates to a cable production scheduling modeling method. The method comprises the following steps: step 1, integration of production data; step 2, integration of order data; step 3, creation of a constraint model of order production; the method comprises a mathematical model of a cable production process and an ant colony optimization algorithm, and the two algorithm models are combined to complete production scheduling of a cable scheduling process.

Description

Technical Field

[0001] The invention relates to a cable production scheduling modeling method, and belongs to the technical field of intelligent scheduling. Background Art

[0002] Production scheduling is of vital importance in the manufacturing industry. Production scheduling can help companies make more effective use of their production resources, such as staff, equipment and materials. Effective scheduling can also ensure the full use of resources, improve production efficiency, and avoid or reduce idle resources. A reasonable and effectively executed production scheduling plan can help companies improve production efficiency and reduce waste, while also helping to ensure the quality of products or services. Production scheduling helps determine the production time and sequence of products, so that companies can complete production tasks within the specified period, meet customer orders on time, and increase customer satisfaction. Effective production scheduling can reduce the inventory level of work-in-progress and finished products and reduce inventory costs. In addition, reasonable production scheduling can also help reduce the additional costs caused by overproduction and production interruptions. By optimizing production processes, improving production efficiency and quality, reducing costs and meeting delivery dates, production scheduling can greatly improve the competitiveness of enterprises, thereby gaining more market share. Implementing effective production scheduling is very important for an enterprise, both from the perspective of improving efficiency and profits and enhancing competitiveness.

[0003] The cable scheduling method aims to use reasonable and appropriate methods to properly plan the cable production and manufacturing process, improve processing efficiency, and reduce unnecessary resource waste. It can be summarized as the scheduling problem of the production and manufacturing workshop. Arthanari and Salvador first proposed the scheduling problem in which the process can be processed on multiple machines in the 1970s, which marked the beginning of the study of flexible flow shop scheduling problems. He and Babayan studied the two-stage flexible flow shop scheduling problem considering delayed product differences, in which there is only one machine in the first stage and multiple identical parallel machines in the second stage. Schulz et al. proposed an improved multi-phase iterative local search algorithm to solve the k-stage mixed flow shop scheduling problem with maximum completion time, total energy consumption and peak load as objective functions.

[0004] Cable manufacturing companies mainly focus on user orders and formulate production plans according to customer requirements. However, as a traditional manufacturing industry, the cable industry still relies heavily on manual scheduling and dispatching during the manufacturing process, which leads to the inability to fully utilize the production efficiency of equipment and affects the overall benefits of the company. In addition, in the actual process of cable manufacturing, there are random disturbance events such as inserting orders, reducing orders, and advance orders, which have a great impact on the production efficiency of the workshop. Therefore, intelligent production scheduling technology can provide a more efficient and reliable solution for the dynamic scheduling plan update of the cable production workshop, effectively reduce production costs, shorten production cycles, ensure production quality, and enhance the overall competitiveness of enterprises.

[0005] However, there are four major difficulties in the cable production process: 1. Difficulty in defining production objects 2. Complex order merging and material substitution 3. Different logic before and after key processes 4. Both process manufacturing and discrete manufacturing characteristics. It is necessary to effectively extract order information, further analyze the scheduling strategy before and after key equipment, and accurately handle the complex changes of order merging and material substitution in the cable production process, so as to achieve an objective and accurate description of the mathematical model for the intelligent scheduling of cable production. Summary of the invention

[0006] The present invention aims at the problems existing in the prior art and provides a cable production scheduling modeling method, which includes a mathematical model of the cable production process and an ant colony optimization algorithm, and combines the two algorithm models to complete the production scheduling of the cable scheduling process.

[0007] This method solves four major difficulties in cable production scheduling: the difficulty in defining production objects, the complexity of order merging and material substitution, the different logic before and after key processes, and the characteristics of both process manufacturing and discrete manufacturing. It realizes intelligent scheduling of cable production, improves scheduling efficiency and production efficiency, and can effectively ensure the utilization rate of key equipment.

