Method for analyzing aircraft assembly batch completion time based on discrete event simulation

Abstract

In view of the problem that existing discrete event simulation tools cannot perform differential simulation analysis on the production state and performance index of specific batches when simulating the process of aircraft assembly work, the application provides a method for analyzing the completion time of aircraft assembly batches based on discrete event simulation. Based on the traditional discrete event simulation framework, the application proposes a method of marking simulation work-in-process elements with corresponding labels by taking aircraft batch information as simulation input, and proposes a simulation model execution mechanism, a promotion framework and a completion time solving method around the batch label. The application can provide control personnel with reliable analysis results of the estimated completion time of each batch on the assembly line, thereby supporting the perception of the assembly state of each batch on site.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of aircraft assembly line simulation analysis, and particularly relates to a method for analyzing aircraft final assembly batch and frame completion time based on discrete event simulation. BACKGROUND

[0002] In order to improve the overall assembly efficiency and assembly quality, most aircraft final assembly lines divide assembly work into multiple stations for assembly, realize the synchronous production of multiple stations and multiple aircraft, and manage the aircraft assembly process from the aspects of planning, process, material and quality based on batch and frame management technology. At the same time, as an order-oriented complex discrete manufacturing process, ensuring that all assembly tasks are completed before the order delivery node is the fundamental goal of aircraft manufacturing enterprises, so that the on-site control personnel pay great attention to the assembly line capacity index. Due to the complexity of the aircraft product itself and the influence of various random factors in the assembly process, it is difficult to obtain reliable capacity index analysis results by using analytical method, so the industry currently generally uses discrete event simulation technology to complete assembly line modeling and analysis and optimization of capacity index. However, due to the existence of synchronous production of multiple aircraft in the field, it is not enough to only pay attention to the overall capacity of the assembly line, and the on-site control personnel are more concerned about the estimated completion time of each batch and frame of aircraft. However, the commonly used discrete event simulation tools can only obtain the overall performance of the assembly line by using a fuzzy, average and universal method when simulating the aircraft assembly process, and cannot obtain the production state and performance index of the specific batch and frame in a differentiated manner. SUMMARY

[0003] Relying on the actual needs of aircraft manufacturing enterprises, in view of the problem that the existing discrete event simulation tools cannot perform differentiated simulation analysis on the production state and performance index of the specific batch and frame when simulating the aircraft assembly process, the present application provides a method for analyzing the final assembly batch and frame completion time of an aircraft based on discrete event simulation. Based on the traditional discrete event simulation framework, the present application proposes a method of giving the simulation work-in-process element corresponding tags by taking the aircraft batch and frame information as the simulation input, and proposes a simulation model execution mechanism, a promotion framework and a completion time solving method around the batch and frame tags. The present application can provide control personnel with a reliable analysis result of the estimated completion time of each batch and frame on the final assembly line, thereby supporting the on-site assembly state perception of each batch and frame.

[0004] To achieve the above object, the present application adopts the following technical solutions:

[0005] A method for analyzing the final assembly batch and frame completion time of an aircraft based on discrete event simulation, characterized in that it comprises the following steps:

[0006] Step S1: count total assembly site production information, and construct batch frame total assembly line whole line operation process simulation model:

[0007] S101, collect and count resource information in each station of the assembly line, thereby using formal symbols to construct resource and resource library related elements in the assembly station operation process simulation model; the resource information includes resource number, name, configuration quantity, and resource library information, wherein the resource library information includes resource configuration information of each station, and information interaction set of the resource library and all work elements in the station;

[0008] S102, collect and count material information in each station of the assembly line, thereby using formal symbols to construct material and material library related elements in the assembly station operation process simulation model; the material information includes material number, name, supply frequency, and material library information, wherein the material library information includes material configuration information of each station, and information interaction set of the material library and all work elements in the station;

[0009] S103, collect and count assembly operation information of each station and each frame of aircraft in the assembly line, combine operation state information and input and output interaction information of operation, and use formal symbols to construct operation related elements in the assembly station operation process simulation model; the assembly operation information includes operation number, name, resource demand and material demand of operation development, rated working hours, and logical sequence constraint between all operations of the same frame in the station; the operation state information refers to the state description of whether the operation is completed, which is a self-defined parameter; the input and output interaction information of operation is the interaction information set between operations, the interaction information set between operation and resource library, and the interaction information set between operation and material library;

[0010] S104, construct the assembly station operation process simulation model: based on the resource, material and assembly operation related elements constructed in steps S101-103 above, using the logical sequence constraint relationship between the assembly operations contained in the station and the assembly operations, the material supply and resource configuration information, the interaction information between the assembly operation and the material library and the resource library, and the state transition information of the station (i.e. the set of all operation state information in the station), formal symbols are used to construct the assembly station operation process simulation model;

[0011] S105, construct the whole line operation process simulation model of the assembly line: based on the operation process simulation model of each station constructed in S104, connect each station according to the logical sequence relationship, thereby constructing a complete whole line operation process simulation model of the assembly line.

