Intelligent flexible production method applied to small part production

Abstract

The application provides an intelligent flexible production method applied to small part production and belonging to the technical field of part machining, and comprises the following steps: S1, establishing a work station: a plurality of machining units are first established, and an operation set is divided into a corresponding number of non-empty subsets, and each stage is allocated to sequentially perform operations; S2, processing the balancing problem in step 1: S201, a method based on a decomposition heuristic; S202, a heuristic method based on an artificial bee colony; S203, a heuristic algorithm based on a constructive method; the intelligent flexible production line provided by the application can solve the balancing problem of the intelligent flexible automatic line, the balancing problem under double uncertainty, the balancing problem of multi-objective robustness, the batch scheduling problem and the like; meanwhile, the intelligent flexible production line can be applied to various engineering practices, small parts, complex parts, multi-variety mass production parts and the like; moreover, the intelligent flexible production line has low cost and high cost performance, and has great competitive advantage in the domestic and foreign markets.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of part processing, and particularly relates to an intelligent flexible production method applied to small part production. BACKGROUND

[0002] With higher and higher requirements for energy saving, environmental protection and safety of automobiles, and rapid technological progress, new automobile parts are launched at an increasingly fast speed. Under this background, revolutionary changes have taken place in automobile part production methods in recent years. Emerging multi-variety mass production methods have posed severe challenges to manufacturing technology, aiming at integrating high flexibility and high efficiency. Thus, flexible automatic lines for multi-variety mass production of small complex parts have emerged.

[0003] Compared with traditional special production lines, the flexible automatic line adopts a numerical control machine tool device capable of flexible processing, can quickly respond to changes in market demand, and can be used for producing multiple models and multiple series of products. There are also assembly line balancing problems, flexible production line balancing problems under uncertain environment, and scheduling problems in mixed flow production lines in the production process.

[0004] There are also assembly line balancing problems in the production process. When the production design capacity (production line beat) is known, the problem of minimizing the production line configuration cost is called simple assembly line balancing problem-I (SALB-I). When the production line configuration and the operation time of each process of the product are known, the problem of maximizing the production line capacity (i.e. minimizing the production line beat) is called simple assembly line balancing problem-II (SALB-II). The former considers building an intelligent flexible production line at the lowest cost under the condition of meeting market demand, and the latter maximizes the capacity of the existing intelligent flexible production line for a new product when responding to changes in market demand.

[0005] Therefore, we provide an intelligent flexible production method applied to small part production to solve the above problems. SUMMARY

[0006] The application aims to provide an intelligent flexible production method applied to small part production, and aims to solve the balancing problem in the intelligent flexible production line in the prior art.

[0007] To achieve the above-mentioned purpose, the application provides the following technical scheme: an intelligent flexible production method applied to small part production, comprising the following steps:

[0008] S1, establishing a work station: first, a plurality of processing units are established, and an operation set is divided into a corresponding number of non-empty subsets and each stage is assigned to perform operations in turn;

[0009] S2, processing the balancing problem in step 1:

[0010] S201, a method based on decomposition heuristic: design three propositions, respectively, sub-problems CCP-1, CCPv2 and CCP-3, to be used;

[0011] S2011, initialization: first calculate the lower limit value of the number of machines in the deterministic environment MpL: (|MpL|, ), p is the number of machines; and make the initialization relaxation variable R = 0;

[0012] S2012, solve the sub-problem: first get the result of CCP-1, then calculate the probability P2 of event 2, and check the result of sub-problem CCP-2 according to the result of CCP-1;

[0013] S2013, check the constraint satisfaction and output the result;

[0014] S202, a heuristic method based on artificial bee colony:

[0015] S2021, first initialize the honey source, and give each employed bee a honey source;

[0016] S2022, then dispatch the employed bees in multiple processing units, calculate the honey value of each scene in the production line and production stage

[0017] S2023, then dispatch the observer bees, perform local non-dominated sorting between the production stages, and finally generate new honey sources by interleaving

[0018] S203, a heuristic algorithm based on construction:

[0019] S2031, obtain the batch of each production line,

[0020] S2032, obtain the partial batch sequence with the optimal target value,

[0021] S2033, select the k partial batch sequences that maximize the target value Obj,

[0022] S2034, select the result output with the optimal Obj value.

[0023] Further, each processing unit in step S1 is composed of one or more numerical control machine tools and a robot.

[0024] Further, each numerical control machine tool is provided with two trays, and each tray can carry two or more parts to be processed.