[0008] In order to achieve the above object, the technical solution of the present invention is as follows: a cable production scheduling modeling method, comprising the following contents:

[0009] Step 1: Integrate production data

[0010] (1) Production time

[0011] During the production process, each unit length needs to be spliced ​​from the next section to the end of the current section, which takes mmin time, and this operation takes exactly the same time on each device, so it is necessary to increase the packaging time of each device. Moreover, different devices have their own production time when completing parts of the same length. When the production program starts, we have to judge the situation of the equipment, which depends on the highest common factor HCF and m of the production time of all equipment, which we call MPTC (number of Minimum processing time cell). In this process, the status of the equipment may change each time it passes through MPTC. Therefore, we use the number of MPTC to describe the production time of the equipment and obtain the equipment update matrix T p ,and It is the number of MPTC per unit production length:

[0012]

[0013] (2) Time-consuming production change

[0014] The amount of MPTC that needs to be replenished when the production process changes is T c

[0015]

[0016] (3) Equipment status

[0017] E S =[1,1,1,1,1,1,1,1] (3)

[0018] 1 means the device is available, 0 means the device is busy or faulty

[0019] (4) Material status

[0020] M jn Represents the quantity of the nth material in the jth process

[0021] Step 2: Integrate order data

[0022] (1) Order integration

[0023] Since cables have both process and discrete characteristics, all process operations must be completed according to the established process during the production process. However, the production process does not complete the processing of a whole cable at one time, but is divided into several sections according to its length for processing. i Represents a cable, P ij represents the jth process in cable i, part ij1 represents the first part of the jth process in cable i, and in different processes kij The value of is different, and each device needs to complete different processing segments. processType represents the processing type, deadline represents the delivery time, and class in Enter the material type, class out Represents the output material type.

[0024] Line i =[P i1 ,P i2 ,P i3 ,P i4 ,P i5 ,P i6 ,P i7 ,P i8 ] (4)

[0025]

[0026] part ijk =[processType,deadline,class in ,class out ] ij (8)

[0027] Step 3: Create a business constraint model for order production

[0028] (1) Hard constraints

[0029] Processing constraints:

[0030] 1) One device can only process one process at a time

[0031] 2) One device can only process one section of an order at a time

[0032] 3) Each section in the order must complete the pre-steps first

[0033] 4) After being processed, the fragments of different orders can be applied to the next processing step of any order that meets the material requirements.

[0034] 5) The production completed by one equipment is only the production of one process in one section of the order

[0035] Change of production constraints:

[0036] In the actual production process, there are four situations in the production process of different orders:

[0037] 1) Different materials, same process;

[0038] 2) Same materials and same processes;

[0039] 3) Different materials and processes;

[0040] 4) Same materials, different processes.

[0041] The changeover time for options 1 and 2 is fixed at m min, which is the time consumed for replacing the tray; while the changeover time for options 3 and 4 is not fixed, depending on different equipment. The time consumed for changing production for different equipment is also different, and scheduling is needed to minimize the total time consumption, while making the key equipment run at full capacity as much as possible to improve the production efficiency of the entire workshop. Another key indicator is that the order must be completed within the deadline.

[0042] (2) Soft Constraints

[0043] Key equipment processing logic constraints:

[0044] Since the cross-linking machine is selected as a key equipment, its production process needs to be stable and uninterrupted, and the process change should be avoided as much as possible to reduce the time consumption of production change. At the same time, it is necessary to ensure the sufficient supply of the corresponding raw materials as much as possible to prevent the situation of forced production change due to insufficient raw materials. Therefore, in the scheduling process, the corresponding part is selected ijk When setting a larger penalty coefficient for key equipment, try to avoid it from selecting parts that require process switching. ijk At the same time, as shown in Table 1, the time taken by the previous equipment of the cross-linking machine to produce a unit length is less than that of the cross-linking machine, so it needs to give priority to the production of all raw materials that may be used by the cross-linking machine, ensuring that there is at least one copy of each raw material; if there are multiple raw materials with less than one copy, the raw materials currently being used by the cross-linking machine will be given priority, provided that there are still fragments of such raw materials in the processing fragment; if each of the raw materials required by equipment 3 has at least one, equipment 2 will be produced according to the part that avoids process conversion as much as possible. ijk For processing.

