CN115545246A - Multi-vehicle batch flexible production line scheduling method - Google Patents

Multi-vehicle batch flexible production line scheduling method Download PDF

Info

Publication number
CN115545246A
CN115545246A CN202110719807.0A CN202110719807A CN115545246A CN 115545246 A CN115545246 A CN 115545246A CN 202110719807 A CN202110719807 A CN 202110719807A CN 115545246 A CN115545246 A CN 115545246A
Authority
CN
China
Prior art keywords
vehicle
batch
vehicle body
sequence
production line
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110719807.0A
Other languages
Chinese (zh)
Inventor
吴泽锐
刘冉
陈晓东
易延洪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202110719807.0A priority Critical patent/CN115545246A/en
Publication of CN115545246A publication Critical patent/CN115545246A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/08—Probabilistic or stochastic CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/18—Manufacturability analysis or optimisation for manufacturability

Abstract

A multi-vehicle type batch flexible production line scheduling method comprises the steps of firstly, taking configuration parameters of a production line, considering the influence of random sampling inspection events, random repair events and random workshop shutdown events, and determining evaluation parameters of solutions and parameters of an inner layer simulated annealing algorithm; then taking the following traveling vehicle type calling sequence as input, and dividing two decision variables of minimum batch and vehicle type sequencing into two layers of optimization problems, wherein: after the outer layer determines the minimum batch, the simulated annealing algorithm of the inner layer solves the sequencing scheduling problem with the minimum batch constraint to obtain the optimal production sequencing solution. The method can solve the production scheduling problem with minimum batch constraint and random event disturbance, and provides powerful decision support for batch flexible production scenes in real environments.