[0025] Further, the movement speed of each robot is constant.

[0026] Further, the processing time of each numerical control machine tool can be controlled.

[0027] Further, the operation in the step S1 has a priority constraint between each other.

[0028] Further, the CCP-1 is For indicating that the probability of CCPv1 is negatively correlated with the total number of machine failures, P1 is the probability of event 1, i, m is the number, (ximxmp) is the event 1 machine failure correlation quantity.

[0029] Further, the CCP-1 is For indicating that the probability of CCP-1 is positively correlated with the total number of machine failures, P2 is the probability of event 2, i, m is the number, (yimymp) is the event 2 machine failure correlation quantity.

[0030] Further, the step S2011 is t is time, and CTd is demand tact.

[0031] Further, the position of the batch in the list in the step S2031 is recorded, and the recorded value is k.

[0032] Compared with the prior art, the beneficial effects of the present application are:

[0033] 1、The intelligent flexible production line provided by the present application can solve the balance problem of the intelligent flexible automatic line, the balance problem under double uncertainty, the balance problem of multi-objective robustness, the batch scheduling problem, etc.; can be applied to various engineering practices, small parts, complex parts, multi-variety mass-produced parts, etc.; and has low cost and high cost performance, so that it has great competitive advantage in the domestic and foreign markets.

[0034] 2、The processing unit composed of multiple numerical control machine tools and automatic robots can realize automatic production, thereby reducing enterprise labor cost and human operation errors, and further improving production line automation level and production quality; each production stage is composed of multiple processing units capable of processing distributed processing operation tasks in parallel, so that the parallel processing units of the same production stage are designed as mutual insurance, so that the reliability of the production line is improved; the use of the numerical control machine tool with a tool magazine and the machining center can realize the flexible capability of the production line for multi-product processing; the movable automatic robot can form a resource sharing and dynamic reconfigurable virtual processing unit with the corresponding processing machine tool, so that the configuration of the production line can be adjusted as needed; through rapid re-adjustment of the composition of the virtual processing unit and the virtual reconfiguration of the production stage, the rapid switching of different products can be realized, and the production line capacity is maximized; space can be reserved for new equipment at the beginning of the design of the production line, and due to the mobility of the robot, the production line can also decide whether to expand the number of production equipment according to the fluctuation of product demand in the market. BRIEF DESCRIPTION OF DRAWINGS

[0035] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of the specification, illustrate embodiments of the application and are used to explain the application, but are not intended to limit the application. In the drawings:

[0036] Figure 1 Flow chart of the work unit in the present application;

[0037] Figure 2 Block diagram of the operation set in the present application. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0039] Embodiment 1

[0040] The production line structure in the present application is composed of multiple production stages, each production stage is composed of multiple parallel processing units, and each processing unit is composed of the same machining tool and robot. Meanwhile, the structure characteristics of the intelligent flexible production line are analyzed, a mathematical model is established, and an effective algorithm is designed, which can solve the balancing problem of the new intelligent flexible production line in a reasonable time. The intelligent flexible production line has a special line configuration, that is, the production line is composed of multiple production stages, each production stage processes a part of the operation set, and each production stage is composed of one or more processing units, and each processing unit is composed of a robot and one or more numerical control machining tools. For the operation set to be allocated for a certain type of product, the priority constraint, mutual exclusion constraint and inclusion constraint between operations are considered. According to the actual situation of most production lines, problems of different scales are designed. At the same time, the computational complexity of this new type of intelligent flexible production line is described, and it is clarified that this problem is an NP-hard problem. In the method of solving similar assembly line balancing problems, one of the most effective methods is the branch and bound method. The present application refers to the assembly line balancing problem and proposes a heuristic method based on branch and bound to solve the intelligent flexible production line balancing problem;

[0041] Considering the double uncertainty of machine tool failure frequency and failure maintenance recovery time, the flexible automatic line balancing problem is established, the corresponding chance constrained programming model is solved, and a decomposition heuristic algorithm is established to solve the chance constrained programming problem. Secondly, the performance indicators between the decomposition heuristic method and the stochastic simulation method are compared. The present application tends to design a production line that meets the production line cycle time CT dA reliable production line with satisfaction degree α, i.e. the production line can meet the production cycle time less than or equal to the required cycle time CT with a probability of α% under the consideration of uncertain machine downtime d The satisfaction degree α is a production line parameter used to describe the reliability of the production line. The design of the production line aims to meet the required production line cycle time with a satisfaction degree α under the consideration of uncertain machine failure frequency and uncertain failure recovery time (repair time), while minimizing the production line configuration cost, i.e. the number of machines and robots, to establish an opportunity constrained programming model.