[0045]

[0046] In the formula, Equipment2 represents the processing section selected by Equipment 2, represents the material used in the current processing of equipment 3, P(M) represents the production segment that produces M raw materials, and [P] is an indicator function, which is 1 when proposition P is true, otherwise it is 0.

[0047] Step 4: Build a scheduling mathematical model based on production process characteristics

[0048] (1) Calculation of the likelihood of processing segments

[0049] Due to the particularity of cable production, most equipment can only process one section at a time, one machine can only execute one cable order unit time period, and one machine can only select one cable order unit time period. The selectable section needs to meet two conditions: sufficient material preparation and the order section has not yet completed processing.

[0050]

[0051] Among them, j represents the corresponding equipment, n represents the nth material type of the corresponding equipment, and it needs to be judged every time the equipment starts processing or completes processing. jn When it is greater than 1, you can select the material M required in production equipment j. jn In the formula, j represents the corresponding equipment, and m represents the mth process mode of the corresponding equipment. Each time the equipment starts processing or completes processing, it needs to be judged. jm When .num>1, equipment j also needs to produce this type of product, Parts jm When the above material preparation status, equipment segment selection status and time segment processing status are all available, you can select Parts jm Carry out production.

[0052] (2) Processing segment selection

[0053] Each Parts jm The probability of being selected is shown in formula 13:

[0054]

[0055] Where WT am Steps j It refers to the time required to change production from the previous production process a to the selected process m on equipment j, and μ is the distance penalty coefficient, so as to obtain the possibility of selection of each processable Parts.

[0056] (3) Equipment\Material Status Update

[0057] After selecting the corresponding Parts, it is necessary to update the production data, production equipment, and remaining unfinished orders to keep all information updated synchronously, as shown in formulas 14, 15, and 16.

[0058]

[0059] Parts ij .num refers to the number of parts that need to be completed in process i in equipment j. When it is selected, its number is reduced by one.

[0060] (4) Scheduling plan records

[0061]

[0062] Record each part switch of each device, fully record the results of the entire simulation production scheduling, and generate a scheduling plan as shown in Formula 17.

[0063] Record the time taken to change production. After completing the analysis of the order, we get the parts, and each part requires corresponding equipment to complete its production and processing. Different parts require different processing technologies. Therefore, we need to use Formula 2 to build a production exchange time matrix to record the production exchange time between different parts and provide guidance for the selection of production and processing technology for parts, as shown in Formula 18, ρ ijk-i'j'k' It refers to the time consumed when the kth processing segment in order i is switched to the kth processing segment in order i` on equipment j. ik (tpye.P j ) refers to the processing method used for the kth processing segment in order i:

[0064]

[0065] Step 5: Iterative optimization based on ant colony algorithm

[0066] The mathematical model constructed based on formulas (1)-(18) can complete a scheduling task, but its scheduling results are highly random and cannot meet the demand for the shortest possible production time. The ant colony algorithm is needed to optimize its scheduling results.

[0067] (1) Heuristic information definition, let ρ ijk-i'j'k' Become and The relevant heuristic value is given by the heuristic value ρ ijk-i'j'k' Come to evaluate The advantages and disadvantages of ijk-i'j'k' yes The reciprocal of is shown in Formula 19.

[0068]

[0069] (2) Update the pheromone matrix, set Pronounced as pheromone, it is a positive real value that is intended to indicate how useful it is in making the following decisions Selected The next operation is processed by the same machine. In the ACO algorithm, M ants learn pheromones through cooperation, so that they can probabilistically find a better scheduling solution. The update rules of pheromones are shown in Equations 20 and 21.

[0070]

[0071] (3) Random selection probability definition, ant r chooses the next The following rule, the action rule, is used when . The random selection is made from the distribution given by the formula. This formula gives the When just selected, ant r chooses as the probability of the next action. Among them, α and β are parameters that weigh the relative importance of the learned pheromone and heuristic value. ijk It refers to whether the order itself meets the conditions for being selected, whether it has been completed and whether the materials are ready.