Description

Multi-vehicle batch flexible production line scheduling method
Technical Field
The invention relates to a technology in the field of flexible workshop production and manufacturing, in particular to a multi-vehicle batch flexible production line scheduling method.
Background
In an actual multi-vehicle flexible production line, a large Block manufacturing mode is required to be combined to realize flexible production during mixed line production. A Block is the number of vehicles that are continuously produced for the same model, also called batch minimum batch, e.g. Block =20, the next produced model can be switched after every 20 cars. In production, unified production blocks need to be established first, and then specific vehicle models produced by each Block are decided. The existing multi-vehicle type mixed line flexible production scheduling method is mostly based on the complete flexible condition of a production line, namely the actual constraint of the minimum batch in batches is not considered. In addition, the existing flexible production scheduling method still lacks consideration on random event disturbance such as spot check, repair, shutdown and the like on a production line, and is difficult to adapt to an actual production scene.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a multi-vehicle type batch flexible production line scheduling method, which can solve the production scheduling problem with minimum batch constraint and random event disturbance and provide powerful decision support for batch flexible production scenes in real environments.
The invention is realized by the following technical scheme:
the invention relates to a multi-vehicle type batch flexible production line scheduling method, which comprises the steps of firstly, calling configuration parameters of a production line, considering the influence of random sampling inspection events, random repair events and random workshop shutdown events, and determining evaluation parameters of a solution and parameters of an inner layer simulated annealing algorithm; then taking the following traveling vehicle type calling sequence as input, and dividing two decision variables of minimum batch and vehicle type sequencing into two layers of optimization problems, wherein: after the outer layer determines the minimum batch, the simulated annealing algorithm of the inner layer solves the sequencing scheduling problem with the minimum batch constraint to obtain the optimal production sequencing solution.
The configuration parameters of the production line include but are not limited to the sequence of calling multiple vehicle types from a downstream assembly workshop to the vehicle body distribution center and the like.
The multi-vehicle batch flexible production line is characterized in that: the production line is provided with S stations, a plurality of vehicle type bodies put in are sequentially processed through the processes of the S stations by taking 1 minute as 1 beat according to a production plan, and enter a body distribution center to be stored as inventory, wherein the number of the vehicle bodies continuously produced by each vehicle type is called as a batch minimum batch B (Block).
The vehicle body distribution center is as follows: and the vehicle body warehouse with the reordering function can receive processed vehicle bodies from the vehicle body workshop, arrange the processed vehicle bodies according to different vehicle types and provide required vehicle types for the final assembly workshop.
The multi-vehicle type calling sequence is as follows: at each beat, the assembly shop calls a vehicle body of a certain vehicle type from the vehicle body distribution center for overall assembly, and the calling sequence is a sequence D with the length of N vehicle bodies known in advance.
The production sequencing solution is as follows: the production plan sequence P with the length of N vehicle bodies indicates the type of the vehicle body which is put into a vehicle body workshop for processing at each beat, and the vehicle body enters a vehicle body distribution center for dispatching after being processed at S stations.
The random event of the spot check refers to: when the vehicle body enters the quality control area in each beat, the number p is used check The vehicle is subjected to probability spot inspection, and the time length T of each spot inspection check A beat. When some automobile bodies are subjected to spot inspection, the follow-up automobile bodies are still in other processes and cannot be replaced in an accelerated mode, and the fact that the automobile bodies are not output in a follow-up automobile body workshop with a certain beat means. After the sampling inspection is finished, in order to reduce the switching of vehicle models, the generator is not inserted immediatelyAnd the production line is inserted into the nearest same-vehicle type sequence, and the work station before the sampling inspection needs to be suspended for waiting during the insertion. At the time of examination, p breakdown The quality problem is found out by the probability of (2), so that the whole workshop is stopped and overhauled.
The repair random event is as follows: the vehicle body has p in the painting process repair The vehicle body is repaired and returned to a vehicle body workshop to repair the flaws, and the repaired vehicle body directly enters a vehicle body distribution center.
The workshop shutdown random event is as follows: workshop has p malfunction The time ratio of (A) is in a production line fault state, and each overhaul generally requires T breakdown And a beat. The probability of shutdown per beat is equal to p repair If the average mean time between shutdowns is longer than
Figure BDA0003136110250000021
The probability of shutdown per beat can be calculated
Figure BDA0003136110250000022