[0042] Please refer to Figures 1-2 The present application provides the following technical solutions:

[0043] An intelligent flexible production method applied to small part production, comprising the following steps:

[0044] S1, establishing a workstation: first, establish multiple processing units, and divide the operation set into a corresponding number of non-empty subsets and assign each stage to perform operations in turn;

[0045] S2, processing the balancing problem in step 1:

[0046] S201, a method based on decomposition heuristic: design three propositions, CCP-1, CCP-2 and CCP-3, to be used;

[0047] S2011, initialization: first, calculate the lower limit value of the number of machines under the deterministic environment (|MpL|, ), and set the initialization relaxation variable R = 0

[0048] S2012, solving the sub-problem: first, obtain the result of solving CCP-1, then calculate the probability P2 of event 2, and check the result of sub-problem CCP-2 according to the result obtained from CCP-1;

[0049] S2013, check the constraint satisfaction and output the result;

[0050] S202, a heuristic method based on artificial bee colony:

[0051] S2021, first, initialize the honey source and give each employed bee a honey source;

[0052] S2022, then dispatch the employed bees in multiple processing units, calculate the honey value of the production line and production stage in each scenario

[0053] S2023, then dispatch the scout bees, perform non-dominated sorting between local honey values of the production stage, and finally generate new honey sources by interleaving

[0054] S203, a heuristic algorithm based on construction

[0055] S2031、Obtain the batch of each production line,

[0056] S2032, obtain the partial batch sequence with the optimal target value,

[0057] S2033, select the k batch partial sequences that maximize the target value Obj,

[0058] S2034, select the result output with the optimal Obj value.

[0059] In a specific embodiment of the application, in order to better test the balancing problem, in this embodiment, each production stage has the same production capacity as each other and only contains one processing machine tool and one robot, while the operation set needs to be divided into a corresponding number of non-empty subsets and each stage is assigned to perform operations in turn, and there are priority constraints between operations, and the target is to minimize the configuration cost of the production line. In this case, each production stage is equivalent to a workstation in the assembly line problem, so the intelligent flexible production line balancing problem is equivalent to the simple assembly line balancing problem-I;

[0060] Based on the second type of balancing problem of intelligent flexible automatic line, the planning targets to minimize the production line beat time, and the corresponding mixed integer programming model is established and solved.

[0061] In a flexible automatic line, the production line beat time is defined as the interval time between the completion of processing of two consecutive products from the intelligent flexible production line. The production line beat is equal to the beat time of the bottleneck production stage, that is, it depends on the maximum beat time in all production stages. Therefore, this study targets to minimize the production line beat, and because the relationship between the beat formulas of the automatic line is nonlinear, a nonlinear programming model is established for the balancing problem of the flexible automatic line.

[0062] By using two typical methods of branch and bound method and tabu search algorithm to solve the established model, and as a reference, the results are compared with the results obtained by the heuristic algorithm based on set partitioning studied, and the feasibility of the proposed algorithm is discussed.

[0063] Intelligent flexible production line multi-objective balancing problem under uncertain processing time environment, different multi-objective robustness dominance standards are proposed and the balancing problem is solved.

[0064] Based on the second type of balancing problem of flexible automatic line, the uncertain factors of processing time are introduced, and the multi-objective problem in the intelligent flexible production line balancing problem is considered, and the following assumptions are made under the original model:

[0065] Each processing unit consists of four identical numerical control machine tools and one robot;

[0066] Each production stage has at least one processing unit;

[0067] Each CNC machine has two pallets, each of which can hold two or more parts to be processed;

[0068] The movement speed of the robot is constant;

[0069] The processing time on the CNC machine varies unpredictably and depends on several factors:

[0070] The approximate processing time of each process on the CNC machine is known;

[0071] The range of processing time changes is known due to the influence of each uncertain factor;

[0072] The probability range of each uncertain factor occurring is known;

[0073] The machines in the same production stage have the same performance, and the change in processing time caused by uncertain factors during processing is the same on all machines in the same production stage, i.e. all machines in the same production stage have the same

[0074] The intelligent flexible production line considered only processes one product;

[0075] Further, according to the modeling of the beat relationship in the flexible automatic line, a target model for optimizing the beat time of the considered production line is established.