[0072]

[0073] Compared with the prior art, the advantages of the present invention are as follows:

[0074] The present invention proposes a cable production scheduling optimization algorithm in a multi-order and multi-product scenario, which solves the problems of low production scheduling efficiency and low resource utilization of traditional production lines. The invention abstracts a mathematical model based on the actual cable production process, accurately expresses the characteristics of processability and discreteness in the production process, and proposes a special scheduling plan for key equipment on the production line to ensure its long-term stable operation. Finally, the ACO algorithm is used to continuously optimize the scheduling results, which improves the production efficiency by 10.04% in the production scenario test case. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 is a flowchart of the algorithm operation in an embodiment of the present invention;

[0076] Figure 2 It is a Gantt chart of scheduling results in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to deepen the understanding of the present invention, the present embodiment is described in detail below with reference to the accompanying drawings.

[0078] Example 1: In order to verify the feasibility and optimization effect of the above method, the production scheduling and optimization of the medium voltage workshop of a cable manufacturer were completed. Different products require different processing techniques in terms of quantity and process parameters, but there are cases where some processes are the same but the raw materials are different, and some raw materials are the same but the processes are different. Therefore, it has caused great difficulties in production scheduling. It is difficult to come up with a suitable scheduling plan based on experience and manual calculations alone, taking into account production efficiency and delivery time, which often causes equipment to idle and wait during the production process.

[0079] Step 1: Integrate production data

[0080] (1) Production time

[0081] Table 1 Equipment processing time

[0082]

[0083] (2) Time-consuming production change

[0084] Table 2 Equipment process switching time

[0085]

[0086] T c =[9,13,64,0,0,13,0,13]

[0087] (3) Equipment status

[0088] E S =[1,1,1,1,1,1,1,1]

[0089] 1 means the device is available, 0 means the device is busy or faulty

[0090] (4) Material status

[0091]

[0092] Step 2: Integrate order data

[0093] (1) Order integration

[0094] To this end, we perform ant colony-based scheduling optimization for the production orders received by the workshop on 2024 / 07 / 18, as shown in Table 3.

[0095] Table 3 Order requirements

[0096]

[0097] Convert the order information into an order matrix and save it in a file for reading.

[0098] Step 3: Construct an industry constraint model for order production, including processing constraints and production change constraints in hard constraints, and key equipment processing logic constraints in soft constraints.

[0099] Step 4: Scheduling plan record

[0100]

[0101] Step 5: The optimization of the existing scheduling model is completed through the ant colony algorithm. The ant colony parameters are set. In this verification, the number of ants is set to 50, the number of cycles is 150, the pheromone coefficient is 10, the distance influence coefficient is 1, the pheromone index parameter is 4, the distance influence index parameter is 1, and the pheromone volatility coefficient is 0.5. Then, according to the information in Table X, the scheduling scheme code is edited according to Figure X and integrated into the ant colony iteration process. Finally, the results of the ant colony iteration process are converted into a Gantt chart to output the total time and scheduling scheme for the production of all orders, such as Figure 1 shown.

[0102] Table 4 Ant colony optimization process

[0103]

[0104]

[0105] The iteration results are shown in Table 4. The result of the first round (589MPTCs) was 2.8632 days, while the actual workshop scheduling required 3 days, which was a significant improvement over the direct manual scheduling result. With further iterations of the optimization, in the 5th iteration, the obtained production schedule (575MPTC) required 2.7951 days to complete production, which was 6.83% higher than the manual scheduling. At this time, the optimization progress began to slow down. In the 10th iteration, the production schedule was further optimized to require (565MPTCs) 2.7465 days to complete the production of all orders. Finally, in the 37th iteration, the optimal solution under the algorithm was found. The generated production schedule required (553MPTCs) 2.6882 days to complete the production of all tasks, which increased the productivity by 10.04% compared with the initial manual scheduling.

[0106] It should be noted that the above embodiments are not intended to limit the protection scope of the present invention, and equivalent changes or substitutions made on the basis of the above technical solutions all fall within the protection scope of the claims of the present invention.