The faults are divided into two types, when the automobile body production line is stopped, the selective inspection and the repair work are not carried out, no new automobile body enters the automobile body distribution center, when the general assembly production line is stopped, the general assembly workshop does not transfer goods from the automobile body distribution center.
The evaluation parameters of the solution refer to: and the coefficient weights of the vehicle type switching times S, the part transportation times T, the part line edge storage quantity R and the vehicle body distribution center stock shortage times L are summed to obtain an evaluation index C of the sequencing solution.
The optimization target is as follows: an evaluation index of a minimum solution min.c = α · S + β · T + δ · R + σ · L, wherein: the number of times of vehicle type switching S is: the number of times of vehicle type switching is +1, an algorithm traverses a production plan sequence from the beginning, and the number of times of switching is +1 when the vehicle types produced in front and back batches are different; the number of times of part transportation T means: the transportation frequency of the parts is that for each part, when a certain vehicle type is about to produce and the corresponding parts of the line edge are insufficient, one box or two boxes of parts are transported according to the usage amount of the parts, each box of parts only corresponds to one vehicle type, and the line edge stores at most two boxes of parts, so the existing parts of the line edge are transported back according to the conditions of the next use. Every time when the parts are transported to come or returned, the transportation times of the parts are plus 1; the number of parts stored on line R means: checking the number of boxes (0 or 1 or 2) stored at the moment at each beat, wherein the number of the stored boxes at the moment is the sum of the number of the boxes; the number of times of stock shortage L of the vehicle body distribution center is as follows: and (4) the times of adjusting goods shortage in the production process, and recording the goods shortage once if the inventory of the vehicle body corresponding to the vehicle type is 0 when the adjustment of the goods is required for the vehicle body distribution center at each beat in the final assembly workshop.
The outer layer traversal comprises the following steps of roughly determining a minimum batch range and precisely searching:
the first step, roughly determining the minimum batch, and firstly determining the possible positive integer value range of the minimum batch according to actual production: (b) 1 ，b 2 ) In this range, a smaller number of simulation cycles M = M is used first 1 Traversal of the specified minimum batch b 1 ≤B≤b 2 Inputting the data into the simulated annealing algorithm of the inner layer to obtain the corresponding inner layer optimal production sequence P B To thereby determine an optimum approximate range of the minimum lot size B
Figure BDA0003136110250000032
Second, an accurate search, at (b) 3 ，b 4 ) In the range of (1), a larger number of simulation times M = M is adopted 2 ＞M 1 Traversal of the specified minimum batch b 3 ≤B≤b 4 Inputting the data into the simulated annealing algorithm of the inner layer to obtain the corresponding optimal production sequence P of the inner layer B To screen out the optimum evaluation value C (P) B ) Corresponding minimum lot B and production order P B And outputting as the optimal solution.
The inner-layer simulated annealing algorithm is based on a general assembly workshop dispatching requirement sequence D and a minimum batch B input from the outer layer, an initial sequencing solution is generated firstly, then a new solution is randomly searched in a solution neighborhood, M times of Monte Carlo simulation are carried out, the quality of the two solutions is compared, the new solution is accepted according to the Metropolis criterion until the iteration of the algorithm reaches the maximum algebraic MaxGen, the new solution is stopped being searched, and the current solution is outputOf optimal rank P B Corresponding evaluation value C (P) B )。
The initial ordering solution is as follows: on the premise of giving the minimum batch, the required quantity of various vehicle types in one day is batched, and the number of vehicles in each batch is not less than the value of the minimum batch. Each vehicle type is preferably divided according to the value of the minimum batch, and if the vehicles of less than one batch are remained after grouping, the vehicles are added into the last batch of the same vehicle type.
The neighborhood of solutions includes: (1) exchanging: randomly exchanging the positions of the two items in the sequence; (2) and (3) shifting: randomly intercepting one section of the sequence and inserting the section of the sequence into any position of the rest sequence; (3) and (3) inversion: randomly intercepting one section of the sequence, and inversely inserting the section of the sequence into the original position; when a neighborhood solution does not fit the minimum batch constraint given by the outer layer, the neighborhood solution is not feasible.
The Metropolis criterion refers to: when the evaluation index Y of the new solution is lower than the evaluation index X of the original solution, the new solution is accepted, otherwise, the probability is used
Figure BDA0003136110250000031
A new solution is accepted. Wherein H = H 0 ×A k Is represented by the initial temperature H 0 And the annealing coefficient A is less than 1, and the current iteration algebra k is less than the current annealing temperature H calculated by MaxGen.