[0076] The purpose of the target optimization problem is to optimize multiple conflicting goals at the same time, and such problems can obtain a set of Pareto solutions rather than a single solution. The general form of a multi-objective problem with multiple conflicting minimization objectives is

[0077] min f o (x), o = 1, 2,..., O

[0078] C j (x) ≤ 0, j = 1, 2,..., J

[0079] D k (x) = 0, k = 1, 2,..., K

[0080] Where f o (x), C j (x), D k (x) are multiple conflicting minimization objective functions, and the optimal multi-objective robustness solution under the corresponding standard is obtained according to the r-type dominance standard and the SR-type dominance standard.

[0081] The problem of dynamic batch scheduling of mixed product types in a multi-flexible automated production line environment:

[0082] Considering hybrid product models, the dynamic demand of customer orders within each planning scope is arranged across multiple production lines. Furthermore, the uncertainty of demand, changeover time between different products, and machine failures must be considered when developing the schedule. This invention proposes a hybrid integer programming model that takes into account uncertain product demand, customer order delivery dates, uncertain machine failures on the production line, and changeover time between different product models. The aim is to maximize the probability of timely completion of all customer orders with varying product demands within the planning scope. Therefore, a constructive heuristic algorithm is proposed to address this problem. The proposed heuristic method can allocate different product batches across different production lines and balance the completion times between production lines. The performance of the constructive heuristic algorithm is primarily compared with that of classic scheduling heuristics (i.e., NEHedd and AGB heuristics) used in scheduling problems with different batch sizes. When a production line does not need to handle other product types, each production line is considered to have different production takt times for producing different products. Each production line has the same number of production stages, but the production takt time for each stage differs across different production lines.

[0083] Balancing Problems in Intelligent Flexible Production Lines - Part 1

[0084] For details, please refer to Figure 1 The intelligent flexible production line with parallel machines studied in this invention considers the balance configuration problem of the intelligent flexible production line under uncertain environments, such as... Figure 1 As shown. From Figure 1 As can be seen, this intelligent flexible production line consists of five processing units for performing different operations. Each unit comprises several identical machine tools that perform the same operations. Within each unit, an automated robot performs necessary auxiliary operations, such as cleaning and transferring finished materials from the machine tools. One or more pallets are positioned near the machine tools to hold product materials; these can be considered buffer zones. For each unit, multiple consecutive machine tools in the production line are aligned, in addition to the material conveyor system. The pallets carry the processed parts from the unit, transferring them to a conveyor belt with special mechanical devices next to the machine tools, moving them to the next unit for subsequent processing operations based on operational priority constraints.

[0085] Specifically, a solution based on the decomposition heuristic.

[0086] This paper addresses the decomposition heuristic (DH) algorithm proposed by CPP. Before proposing the decomposition heuristic algorithm, the following propositions are introduced.

[0087] Proposition 1: shows that the probability of event 1 is negatively related to the total number of machine failures.

[0088] Proposition 2: shows that the probability of event 2 is positively related to the total number of machine failures in event 2.

[0089] Proposition 3: If the machine failure parameter {xmp} of event 1 is a feasible solution that satisfies both sub-problems (CCP-1 and CCP-2), then it is also a feasible solution of the original CCP problem.

[0090] Proposition 1 and Proposition 2 give the relationship between the total number of machine failures and the two probability events. Proposition 3 shows the equivalence of the two sub-problems and the original problem. Since CCP-1 and CCP-2 are computable, the following heuristic method is given to solve the CCP problem:

[0091] Step 1: Initialization.

[0092] Step 1.1: Calculate the lower bound of the number of machines (|MpL|, under the deterministic environment. Generally, |MpL| = |Mp| - |Mp| * (1 - P1) * (1 - P2) * (1 - P3) *... * (1 - Pn) where n is the number of events. t is the time, and CTd is the demand cycle time.

[0093] Step 1.2: Initialize the relaxation variable R = 0.

[0094] Step 2: Solve the two sub-problems.

[0095] Step 2.1: Obtain the result of CCP-1 solution.

[0096] Step 2.2: Calculate the probability P2 of event 2 and check the result of sub-problem CCP-2 according to the result obtained from CCP-1.

[0097] Step 3: Check the constraints and output the result.

[0098] Step 3.1: If the result is a feasible solution of CCP-2, output the result; otherwise, proceed to the next step.