Claims

1. A cable production scheduling modeling method, characterized in that: The method comprises the following steps: Step 1: Integrate production data, Step 2: Integrate order data, Step 3: Create a constraint model for order production, Step 4: Construct a scheduling mathematical model based on production process characteristics. Step 5: Perform iterative optimization based on ant colony algorithm.

2. A cable production scheduling modeling method according to claim 1, characterized in that: Step 1: Integrate production data as follows: (1) Production time In the production process, each unit length needs to be spliced ​​from the next section to the end of the current section, which takes m min. This operation takes exactly the same time on each device, so it is necessary to add the packaging time of each device. Different devices also have their own production time when completing parts of the same length. When the production program starts, the situation of the equipment must be judged. It depends on the highest common factor HCF and m of the production time of all equipment, which is called MPTC (minimum processing time unit number). In this process, the equipment status may change each time after passing MPTC. The number of MPTC is used to describe the production time of the equipment, and the equipment update matrix T is obtained. p ,and It is the number of MPTC per unit production length: (2) It takes time to change production. The amount of MPTC that needs to be replenished when the production process changes is T c (3) Equipment status, E S ＝[1,1,1,1,1,1,1,1] (3) 1 means the device is available, 0 means the device is busy or faulty. (4) Material status, M jn Represents the quantity of the nth material in the jth process.

3. A cable production scheduling modeling method according to claim 1, characterized in that: Step 2: Integrate order data, as follows: (1) Order integration, Since cables have the dual characteristics of process and discreteness, all process operations are completed according to the established process during the production process. During the production process, they are divided into several sections for processing according to their length. In the formula, Line i Represents a cable, P ij represents the jth process in cable i, part ij1 represents the first part of the jth process in cable i, and in different processes k ij The value of is different, and each device needs to complete different processing segments. processType represents the processing technology type, deadline represents the delivery time, and class in Enter the material type, class out Represents the output material type, Line i =[P i1 ,P i2 ,P i3 ,P i4 ,P i5 ,P i6 ,P i7 ,P i8 ] (4) part ijk =[processType,deadline,class in ,class out ] ij (8) 4. A cable production scheduling modeling method according to claim 1, characterized in that: Step 3: Create a business constraint model for order production, as follows: (1) Hard constraints Processing constraints: 1) One device can only process one process at a time. 2) One device can only process one section of an order at a time. 3) Each section in the order must complete the pre-steps first. 4) After being processed, the fragments of different orders can be applied to the next processing step of any order that meets the material requirements. 5) The production completed by one equipment is only the production of one process in one section of the order Change of production constraints: In the actual production process, there are four situations in the production process of different orders: 1) Different materials, same process; 2) Same materials and same processes; 3) Different materials and processes; 4) Same materials, different processes. The changeover time for options 1 and 2 is fixed at m min, which is the time consumed for replacing the tray. The changeover time for options 3 and 4 is not fixed, and depends on different equipment. The time consumed for changing equipment is also different, and scheduling is needed to minimize the total time consumption, while making the key equipment run at full capacity as much as possible to improve the production efficiency of the entire workshop. Another key indicator is that the order must be completed within the deadline. (2) Soft Constraints Key equipment processing logic constraints: Since the cross-linking machine is selected as the key equipment, its production process needs to be stable and uninterrupted, and the process change should be avoided as much as possible to reduce the time consumption of production change. At the same time, it is necessary to ensure the sufficient supply of the corresponding raw materials as much as possible to prevent the situation of forced production change due to insufficient raw materials. Therefore, in the scheduling process, the corresponding part is selected for production. ijk When setting a larger penalty coefficient for key equipment, try to avoid it from selecting parts that require process switching. ijk , the time taken by the previous equipment of the cross-linking machine to produce a unit length is less than that of the cross-linking machine, so it needs to give priority to the production of all raw materials that may be used by the cross-linking machine, ensuring that there is at least one copy of each raw material; if there are multiple raw materials with less than 1 copy, the raw materials currently used by the cross-linking machine will be given priority, provided that there are still fragments of such raw materials in the processing fragment; if each raw material required by equipment 3 has at least one, equipment 2 will follow the part that avoids process conversion as much as possible. ijk Processing, In the formula, Equipment2 represents the processing section selected by Equipment 2, represents the material used in the current processing of equipment 3, P(M) represents the production segment that produces M raw materials, and [P] is an indicator function, which is 1 when proposition P is true, otherwise it is 0.