The invention relates to a system for realizing the method, which comprises the following steps: the system comprises a production line database module, a general assembly vehicle model calling sequence input module, a scheduling algorithm parameter setting module, a scheduling algorithm solving module and a scheme output display module, wherein: the production line database module and the vehicle model calling sequence input module of the final assembly provide problem information for the scheduling algorithm solving module, various parameters of an optimization algorithm are set through the scheduling algorithm parameter setting module in production, the scheduling algorithm solving module is connected with the scheme output display module, and the optimal scheduling plan sequence obtained through solving is converted into a visual and visible production sequence.
Technical effects
The invention integrally solves the problems that the prior art does not consider the minimum batch size and can not carry out flexible production line and semi-flexible batch production line in the automobile manufacturing industry; the minimum batch is used as a decision variable and constraint of an inner-layer simulated annealing algorithm instead of a random result after scheduling through the minimum batch and specific production sequencing combined double-layer optimization, and the influence of random disturbance in production is carefully considered. Compared with the prior art, the method takes the minimum batch as one dimension of the decision variable on the whole, and designs the sequencing scheduling optimization algorithm with the minimum batch constraint at the inner layer, so that the production scheduling can be carried out according to the physical characteristic limit of the production line.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a flow diagram of an embodiment Monte Carlo simulation.
Detailed Description
In the embodiment, the production line has 133 stations, and according to a production plan, with 1 minute as 1 beat, the delivered 6 vehicle-type bodies are sequentially processed through the steps of 133 stations, and enter a body distribution center to be stored as inventory, wherein the number of the bodies continuously produced by each vehicle-type is called as a minimum batch B (Block) in batches, and since the existing fully flexible scheduling technology cannot perform production scheduling on the problem, B =20 in the current enterprise tentative production plan.
Through a production line database module and a general assembly vehicle model calling sequence input module, initial stocks of 6 vehicle models of a vehicle body distribution center are read to be respectively 50, 10, 20, 10 and 10, the total dispatching quantity of a general assembly workshop in one day is N =1200 vehicles, the calling total quantities of the 6 vehicle models are respectively 500, 100, 200, 100 and 100 vehicles, and the dispatching sequence D is as follows: firstly, the method is to mix 25 vehicle types 1, then mix 5 vehicle types 2, 10 vehicle types 3, 10 vehicle types 4,5 vehicle types 5,5 vehicle types 6, namely to mix goods circularly according to a mode of 60 vehicle types in 1 hour of one period.
According to the statistical data in the actual production, regarding random events of sampling inspection, when the vehicle body enters the quality control area in each beat, the random events are expressed as p check Probability of 3% sampling inspection time T of the vehicle check =6 beats. At the time of examination
Figure BDA0003136110250000041
The quality problem is found according to the probability of the fault, and the whole workshop is stopped and overhauled.
According to the statistical data in the actual production, the vehicle body has p in the painting process and the repair random event repair Repair occurs with a probability of =17%, the repairing is returned to a vehicle body workshop to repair the flaw, the average value E (X) of repair time duration is 10 minutes (namely 10 beats), the repair time duration is in lognormal distribution, namely the parameter is mu repair ＝1.8，σ repair =1. The repaired vehicle body directly enters a vehicle body distribution center.
According to the statistical data in the actual production, the workshop has p shutdown random events malfunction The time ratio of =6% can be in a state of production line fault, and each overhaul generally requires T breakdown =60 beats. According to the equal probability of shutdown of each beat, the shutdown probability p of each beat can be calculated repair 0.001. The faults are divided into two types, when the automobile body production line is stopped, the selective inspection and the repair work are not carried out, no new automobile body enters the automobile body distribution center, when the general assembly production line is stopped, the general assembly workshop does not transfer goods from the automobile body distribution center.
As shown in fig. 1, the present embodiment implements the scheduling method for a multi-vehicle batch flexible production line in a body shop of a certain automobile manufacturing enterprise, including:
step S1, determining an evaluation coefficient of a solution: the coefficient α =100 of the number of times S of vehicle type switching, the coefficient β =0.5 of the number of times T of parts transportation, the coefficient δ =0.5 of the number R of parts line side storage, and the coefficient σ =30 of the number L of times L of vehicle body distribution center stock out. In the evaluation indexes, about 50% is the number of times of stock shortage, 15% is the number of times of vehicle type switching, 18% is the number of times of part transportation, and 17% is the number of parts stored on the line side.
S2, determining parameters of an inner layer simulated annealing algorithm: maximum iteration algebra MaxGen =30000, initial temperature H 0 =1000, annealing coefficient a =0.95.