[0012] The above steps S101 and S102 are not in sequence and can be interchanged.

[0013] Step S2, make the execution mechanism of the total assembly line whole line job process simulation model constructed in step S1:

[0014] S201, make the batch execution mechanism of the total assembly line whole line job simulation model:

[0015] In the simulation process, the nested mode of internal and external loops is adopted, the external loop is used to judge the number of times the batch aircraft enters the assembly line, and the internal loop judges the execution progress of the batch aircraft entering the assembly line each time by recording the flight information;

[0016] S202, make the execution mechanism of the job simulation element in the total assembly line whole line simulation model:

[0017] The job dynamic execution first carries out conditional judgment according to conditions one to three as follows,

[0018] One is to check whether the materials required for the job are complete, if the materials are complete, get the materials, otherwise, update the job state to the waiting material state;

[0019] Two is to check whether the equipment, tools / decoration required for the job meet the demand, if the equipment, tools / decoration meet the demand, get the equipment, tools / decoration, otherwise, update the job state to the waiting equipment, tools / decoration state;

[0020] Three is to check whether the human resources required for the job meet the demand, if the human resources meet the demand, get the human resources, otherwise, update the job state to the waiting human resources state;

[0021] After completing the conditional judgment, if all the above three conditions are met, update the state of the job element to "executing", and grab the corresponding materials, equipment, tools / decoration, and human resources from the material library and resource library, after the job execution is completed, release the equipment, tools / decoration, and human resources occupied by the job, update the state of the job to "completed" and count the job end time;

[0022] S203, generation mechanism of random variables in the total assembly line whole line simulation model:

[0023] In the simulation process, the actual working hours of the job, and the device failure, personnel leave, and material shortage events are set as random events, among which, according to the application habit of traditional discrete event simulation in the field of production system, the actual working hours of the job are set as subject to normal distribution, the device failure, personnel leave, and material shortage are set as subject to exponential distribution, the ordinary skilled person in the art can also use other random variable distribution functions to express the above random events according to the actual situation;

[0024] Step S3, make the total assembly line whole line job process simulation model constructed in step S1:

[0025] S301, input of simulation data:

[0026] The foregoing step S1 constructs a static assembly line simulation model for simplifying the expression of an actual production line. The purpose of step S3 is to formulate a propulsion framework for dynamically propelling the simulation running of the assembly line simulation model established in step S1. Therefore, simulation data needs to be first input into the assembly line simulation model established in step S1. The simulation data are overall parameters of the simulation process, including the simulation cycle number, batch data, and simulation start time. The batch data include the batch number, frame information, and total quantity.

[0027] S302, initialization of the simulation process:

[0028] According to the batch data, the input order and quantity of products (i.e., aircraft) in the assembly line simulation model established in step S1 are initialized, and the configuration of all materials and resources is initialized. Meanwhile, a simulation clock and a future event table are established and initialized. The simulation clock is used to record the absolute time in the simulation propulsion process and is an inherent attribute of the entire simulation system, which is initialized as the simulation start time input in step S301. The future event table is used to store various events (job execution, job completion, resource release, etc.) to be occurred in the future in chronological order, which is initialized as an empty table.

[0029] S303, scanning of the simulation system state:

[0030] The scanning of the simulation system state includes scanning the configuration quantity of all materials, the available state of various resources, the location of each frame of aircraft, and the execution of all jobs at each station.

[0031] S304, event conversion:

[0032] The future event table is updated according to the results obtained in step S303.

[0033] S305, simulation propulsion:

[0034] The next job to be executed is selected from the future event table by a decision rule in combination with the current state of all jobs, and the simulation clock is propelled to the time when the job occurs. The decision rule refers to a job priority order sorting rule.