[0099] Step 3.2: Set R = R + 1 and go to Step 2.

[0100] The above decomposition heuristic can output the optimal solution for CCP by gradually expanding the relaxation value R, which provides more machines. CCP-1 and CCP-2 are solved according to the increased relaxation value R in the loop, and the result is checked to see if it is a valid solution. When the first time CCP-1 and CCP-2 obtain a valid solution that satisfies the constraints, the minimum value of R is obtained, and the minimum number of machines in the problem is also obtained.

[0101] For details, see the heuristic method based on artificial bee colony.

[0102] The artificial bee colony (ABC) algorithm proposed by Karaboga has been applied to various optimization problems and has been proved to be an effective algorithm for solving optimization problems. In existing research, different local search methods are introduced to improve the algorithm, and cross-neighborhood search is used to improve the information sharing between bee colonies. The improved ABC algorithm of the application adopts a heuristic search mechanism of HBB. According to HBB, the search mechanism is guided, and the honey source of each hired bee is similar to the node of the branch and bound method.

[0103] Specifically, the heuristic algorithm is constructed

[0104] Step 1: Obtain the batch of each production line;

[0105] Step 2: For all production lines, obtain the delivery date DDil of each batch in the batch sequence;

[0106] Step 3: Make a batch list in the order of delivery date, and k represents the position of the batch in the list; Step 4: For all production lines, select the batches at positions k=1 and k=2 in their lists, and obtain

[0107] all possible sequences;

[0108] Step 5: Calculate the completion time of the batches in all possible sequences of the batches;

[0109] Step 6: Obtain the partial batch sequence with the optimal objective value;

[0110] Step 7: For each production line, the batch at sequence position k>2:

[0111] Step 7.1: Select the batch at position k from the list in step 3, and insert it into the current partial sequence in step 6

[0112] obtained from all k possible time slots in the current partial sequence, which consists of k-1 batches;

[0113] Step 7.2: Select the k batch partial sequence that maximizes the objective value Obj as the current best partial sequence;

[0114] Step 7.3: Let i=1;

[0115] Step 7.4: Remove batch i (i.e. the i-th batch in the sorted list obtained in step 3) from the current partial sequence and insert it into the k = 1 position of the remaining partial sequence. i = i + 1, repeat this step until i = k - 1. From the new sequence of partial sequences just created, remove common sequences (i.e. sequences that appear more than once, remove them and keep one of the sequences to avoid an increase in computation time) and remove sequences from the new sequence that are identical to the partial sequences. Calculate the corresponding Obj for all new sequences and obtain the best partial batch sequence that gives the largest Obj;

[0116] Step 7.5: k = k + 1, go back to step 7.1 until k = II, II being the number of batches;

[0117] Step 8: Select the batch with the smallest batch size from the production line with the largest Obj. Add the selected batch to the batches of the same order and the same product with a batch size smaller than the batch size Gmolτ in the production line with the smallest Obj. If there is no batch with a batch size smaller than Gmolτ, select the first batch in the same order and the same product from the production line with the smallest Obj. Recalculate the Obj and the completion time for both production lines using the best partial sequence from step 7. If the deviation of the completion time of all production lines (i.e. the difference between the completion time of that line and the average completion time of all lines) is smaller, keep this situation and repeat this step until the deviation of the completion time of all production lines no longer improves;

[0118] Step 9: Exchange the batch with the largest Obj in the production line with the largest Obj with the batch with the smallest Obj in the production line with the smallest Obj. Repeat steps 4 to 7 for both production lines;

[0119] Step 10: Repeat step 9 x times and select the result with the optimal Obj value as the output.

[0120] Contrast verification: see Figure 2 The present application gives a simple problem solving algorithm for an intelligent flexible production line containing eight operation sets. Figure 2 As shown, for 8 operation sets, wherein Fig. a is a mutual exclusion diagram of 8 operation sets, Fig. b is obtained by finding the complement of the diagram. Then, the unconnected part of Fig. b is blocked to obtain the corresponding sub-diagram, as shown in Fig. c. The complement of each sub-diagram in Fig. c is obtained, the operations with inclusion relationship are merged to obtain Fig. e, considering the inclusion relationship between operations. At this time, the lower bound value of the production stage can be obtained through Fig. e. Then, for the results in Fig. e, the unassigned operations are searched according to the priority constraint relationship (such as Fig. g) using the branch and bound method, and the optimal result is finally obtained, as shown in Fig. h.