Step S3, first double-layer optimization: the possible value range of the minimum batch specified by the outer layer is (18, 60), and firstly, in this range, the monte carlo simulation is required to be performed on each solution of the inner layer algorithm for M =100 times, which specifically includes:
in step S301, the outer layer specifies a minimum batch B.
Step S302, initializing an inner layer, and generating an initial sequencing solution P according to a given minimum batch B B (0) And setting the current iteration number k =0 and the temperature H = H 0 ＝1000。
Step S303, evaluating the current solution as a weighted average C (P) of four indexes B (k) - = α · S + β · T + δ · R + σ · L, wherein: the number of times of vehicle type switching S is: the number of times of vehicle type switching is +1, an algorithm traverses a production plan sequence from the beginning, and the number of times of switching is +1 when the vehicle types produced in front and back batches are different; the number of times of part transportation T means: the transportation frequency of the parts is that for each part, when a certain vehicle type is about to produce and the corresponding parts of the line edge are insufficient, one box or two boxes of parts are transported according to the usage amount of the parts, each box of parts only corresponds to one vehicle type, and the line edge stores at most two boxes of parts, so the existing parts of the line edge are transported back according to the conditions of the next use. Every time when the parts are transported in or back, the number of times of transporting the parts is +1; the number of parts stored on the line side R is as follows: checking the number of boxes (0 or 1 or 2) stored at the moment at each beat, wherein the number of the stored boxes at the moment is the sum of the number of the boxes; the number of times of stock shortage L of the vehicle body distribution center is as follows: and (3) adjusting the times of the shortage of goods in the production process, and recording the shortage of goods once if the inventory of the car body of the corresponding car type is 0 when the dispatching of the goods is required for the car body distribution center at each beat in the final assembly workshop.
The disturbance of the random event is considered when evaluating the solution, and the average is obtained by simulating M times by using monte carlo simulation, as shown in fig. 2.
Step S304, a new solution P is randomly searched in the neighborhoods of (1) exchanging, (2) shifting and (3) inverting the three solutions B (new), checking the minimum batch constraint, and if the minimum batch of the new solution does not accord with the value B specified by the outer layer, searching the new solution again; if the minimum batch constraint is met, obtaining an evaluation index C (P) of a new solution according to the method of the step S303 B (new))。
Step S305, accepting according to Metropolis criterionAnd (3) new solution: when C (P) B (new))＜C(P B (k) Accept the new solution P) B (k+1)＝P B (new), otherwise with probability
Figure BDA0003136110250000051
A new solution is accepted.
Setting the iteration times k = k +1, the iteration temperature H = H × A, repeating the processes from S304 to S305 until k = MaxGen, terminating the algorithm, and outputting the optimal production sequence P under the current minimum batch B B And a corresponding index C (P) B )。
In step S306, since the evaluation index value of the solution changes to a concave curve along with the minimum batch, the minimum batch value of 10% of the lowest analysis index value is taken as the optimal range (b) of the minimum batch 3 ，b 4 )。
Step S4, second double-layer optimization is carried out in step (b) 3 ，b 4 ) The minimum batch is traversed within the range of (3), the simulation times M =5000 times of the inner layer Monte Carlo are required, and the process of S3 is repeated to obtain an accurate and optimal scheduling result.
Through specific actual experiments, the method is operated based on the actual production environment data of the enterprise, the obtained experimental result data are shown in table 1, the solution time is 10 hours, and the (b) is obtained through first double-layer optimization 3 ，b 4 ) = (45, 55), the second double-layer optimization yields the optimal value B =48 for the minimum batch, with the specific vehicle type ranking as (the number indicates the vehicle type produced by the batch): 1-1-1-1-3-3-3-3-1-4-4-4-4-1-1-1-6-1-1-2-2-5-5-6. Since the existing batch flexible scheduling technology does not exist, the enterprise tentative scheme is the minimum batch B =20, which is roughly obtained by a table static deduction method, and the specific vehicle type sequence is also difficult to solve by the existing technology, but is solved by using the inner-layer optimization algorithm of the embodiment under the condition of B =20.
Table 1:
Figure BDA0003136110250000061
as described above, the method provides a production scheduling method capable of considering minimum batch constraints for the first time, and is a new technology never appearing in the field, and compared with a scheduling scheme designed by an enterprise by using a form static deduction method, the performance index of the method is improved in that the influence of random event disturbance is considered, the obtained optimal solution quality is improved by 45.87% compared with the enterprise scheme, in addition, the algorithm solution only needs 10 hours, and the production plan of the 2 nd day can be made 1 day in advance.
The foregoing embodiments may be modified in many different ways by one skilled in the art without departing from the spirit and scope of the invention, which is defined by the appended claims and not by the preceding embodiments, and all embodiments within their scope are intended to be limited by the scope of the invention.