[0035] S306, event processing:

[0036] The next job information obtained in the step S305 is received, first, the material and resource request signals are sent to the material store and the resource store of the station where the job to be executed is located respectively; second, the feedback signals of the material store and the resource store are obtained, and it is determined whether to start the job and execute it or not; finally, after the job is executed, the resources are released, the signal indicating that the occupied resources are released is sent to the resource store, and the state of the job is changed to "completed";

[0037] S307, the simulation system state is updated:

[0038] After a job is completed, all material configuration quantities, the available states of various resources, the positions of each aircraft, and the execution of all jobs at each station are recalculated, thereby updating the system state;

[0039] S308, the simulation progress is judged:

[0040] The simulation progress is judged according to the simulation cycle number and the aircraft serial number, wherein the simulation cycle number is used to judge the total progress of the simulation, and the aircraft serial number is used to judge whether a simulation cycle is completed or not, when the aircraft with the largest serial number is in the last station and all assembly jobs at the station are completed, it indicates that a cycle is completed;

[0041] S309, the simulation is ended:

[0042] When the simulation cycle number reaches the maximum and the last aircraft in the simulation cycle is completed, the simulation is ended; in the field of discrete event simulation, the maximum simulation cycle number is defined according to the actual simulation accuracy requirement, which is known technology.

[0043] Step S4, the aircraft completion time index is calculated:

[0044] S401, the total assembly period of each aircraft is converted into a variable in days based on the simulation start time, and is represented as D i =(d i1 ,d i2 ,…,d ij ), wherein d ij represents the completion of the assembly of the jth aircraft in the ith cycle at the d ij th day from the simulation start time;

[0045] S402, the total simulation cycle number is represented as n, thereby reconstructing the statistic d x =(d 1x ,d 2x ,…,d nx ), wherein d x is the set of assembly period days of the xth aircraft, and d nxThis represents the assembly time in days for the x-th aircraft relative to the simulation start time in the nth simulation cycle. The simulation result d... 1x ,d 2x ,…,d nx It can be considered as a random variable that follows a normal distribution;

[0046] S403. Using the interval estimation method, by running the simulation a certain number of times, the most likely completion date of each aircraft can be obtained in a confidence interval of 1-α, where α represents the significance level. In most cases, it will be set to 0.05. The resulting confidence interval is a 95% confidence interval.

[0047] S404. After each simulation cycle, compare whether the confidence interval of the current completion time calculation for each flight meets the accuracy requirements. If it does not meet the requirements, increment the total number of simulation cycles by 1, return to step S3 to restart the simulation process, and perform simulation analysis. Iterate in this way until the flight completion time calculation result that meets the accuracy requirements is obtained.

[0048] Based on the above technical solutions, the present invention also makes the following optimizations and further limitations:

[0049] Furthermore, in step S101, each resource element in the simulation model can be represented as a triple: r i ={id i Nam i ,Num i}, where id i Nam is the unique identifier for the i-th resource; i Let Num be the name of the i-th resource; i Represents the quantity of the i-th type of resource; each resource pool is counted and represented as a tuple: RC i ={rc i ,rcl i}, where rc i =(r i1 ,r i2 ,…,r ij Let r be the set of resource configuration information for the i-th station. ij This represents the configuration information of the j-th resource at the i-th station; rcl i =(rcl i1 ,rcl i2 ,…,rcl ij Let be the set of information exchanges between the resource repository at the i-th station and all operations within the i-th station.

[0050] Furthermore, in step S102, each material-related element in the simulation model can be represented as a triple: m j ={idj nam j IAT j , where id j is the unique number of the jth material; nam j is the name of the jth material; IAT j represents the supply frequency of the jth material. Each material inventory is represented as a binary tuple: MC i = {mc i , mcl i}, where mc i = (m i1 , m i2 , …, m ij ) is the set of material configuration information at the ith station, m ij represents the configuration information of the jth material at the ith station; mcl i = (mcl i1 , mcl i2 , …, mcl ik ) is the set of information interaction between the material inventory at the ith station and all the jobs in the ith station.

[0051] Further, in step S103, each job-related element in the simulation model can be represented as a seven-tuple: where Id i is the unique number of the xth aircraft for this job; n i is the name of the xth aircraft for this job; R i = (H i , l i ) represents the set of resource requirements for starting the xth aircraft for this job, where column vector H i = (id1, id2, …, id j ) T and l i = (Num1, Num2, …, Num j ) T represent the numbers and quantities of the required resources, respectively; represents the set of material requirements for executing the xth aircraft for this job, where column vector K i = (id1, id2, …, id k ) T and represent the numbers and quantities of the required materials, respectively; t i = (t ir , t it ) is the set of time for the xth aircraft for this job, t ir is the rated working hours for the xth aircraft for this job, t itSimulation man-hours of the xth aircraft for the task; st i st represents the current state of the xth aircraft for the task, and the initial state is st i = 0, and st i = 1 if the task is completed; If i = (PI i , PO i ) represents the interaction information set of the xth aircraft for the task, wherein PI i is an input information set, and PO i is an output information set; meanwhile, the logical sequence constraint matrix in the same station between different aircrafts is the same, so the logical sequence constraint matrix of all tasks in the station is uniformly defined as:

[0052] PR= (pr ij ) n×n (1)

[0053] wherein, i∈{1,2,...,n}, j∈{1,2,...,n}, and i≠j, pr i is a task-related element in the simulation model, and since the logical sequence constraint matrix in the same station between different aircrafts is the same, no aircraft label x is used for distinction here.