5. A cable production scheduling modeling method according to claim 1, characterized in that: Step 4: Construct a scheduling mathematical model based on the production process characteristics, as follows: (1) Calculation of the likelihood of processing fragments, Due to the particularity of cable production, most equipment can only process one section at the same time, one machine can only execute one cable order unit time period, and one machine can only select one cable order unit time period. At the same time, the selectable section needs to meet two conditions: sufficient material preparation and the order section has not yet completed processing. Among them, j represents the corresponding equipment, n represents the nth material type of the corresponding equipment, and it needs to be judged every time the equipment starts processing or completes processing. jn When it is greater than 1, you can select the material M required in production equipment j. jn In the formula, j represents the corresponding equipment, and m represents the mth process mode of the corresponding equipment. Each time the equipment starts processing or completes processing, it needs to be judged. When Parts jm When .num>1, equipment j also needs to produce this type of product, Parts jm You can select this equipment for production. When the above material preparation status, equipment segment selection status and time segment processing status are all available, you can select Parts jm To produce, (2) Processing segment selection, Each Parts jm The probability of being selected is shown in formula 13: Where WT am Steps j It refers to the time required to change production from the previous production process a to the selected process m on equipment j, and μ is the distance penalty coefficient, so as to obtain the possibility of selection of each processable Parts. (3) Equipment\Material Status Update After selecting the corresponding Parts, it is necessary to update the production data, production equipment, and remaining unfinished orders to keep all information updated synchronously, as shown in formulas 14, 15, and 16. Parts ij .num=Parts ij .num-1 (16) Parts ij .num refers to the number of parts that need to be completed in process i in equipment j. When it is selected, its number decreases by one. (4) Scheduling plan records Record each part switch of each device, fully record the results of the entire simulation production scheduling, and generate a scheduling plan as shown in formula 17. Record the time taken to change production. After completing the analysis of the order, we get the parts. Each part requires corresponding equipment to complete its production and processing. Different parts require different processing technologies. Therefore, we need to use Formula 2 to build a production exchange time matrix to record the production exchange time between different parts and provide guidance for the selection of production and processing technologies for parts, as shown in Formula 18. It refers to the time consumed when the kth processing segment in order i is switched to the kth processing segment in order i` on equipment j. ik (tpye.P j ) refers to the processing method used for the kth processing segment in order i:

6. A cable production scheduling modeling method according to claim 4, characterized in that: Step 5: Perform iterative optimization based on the ant colony algorithm, as follows: The mathematical model constructed based on formulas (1)-(18) completes a scheduling task, but its scheduling results are highly random and cannot meet the requirement of shortening production time as much as possible. The ant colony algorithm is needed to optimize its scheduling results. (1) Heuristic information definition, let ρ ijk-i'j'k' Become and The relevant heuristic value is given by the heuristic value ρ ijk-i'j'k' Come to evaluate The advantages and disadvantages of ijk-i'j'k' yes The reciprocal of is shown in formula 19. (2) Update the pheromone matrix, set Pronounced as pheromone, it is a positive real value that is intended to indicate how useful it is in making the following decisions Selected The next operation is processed by the same machine. In the ACO algorithm, M ants learn pheromones through cooperation. In this way, they can probabilistically find a better scheduling solution. The pheromone update rule is shown in Equations 20 and 21. (3) Random selection probability definition, ant r chooses the next The following rule, the action rule, is used when the random selection is made from the distribution given by the formula, which gives the When just selected, ant r chooses As the probability of the next action, where α and β are parameters that weigh the relative importance of the learned pheromone and heuristic values, z ijk It refers to whether the order itself meets the conditions for being selected, whether it has been completed and whether the materials are ready.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, a cable production scheduling modeling method as described in any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, a cable production scheduling modeling method as described in any one of claims 1-6 is implemented.

Images