[0121] Specifically, the intelligent flexible production line balancing problem under double uncertainty is studied:

[0122] The present application will be realized by MS visual C++, MATLAB 2016 coding to realize the decomposition heuristic algorithm, in the configuration Inter Core i7 processor, CPU is 2.5GHz and 16G memory computer, to the opportunity constraint programming model proposed in solving. Test the accuracy and performance of the CCP model and DH algorithm proposed in the present application. The solution obtained by the decomposition heuristic algorithm is compared with the stochastic simulation method.

[0123] The intelligent flexible production line lot-sizing problem under demand uncertainty is studied:

[0124] The proposed constructive heuristic for lot-sizing problem (CHLP) is compared with the classical AGB and NEHedd heuristics. The performance of the algorithms is measured in terms of the quality of the solutions obtained and the computation time used by the algorithms to solve the problem. Since the lot-sizing problem considered involves many factors and parameters, in order to fairly compare the CHLP with the heuristics in other studies, different problem parameters need to be considered, including the number of intelligent flexible production lines, the number of customer orders, the estimated demand of customer orders, and the density and due date of customer orders.

[0125] The present application can solve the intelligent flexible automatic line balancing problem, the balancing problem under double uncertainty, the multi-objective robust balancing problem, the lot-sizing problem, etc. The present application can be applied to various engineering practices, small parts, complex parts, multi-variety mass production parts, etc. The present application improves the production efficiency, energy consumption utilization, etc. to some extent, fills the technical gap of the complex intelligent flexible production line in the domestic market, and describes that the computational complexity is NP-hard. The present application has low cost and high cost performance, and has great competitive advantage in the domestic and foreign markets.

[0126] Finally, it should be noted that: the above only for the preferred embodiments of the present application, and not for the limitation of the present application, although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, it still can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent flexible production method applied to small parts production, characterized in that, The method comprises the following steps: S1, establishing a workstation: first, a plurality of processing units are established, and the operation set is divided into a corresponding number of non-empty subsets and each stage is assigned to perform operations in turn; S2, processing the balancing problem in step 1: S201, a method based on a decomposition heuristic: design three propositions, respectively, sub-problems CCP-1, CCP-2 and CCP-3, to be used; S2011, initialization: first calculate the lower limit value of the number of machines under the deterministic environment :( ), p is the number of machines; and make the initialization relaxation variable = 0; S2012, solving sub-problem: first, get the result of solving sub-problem CCP-1, then calculate the probability of event 2 2, and according to the result obtained from sub-problem CCP-1, check the result of sub-problem CCP-2; S2013, check the constraint satisfaction and output the result; S202, a heuristic method based on artificial bee colony: S2021, first, initialize the honey source, and give each employed bee a honey source; S2022, then dispatch the employed bees in the plurality of processing units, calculate the honey value of each scene in the production line and production stage S2023, then dispatch the scout bees, perform non-dominated sorting between local honey values of the production stage, and finally generate new honey sources by interleaving S203, a heuristic algorithm based on construction: S2031, obtain the batch of each production line, S2032, obtain the partial batch sequence with the optimal target value, S2033, selecting a target value maximal one batch partial sequence, S2034, selecting the result output with the optimal value. Wherein: each processing unit in the step S1 is composed of one or more numerical control machine tools and a robot; The CCP-1 is , for indicating that the probability of CCP-1 is negatively correlated with the total number of machine failures, 1 is the probability of event 1, , is the number of times, ) is the event 1 machine failure related quantity; The CCP-1 is , for indicating that the probability of CCP-1 is positively correlated with the total number of machine failures, 2 is the probability of event 2, , is the number of times, ) is the event 2 machine failure related quantity.

2. The intelligent flexible production method for small parts production according to claim 1, characterized in that, Each numerical control machine tool is provided with two trays, and each tray can carry two or more parts to be processed.

3. The intelligent flexible production method for small parts production according to claim 1, characterized in that, The movement speed of each robot is constant.

4. The intelligent flexible production method for small parts production according to claim 1, characterized in that, The processing time of each numerical control machine tool can be controlled.

5. The intelligent flexible production method for small parts production according to claim 1, wherein, The operations in the step S1 have priority constraints between each other.

6. The intelligent flexible production method for small parts production according to claim 3, wherein, In the step S2011 is time, is the demand beat.​​ 7. The intelligent flexible production method for small parts production according to claim 5, characterized in that, The position of the batch in the list in step S2031 is recorded, with the value .

Images