Claims (10)

1. A multi-vehicle type batch flexible production line scheduling method is characterized in that configuration parameters of a production line are firstly called, the influence of random inspection events, random repair events and random workshop shutdown events is considered, and evaluation parameters of a solution and inner layer simulated annealing algorithm parameters are determined; then taking the following traveling vehicle type calling sequence as input, and dividing two decision variables of minimum batch and vehicle type sequencing into two layers of optimization problems, wherein: after the outer layer determines the minimum batch, the simulated annealing algorithm of the inner layer solves the sequencing scheduling problem with the minimum batch constraint to obtain the optimal production sequencing solution;
the multi-vehicle batch flexible production line is characterized in that: the production line is provided with S stations, a plurality of vehicle-type bodies put in are sequentially processed through the working procedures of the S stations according to a production plan and with 1 minute as 1 beat, and enter a vehicle body distribution center to be stored as inventory, wherein the quantity of the continuously produced vehicle bodies of each vehicle type is called as the minimum batch B of batches;
the random event of the spot check refers to: when the vehicle body enters the quality control area in each beat, the number p is used check The vehicle is subjected to probability spot inspection, and the time length T of each spot inspection check A beat; when the vehicle body is subjected to spot inspection, the subsequent vehicle body is still in other processes and cannot be replaced in an accelerated way, which means that a vehicle body workshop with a certain beat will not output the vehicle body; after the sampling inspection is finished, the method comprises the following stepsThe switching of vehicle types is reduced, the vehicle types are not immediately inserted into a production line but are inserted into a nearest same vehicle type sequence, and stations before spot inspection need to pause for waiting during insertion; at the time of examination, p breakdown The quality problem is found out, so that the whole workshop is stopped and overhauled;
the repair random event is as follows: the vehicle body has p in the painting process repair The vehicle body is repaired and returned to a vehicle body workshop to repair the flaws, and the repaired vehicle body directly enters a vehicle body distribution center;
the workshop shutdown random event is as follows: workshop has p malfunction The time ratio of (2) is in the state of production line fault, and each overhaul generally requires T breakdown A beat; the probability of shutdown per beat is equal to p repair If the average mean time between shutdowns is longer than
Figure FDA0003136110240000011
The probability of shutdown per beat can be calculated
Figure FDA0003136110240000012
When the vehicle body production line is stopped, the selective inspection and the repair work are not carried out, no new vehicle body enters a vehicle body distribution center, and when the general assembly production line is stopped, the general assembly workshop does not transfer goods from the vehicle body distribution center;
the evaluation parameters of the solution refer to: the method comprises the following steps of (1) obtaining a model switching frequency S, a part transportation frequency T, a part line edge storage quantity R and a coefficient weight of a vehicle body distribution center stock shortage frequency L, and summing to obtain an evaluation index C of a sequencing solution;
the vehicle body distribution center is as follows: a vehicle body warehouse with a reordering function is capable of receiving processed vehicle bodies from a vehicle body workshop, arranging the processed vehicle bodies according to different vehicle types, and providing required vehicle types for a final assembly workshop.
2. The multi-vehicle type batch flexible production line scheduling method of claim 1, wherein the multi-vehicle type scheduling sequence is as follows: at each time, the general assembly workshop calls a vehicle body of a certain vehicle type from the vehicle body distribution center for general assembly, and the calling sequence is a sequence D with the length of N vehicle bodies known in advance.
3. The multi-vehicle type batch flexible production line scheduling method of claim 1, wherein the production sequencing solution is: the production plan sequence P with the length of N vehicle bodies indicates the type of the vehicle body which is put into a vehicle body workshop for processing at each beat, and the vehicle body enters a vehicle body distribution center for dispatching after being processed at S stations.
4. The multi-vehicle batch flexible production line scheduling method of claim 1, wherein the optimization objective is: the evaluation index of the minimum solution min.c = α · S + β · T + δ · R + σ · L, wherein: the number of times s of vehicle type switching is: the number of times of vehicle type switching is +1, an algorithm traverses a production plan sequence from the beginning, and the number of times of switching is +1 when the vehicle types produced in front and back batches are different; the number of times of part transportation T means: the transportation frequency of the parts, for each part, when a certain vehicle type is about to produce and the corresponding parts of the line edge are insufficient, the parts are transported to one box or two boxes of parts according to the usage amount of the parts next, each box of parts only corresponds to one vehicle type, and the line edge stores at most two boxes of parts, so the existing parts of the line edge are transported back according to the next usage condition, and the transportation frequency of the parts is plus 1 when every part is transported or transported back; the number of parts stored on line R means: checking the number of boxes stored at the moment at each beat, wherein the number of the boxes stored at the moment is the sum of the number of the boxes; the number L of times of stock shortage of the vehicle body distribution center refers to: and (3) adjusting the times of the shortage of goods in the production process, and recording the shortage of goods once if the inventory of the car body of the corresponding car type is 0 when the dispatching of the goods is required for the car body distribution center at each beat in the final assembly workshop.