[0054] Further, in step S104, the assembly station task process simulation model can be represented as a six-tuple: wherein ID i is a unique identifier of the ith station on the assembly line; ∑ i = (P i , Rl i ) is a task relationship network in the ith station, wherein P i = (p i1 (x) , p i2 (x) ,…, p it (x) ) is a set of all tasks in the ith station, and p it (x) represents the xth aircraft in the ith station for the tth task, and the logical relationship matrix Rl i = (Rl i(j,k) ) t×t represents the precedence relationship of all tasks in the ith station, and if task p ij (x) is the immediately preceding task of task p ik (x) , then Rl i(j,k) = 1, and otherwise, Rl i(j,k)= 0; RC i (x) denotes the resource pool of the i-th station when the x-th aircraft is assembled at the i-th station; MC i (x) denotes the material pool of the i-th station when the x-th aircraft is assembled at the i-th station; ST i (x) denotes the state of the i-th station, which is determined by the states of all the jobs in the i-th station at time t, IF i (x) = (SI i , SO i ) denotes the interaction information set of the i-th station, wherein SI i is the input information set, including the finished product information of the previous station and the finished product information of the previous aircraft at the station, and SO i is the output information set, i.e., the finished product information of the current station.

[0055] Further, in step S105, the assembly line integrated job process simulation model can be represented as a four-tuple: AL (x) = {S (x) , R (x) , W (x) , ST (x)}, wherein S is the set of stations on the assembly line, and i is the total number of stations on the assembly line; R (x) = (RC1 (x) , RC2 (x) , …, RC j (x) is the configuration information set of all resources on the assembly line; W (x) = (MC1 (x) , MC2 (x) , …, MC k (x) is the supply information set of all materials on the assembly line; ST (x) = (ST1 (x) , ST2 (x) , …, ST i (x) is the state of the assembly line, which is determined by the states of the stations.

[0056] Further, in step S305, the decision rule refers to the job priority order sorting rule of the comprehensive job working hour coefficient, the job manpower resource demand coefficient, and the job working hour coefficient of the next job, and the establishment method is as follows: assuming that the job set P c = {p c1 , pc2 ,…,p ck}, the descriptions of the three types of coefficients are as follows:

[0057]

[0058] wherein, CJF ci is the operation man-hour coefficient, t ci represents the rated man-hour of the operation p ci ; 1≤i≤k;

[0059]

[0060] wherein, HDF ci is the operation human resource demand coefficient, represents the total number of resource demands of the operation p ci , and J represents the total number of resource types;

[0061]

[0062] wherein, OTF ci is the immediate post-operation man-hour coefficient, t' ci represents the total immediate post-operation man-hour of the operation p ci ; according to the above three coefficients, the assembly operation priority calculation formula is established:

[0063] PEC i = λ1CJF ci + λ2HDF ci + λ3OTF ci (5)

[0064] wherein, λ1-λ3 are weight coefficients, and usually, λ1 and λ3 are set to be the same size, and λ2 can take a value in the range of 0-0.5, and the value is determined according to the degree of attention of the practitioner to human resources in the practice process;

[0065] The priority of each assembly operation is calculated according to the above formula (5), and the operation priority order sorting rule is obtained according to the priority size.

[0066] Further, in step S401, the total assembly period of each rack is calculated according to the following formula:

[0067]

[0068] wherein, represents the assembly period of the xth rack at the jth station in the τth simulation cycle, and m represents the total number of stations in the total assembly line.

[0069] Further, in step S403, the interval estimation method is used to estimate the confidence interval of the set d of the assembly duration days of the xth aircraft x The confidence interval of the confidence degree of 1-alpha is:

[0070]

[0071] Wherein, The mean of the set of assembly duration days is represented by The variance of the set of assembly duration days is represented by The upper side quantile of the t distribution with n-1 degrees of freedom is represented by

[0072] Compared with the prior art, the present application has the following beneficial effects:

[0073] Based on the traditional discrete event simulation technology, the present application constructs the assembly line operation process simulation model through the statistics of the field batch aircraft operation, resources and material information, formulates the simulation promotion framework and simulation execution mechanism considering the batch aircraft information, and gives the aircraft completion time analysis method based on the simulation calculation result, so that the present application can solve the defects that the traditional discrete event simulation technology cannot perform the batch aircraft differentiation reasoning and analysis, and reliably provides the confidence interval of the predicted completion date of each batch aircraft for the practitioners. BRIEF DESCRIPTION OF DRAWINGS

[0074] Figure 1 is the operation process diagram between the stations of the assembly line in the embodiment of the present application.

[0075] Figure 2 is the operation information obtained by statistics in the first station of the assembly line in the embodiment of the present application.

[0076] Figure 3 is the resource information obtained by statistics in the first station of the assembly line in the embodiment of the present application.

[0077] Figure 4 is the material information obtained by statistics in the first station of the assembly line in the embodiment of the present application.

[0078] Figure 5 is the network diagram of the operation process simulation model of the first station of the assembly line in the embodiment of the present application.

[0079] Figure 6 is the network diagram of the whole line operation process simulation model of the assembly line in the embodiment of the present application.

[0080] Figure 7 is the batch processing execution mechanism diagram of the simulation model in the embodiment of the present application.

[0081] Figure 8 is the confidence interval diagram of the completion time of each aircraft first calculated in the embodiment of the present application.​

[0082] Figure 9 is the confidence interval chart of each batch completion time and date after precision adjustment in the embodiment of the application.

[0083] Figure 10 is the distribution chart of each batch completion time in the embodiment of the application. DETAILED DESCRIPTION

[0084] Embodiment:

[0085] A certain type of aircraft assembly line is selected as the object, which contains five stations, each station completes partial assembly work, and the main process between stations is as shown in Figure 1 , which is a serial logical relationship.

[0086] Step S1: Statistics Figure 1 The actual resources, materials, operations, and batch information of each station of the assembly line are shown in Figure 2 , a batch assembly line whole line operation process simulation model is constructed, taking the first station as an example, 22 operation information are obtained, as shown in Figure 2 ; from the statistical resource information, 5 kinds of key equipment which have greater influence on assembly progress and 8 kinds of professional human resources are selected, which together form 13 resource information, as shown in Figure 3 ; from the statistical material information, 10 kinds of key materials which may not be supplied in time in the production process are selected, forming 10 key material information, as shown in Figure 4 ; based on the above statistical information, the operation process simulation model of the first station S1 is constructed into a six-tuple using formal symbols:

[0087]

[0088] Among them: ID1=1, indicating the first station of the assembly line; ∑1=(P1,Rl1), P1=(p 11 (x) ,p 12 (x) ,…,p 1,21 (x) ,p 1,22 (x) ) is the first station 22 operation set, Rl1 is the first station 22 operation logic sequence matrix;

[0089]

[0090] RC1 (x)= {rc1, rcl1} represents the resource library of the first station when the xth aircraft is assembled at the first station, rc1 is the first station 13 resource configuration information set, rcl1 is the information interaction set of the first station resource library and all jobs (22 items) in the first station; MC1 (x) = {mc1, mcl1} represents the material library of the first station when the xth aircraft is assembled at the first station, mc1 represents the material configuration information set, and mcl1 represents the interaction information set of the first station material library and all jobs (22 items) in the first station; represents the state of the first station at time t, which is composed of 22 job states; IF1 (x) = (SI1, SO1) represents the first station input / output interaction information set. Among them, the information interaction set rcl1 of the first station resource library and all jobs in the first station, the information interaction set mcl1 of the first station material library and all jobs in the first station, the state ST1 of the first station at time t (x) (t), the first station input / output interaction information set IF1 (x) and the like are dynamically changed with the simulation running.

[0091] In order to intuitively display the job process simulation model of the first station in a graphical form, the material library, the resource library and the information interaction of the job are represented by small squares, and the information transmission direction is represented by a dashed line with an arrow; the information interaction between jobs is represented by a long square, and the information transmission direction is represented by a solid line with an arrow, thereby drawing a network diagram of the first station job process simulation model, as shown in Figure 5 The five stations are connected in series to obtain the total assembly line job process simulation model, and the internal interaction information expression of each station is simplified to draw a network diagram of the total assembly line job process simulation model, as shown in Figure 6

[0092] Step S2: Formulate the execution mechanism of the simulation model:

[0093] In this embodiment, each batch contains 3 aircrafts, and the batch processing execution mechanism of this embodiment is formulated in a nested manner, as shown in Figure 7 The external n simulation loops correspond to the batch entering the total assembly line simulation model a total of n times, and in each batch simulation, the execution progress of the batch aircraft in the current simulation cycle is judged by recording the current aircraft information at each station, for example, in this embodiment, if the third aircraft in the fifth station has been completely completed, it represents that the current cycle of the batch has been completed. The execution mechanism of the simulation model job element and the generation mechanism of the random variable are formulated according to the general method described in the technical scheme of the present application.

[0094] Step S3: Formulate the simulation model promotion framework:​

[0095] The simulation input data in this embodiment is as follows: the order contains 3 aircrafts, the batch is named X, the flight numbers are named X01, X02 and X03 respectively, the simulation start time is set to 2022-03-01 08:00, and the initial cycle number is set to 30. The remaining sub-steps are formulated according to the general method described in the technical solution of the present application.

[0096] Step S4: Calculate the flight completion time index:

[0097] In this embodiment, the simulation data is first processed to obtain the production period of each cycle X01, X02 and X03, then the mean and variance of the production period of each flight are calculated, and finally α = 0.05 is selected to obtain The confidence interval of the completion time of each flight is calculated, as shown in Figure 8 In order to obtain a higher precision result, the critical value of the solution precision is limited to ε = 0.5, and the simulation cycle number is gradually increased until the cycle number reaches 187, and the confidence interval of the completion time and date of each flight after precision adjustment that meets the precision requirement is obtained, as shown in Figure 9 The assembly cycle in days is taken as the horizontal axis and the distribution probability is taken as the vertical axis, a probability distribution graph is drawn, and the distribution probability of each day in the assembly completion time set of each flight is drawn on the graph, as shown in Figure 10 The three curves from left to right in the figure correspond to X01, X02 and X03 flights respectively.

[0098] Through the above method, the confidence degree of the predicted completion time of the three flights in the batch is obtained as a 95% confidence interval, that is, the true value of the completion time has a 95% probability of falling within the interval measured by the embodiment of the present application.

Claims

1. A method for analyzing aircraft final assembly cell throughput based on discrete event simulation, characterized in that, The method comprises the following steps: Step S1: statistics total assembly site production information, and constructing a batch of assembly line whole line operation process simulation model: S101, collect and statistics total assembly line each station in the resource information, using formalized symbol to build the resource and resource library related elements in the assembly station operation process simulation model; S102, collect and statistics total assembly line each station in the material information, using formalized symbol to build the material and material library related elements in the assembly station operation process simulation model; S103, collect and statistics total assembly line each station in, each batch of aircraft assembly operation information, the work state information and the input and output interaction information of work are combined, and formalized symbol is used to build the work related elements in the assembly station operation process simulation model; S104, constructing assembly station operation process simulation model: Based on the resource, material and assembly operation related elements, the logical sequence constraint relationship between the assembly operation contained in the station and the assembly operation, the material supply and resource allocation information, the interaction information of assembly operation and material library and resource library, and the state transition information of station, the assembly station operation process simulation model is constructed by using formalized symbol; S105, constructing total assembly line whole line operation process simulation model: On the basis of the each station operation process simulation model, the each station is connected according to the logical sequence relationship, and the total assembly line whole line operation process simulation model is obtained; The above steps S101 and S102 can be interchanged in sequence; Step S2, the execution mechanism of the total assembly line whole line operation process simulation model constructed in step S1 is formulated: S201, the batch processing execution mechanism of the total assembly line whole line operation simulation model is formulated: In the simulation process, the internal and external loop nesting mode is adopted, the external loop is used to judge the number of times that the batch aircraft enters the total assembly line, and the internal loop is used to judge the execution progress of the batch aircraft entering the total assembly line each time by recording the batch information; S202, the execution mechanism of the operation simulation element in the total assembly line whole line simulation model is formulated: The operation dynamic execution is carried out according to the following conditions one to three in turn, One is to check whether the required materials for operation are complete, if yes, the materials are obtained, otherwise, the operation state is updated to the waiting material state; Two is to check whether the required equipment, tools and accessories meet the requirements, if yes, the equipment, tools and accessories are obtained, otherwise, the operation state is updated to the waiting equipment, tools and accessories state; Three is to check whether the required human resources meet the requirements, if yes, the human resources are obtained, otherwise, the operation state is updated to the waiting human resources state; If the above three conditions are all met, the state of the operation element is updated to "executing", and the corresponding materials and resources are grabbed from the material library and the resource library, after the operation is completed, the materials and resources occupied by the operation are released, the state of the operation is updated to "completed" and the operation end time is counted; S203, the generation mechanism of random variables in the total assembly line whole line simulation model: The actual working hours of the job, equipment failure, personnel leave, and material shortage events are set as random events, wherein the actual working hours of the job are set as subject to normal distribution, and the equipment failure, personnel leave, and material shortage are set as subject to exponential distribution; Step S3, establishing a propulsion framework of the total assembly line job process simulation model constructed in step S1: S301, simulation data input: The simulation data are overall parameters of the simulation process, including simulation cycle number, batch data, and simulation start time; the batch data include batch number, frame information, and total quantity; S302, simulation process initialization: According to the batch data, the input order and quantity of products in the total assembly line simulation model established in step S1 are initialized, the configuration of all materials and resources is initialized, and the simulation clock and future event table are established and initialized; S303, simulation system state scanning: The simulation system state scanning includes scanning the configuration quantity of all materials, the available state of various resources, the position of each frame aircraft, and the execution of all jobs at each station; S304, event conversion: The future event table is updated according to the results obtained in step S303; S305, simulation propulsion: The next job to be executed is selected from the future event table by a decision rule according to the state of all current jobs, and the simulation clock is propelled to the time when the job occurs; the decision rule refers to a job priority order sorting rule; S306, event processing: The next job to be executed obtained in step S305 is received, and first, material and resource request signals are sent to the material library and resource library of the station where the job to be executed is located; second, the feedback signals of the material library and resource library are obtained, and it is decided whether to start and execute the job; finally, after the job is executed, the resources are released, a signal representing that the occupied resources have been released is sent to the resource library, and the state of the job is changed to "completed"; S307, simulation system state updating: After a job is completed, the configuration quantity of all materials, the available state of various resources, the position of each frame aircraft, and the execution of all jobs at each station are recalculated, and the system state is updated; S308, simulation progress judgment: The simulation progress is judged according to the simulation cycle number and frame label number, wherein the simulation cycle number is used to judge the total progress of the simulation, and the frame label number is used to judge whether a simulation cycle is completed; when the aircraft with the largest frame label number is at the last station and all assembly jobs are completed, it is marked that a cycle is completed; S309, simulation end: When the simulation cycle number reaches the maximum and the last frame of the simulation cycle is completed, the simulation is ended; Step S4, calculating frame completion time index: S401. Using the simulation start time as a baseline, convert the total assembly duration of each flight into a variable in days, using D. i =(d i1 ,d i2 ,…,d ij ) represents the simulation result of the i-th simulation cycle, where d ij This indicates that the j-th flight in the i-th cycle occurs at time d from the start of the simulation. ij Assembly completed within one day; S402, express the total simulation cycle number as n, thereby reconfigure the statistics d x = (d 1x ,d 2x ,…,d nx ), d x is the set of the xth aircraft assembly duration days, wherein d nx indicates the assembly duration days of the xth aircraft in the nth simulation cycle relative to the simulation start time, the simulation result d 1x ,d 2x ,…,d nx is regarded as a random variable subject to normal distribution; S403, by using interval estimation method, the most likely completion date of each aircraft is obtained in a confidence interval with a confidence level of 1-α through multiple simulation cycle operations, wherein α represents the significance level; S404, after each simulation cycle, compare the current confidence interval of the calculation of the time to complete each batch to see if it meets the accuracy requirements, if not, add 1 to the total simulation cycle number, return to step S3 to re-propel the simulation process, perform simulation analysis, and iterate until the time to complete each batch meets the accuracy requirements.

2. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 1, wherein: In step S101, each resource element in the simulation model is represented as a triple: r i = {id i , Nam i , Num i}, where id i is the unique number of the i-th resource; Nam i is the name of the i-th resource; Num i represents the number of the i-th resource; each resource pool is counted and represented as a two-tuple: RC i = {rc i , rcl i}, where rc i = (r i1 , r i2 , …, r ij ) is the set of resource configuration information at the i-th station, r ij represents the configuration information of the j-th resource at the i-th station; rcl i = (rcl i1 , rcl i2 , …, rcl ij ) is the set of information interaction between the resource pool at the i-th station and all the jobs in the i-th station.

3. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 2, wherein: In step S102, each material element in the simulation model is represented as a triple: m j = {id j , nam j , IAT j}, where id j is the unique number of the jth material; nam j is the name of the jth material; and IAT j represents the supply frequency of the jth material; and each material library is represented as a pair: MC i = {mc i , mcl i}, where mc i = (m i1 , m i2 , …, m ij ) is the set of material configuration information at the ith station, m ij represents the configuration information of the jth material at the ith station; and mcl i = (mcl i1 , mcl i2 , …, mcl ik ) is the set of information interaction between the material library at the ith station and all the jobs in the ith station.

4. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 3, wherein: In step S103, each task-related element in the simulation model is represented as a seven-tuple: Where Id i This is the unique identifier for the xth aircraft in this operation; n i The name of this operation for the xth aircraft; R i =(Η i ,λ i H represents the set of resource requirements for the commencement of this operation for the xth aircraft, where the column vector H i =(id1,id2,…,id j ) T and λ i =(Num1,Num2,…,Num j ) T These represent the required resource number and quantity, respectively; Let K represent the set of material requirements for the x-th aircraft to perform this operation, where column vector K i =(id1,id2,…,id k ) T and These represent the required material number and quantity, respectively; t i =(t ir ,t it ) represents the time set for the operation of the x-th aircraft, t ir For the rated man-hours of this operation for the xth aircraft, t it The simulated man-hours for this operation on the xth aircraft in the simulation; st i This indicates the current state of the operation for the xth aircraft. Initially, st... i =0, if the task is completed, st i =1; If i =(PI) i ,PO i ) represents the set of interactive information for the x-th aircraft in this operation, where PI i For the input information set, PO i To output the information set; at the same time, the logical order constraint matrix is ​​the same for different flights within the same station, and the logical order constraint matrix for all operations within the station is uniformly defined as follows: PR = (pr ij ) n×n (1) wherein, and i≠j, p i is a certain job-related element in the simulation model, and since the logical sequence constraint matrix is the same for all flights in the same station, no flight label x is used here to distinguish.

5. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 4, wherein: In step S104, the assembly station operation process simulation model is expressed as a six-tuple: where ID i is the unique identification of the i-th station on the assembly line;∑ i = (P i , Rl i ) is the operation relationship network in the i-th station, where P i = (p i1 (x) , p i2 (x) , …, p it (x) ) is the set of all operations in the i-th station, p it (x) represents the t-th operation of the x-th aircraft in the i-th station, the logical relationship matrix Rl i = (Rl i(j,k) ) t×t represents the precedence relationship of all operations in the i-th station, if operation p ij (x) is the immediate preceding operation of operation p ik (x) , then Rl i(j,k) = 1, otherwise Rl i(j,k) = 0; RC i (x) represents the resource pool of the i-th station when the x-th aircraft is being assembled in the i-th station; MC i (x) represents the material pool of the i-th station when the x-th aircraft is being assembled in the i-th station; ST i (x) represents the state of the i-th station, the state of the i-th station at time t is determined by the state of all operations in the i-th station at time t, IF i (x) = (SI i , SO i ) represents the interaction information set of the i-th station, where SI i is the input information set, including the finished information of the previous station and the finished information of the previous aircraft in the station, SO i is the output information set, i.e. the finished information of the current station.

6. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 5, wherein: In step S105, the total assembly line whole line operation process simulation model is expressed as a four-tuple: AL (x) = {S (x) , R (x) , W (x) , ST (x)}, wherein is a set of stations on the total assembly line, i is the total number of stations on the total assembly line; R (x) = (RC1 (x) , RC2 (x) , …, RC j (x) is a set of configuration information of all resources on the total assembly line; W (x) = (MC1 (x) , MC2 (x) , …, MC k (x) is a set of supply information of all materials on the total assembly line; ST (x) = (ST1 (x) , ST2 (x) , …, ST i (x) is a state of the total assembly line, and the state of the total assembly line is determined by the state of each station.

7. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 6, wherein: In step S305, the decision rule refers to the operation priority order sorting rule of the comprehensive operation man-hour coefficient, the operation human resource demand coefficient, and the operation man-hour coefficient after the operation. The establishment method is as follows: Let P be the set of jobs in the station at a simulation time t c = {p c1 , p c2 , …, p ck} and the three coefficients are described as follows: Among them, CJF ci It is the working time coefficient, t ci Indicates assignment p ci The rated working hours; 1≤i≤k; wherein HDF ci is the job human resource demand coefficient, represents the total number of resource demands for job p ci , and J represents the total number of resource types; where OTF ci is the tight following work hour coefficient, t' ci represents the total work hour sum of the work p ci . According to the above three coefficients, the assembly operation priority calculation formula is established: PEC i = λ1CJF ci + λ2HDF ci + λ3OTF ci (5) Wherein, λ1~λ3 are weight coefficients, and Generally, λ1 and λ3 are set to be the same size, and λ2 can be valued in 0~0.5, and its value is determined according to the degree of attention of the practitioner to human resources in the practice process. According to the above formula (5), the priority of each assembly operation is calculated, and the operation priority order sorting rule is obtained according to the priority size.

8. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 7, wherein: In step S401, the total assembly period of each batch is calculated according to the following formula: wherein, denotes the assembly duration of the xth aircraft at the jth station at the τth simulation cycle, and m denotes the total number of stations in the assembly line.

9. The discrete event simulation-based aircraft final assembly cell flow-through time analysis method of claim 8, wherein: In step S403, the interval estimation method is used to estimate the set d of the assembly man-days of the xth aircraft x The confidence interval of the confidence degree 1-α is: wherein, denotes the mean of the set of assembly duration days, denotes the variance of the set of assembly duration days, denotes the upper quantile of the t-distribution with n-1 degrees of freedom with respect to the mean.

Images