5. The method as claimed in claim 1, wherein the outer layer traversal comprises coarse determination of minimum batch range and fine search:
first, roughly determine the minimum batch size, from the production lotThe possible minimum batch positive integer value range is firstly determined: (b) 1 ，b 2 ) In this range, firstly, a small number of times of simulation M = M is used 1 Traversal of the specified minimum batch b 1 ≤B≤b 2 Inputting the data into the simulated annealing algorithm of the inner layer to obtain the corresponding inner layer optimal production sequence P B To thereby determine an optimum approximate range of the minimum lot size B
Figure FDA0003136110240000021
Second, an accurate search, at (b) 3 ，b 4 ) In the range of (1), a larger number of times of simulation is adopted, M = M 2 ＞M 1 Traversal of the specified minimum batch b 3 ≤B≤b 4 Inputting the data into the simulated annealing algorithm of the inner layer to obtain the corresponding inner layer optimal production sequence P B Screening out the optimum evaluation value C (P) B ) Corresponding minimum lot B and production order P B And outputting as the optimal solution.
6. The multi-vehicle batch flexible production line scheduling method as claimed in claim 1, wherein the inner-layer simulated annealing algorithm is based on a general assembly plant dispatching requirement sequence D and a minimum batch B input from an outer layer, an initial sequencing solution is generated firstly, then a new solution is randomly searched in a neighborhood of the solution, M times of Monte Carlo simulation are carried out, the quality of the two solutions is compared, the new solution is accepted according to Metropolis criterion until the algorithm iteration reaches a maximum algebra MaxGen, the new solution is stopped being searched, and the current optimal sequencing P is output B Corresponding evaluation value C (P) B )。
7. The multi-vehicle type batch flexible production line scheduling method as claimed in claim 1, wherein the initial sequencing solution is: on the premise of giving the minimum batch, the required quantity of various vehicle types in one day is batched, the number of vehicles in each batch is not less than the value of the minimum batch, each vehicle type is preferably divided according to the value of the minimum batch, and the remaining vehicles in less than one batch are added into the last batch of the same vehicle type after grouping.
8. The multi-vehicle batch flexible production line scheduling method of claim 1, wherein said neighborhood of solutions comprises: (1) exchanging: randomly exchanging the positions of the two items in the sequence; (2) and (3) displacement: randomly intercepting one section of the sequence and inserting the section of the sequence into any position of the rest sequence; (3) and (3) inversion: randomly intercepting one section of the sequence, and inversely inserting the section of the sequence into the original position; when a neighborhood solution does not fit the minimum batch constraint given by the outer layer, the neighborhood solution is not feasible.
9. The method as claimed in claim 1, wherein the Metropolis criterion is that: when the evaluation index Y of the new solution is lower than the evaluation index X of the original solution, the new solution is accepted, otherwise, the probability is used for judging whether the evaluation index Y of the new solution is lower than the evaluation index X of the original solution or not
Figure FDA0003136110240000031
Accepting a new solution, where H = H 0 ×A k Is represented by the initial temperature H 0 And the annealing coefficient A is less than 1, and the current iteration algebra k is less than the current annealing temperature H calculated by MaxGen.
10. A system for implementing the multi-vehicle batch flexible production line scheduling method of any one of claims 1 to 9, comprising: the system comprises a production line database module, a general assembly vehicle model calling sequence input module, a scheduling algorithm parameter setting module, a scheduling algorithm solving module and a scheme output display module, wherein: the production line database module and the general assembly vehicle model calling sequence input module provide problem information for the scheduling algorithm solving module, various parameters of the optimization algorithm are set through the scheduling algorithm parameter setting module in production, the scheduling algorithm solving module is connected with the scheme output display module, and the optimal scheduling plan sequence obtained through solving is converted into a visual and visible production sequence.
CN202110719807.0A 2021-06-28 2021-06-28 Multi-vehicle batch flexible production line scheduling method Pending CN115545246A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110719807.0A CN115545246A (en) 2021-06-28 2021-06-28 Multi-vehicle batch flexible production line scheduling method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110719807.0A CN115545246A (en) 2021-06-28 2021-06-28 Multi-vehicle batch flexible production line scheduling method

Publications (1)

Publication Number Publication Date
CN115545246A true CN115545246A (en) 2022-12-30

Family

ID=84717360

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110719807.0A Pending CN115545246A (en) 2021-06-28 2021-06-28 Multi-vehicle batch flexible production line scheduling method

Country Status (1)

Country Link
CN (1) CN115545246A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116029623A (en) * 2023-02-17 2023-04-28 希维科技(广州)有限公司 Method, apparatus and storage medium for constructing check flow model
CN117273392A (en) * 2023-11-16 2023-12-22 四川省致链数字科技有限公司 Furniture production decision method and device, electronic equipment and storage medium
CN117314142A (en) * 2023-09-15 2023-12-29 中国人民解放军海军工程大学 Product line process sequence optimization method

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116029623A (en) * 2023-02-17 2023-04-28 希维科技(广州)有限公司 Method, apparatus and storage medium for constructing check flow model
CN117314142A (en) * 2023-09-15 2023-12-29 中国人民解放军海军工程大学 Product line process sequence optimization method
CN117273392A (en) * 2023-11-16 2023-12-22 四川省致链数字科技有限公司 Furniture production decision method and device, electronic equipment and storage medium
CN117273392B (en) * 2023-11-16 2024-02-09 四川省致链数字科技有限公司 Furniture production decision method and device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN115545246A (en) Multi-vehicle batch flexible production line scheduling method
CN110794788B (en) Production scheduling device, method, equipment and computer readable storage medium
CN111191822A (en) Steel parameter prediction method and terminal equipment
CN110245842B (en) Production line risk scheduling method for sudden large faults of equipment
CN115775123A (en) Scheduling plan generation method and system, electronic device and readable storage medium
CN112818484A (en) Physical entity digital twin comprehensive implementation capability assessment method and system
Davari et al. An integrated simulation-fuzzy model for preventive maintenance optimisation in multi-product production firms
Apornak et al. A simulation-based multi-objective optimisation approach in flexible manufacturing system planning
Aydarov et al. Remote monitoring system for quality assessment of car service enterprises
Lim et al. Hybrid approach to distribution planning reflecting a stochastic supply chain
CN114819379A (en) Multi-model position marker collinear installation, debugging and rescheduling method based on improved NSGA-II algorithm
Anthony et al. The History of Lean Six Sigma
Smith et al. Discrete-event simulation and machine learning for prototype composites manufacture lead time predictions
Imai Front Loading: Key Concept of Strategy for Business Innovation in Japanese Automobile Industry
Kazemi Esfeh et al. Flexible flow shop scheduling problem with reliable transporters and intermediate limited buffers via considering learning effects and budget constraint
Alszer et al. Complex approach to the control of car body buffering in the paint shop
JP2002073146A (en) Facility load simulation device
Krstovski et al. Design of experiment base scenario analysis in discrete event simulation of multi-path manufacturing system
Li et al. A generalized stochastic petri-net model for performance analysis and allocation optimization of a particular repair system
CN116384257B (en) Method for predicting assembly errors and optimizing tolerance of air separation integral cold box
Ringen Value chain data integrity and quality–a case study
Rahmani et al. Designing the retouch line in an automobile factory using a simulation-based optimization technique; a real case study
CN114115156B (en) Rescheduling method for flexible job shop
Åström et al. FLOW SIMULATION OF AN OUTPUT SHAFT LINE: Capacity analysis and optimization of the overall process efficiency
Yilmazlar Modeling and Solution Methodologies for Mixed-Model Sequencing in Automobile Industry

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination