Disclosure of Invention
In order to solve the problem of influence of various disturbance factors of a production line on production line scheduling, the invention provides an optimized scheduling method and system based on disturbance prediction, and the specific scheme is as follows:
an optimized scheduling method based on disturbance prediction comprises the following steps:
s1: constructing a probability model of non-explosion of a split line cache and non-disconnection of a main line production;
s2: optimizing production line production scheduling based on a normalization algorithm;
s3: repeating S2 according to the different waiting time between two adjacent production sequences, obtaining the combination of the waiting time and the actual production quantity on the split line, thereby carrying out global optimization.
Preferably, the constructing of the probabilistic model of the tiled line cache non-explosion and the mainline production non-break in the step S1 includes the following steps:
s11: collecting historical data, and simulating the probability distribution p (n) of the number n of the produced parts in the known unit time T and the piecing random inspection probability p (a)j) The probability p (b) of putting back in piecesj) Main lineProbability of spot check p (a)0) And a main line put back probability p (b)0) (ii) a And simulating the production quantity Dj (T) of the type j vehicle in the main line sequence at any time T0,t),T0Representing the time when the split line production of the last production sequence is completed;
s12: simulating the actual production quantity Qj on the split line at time t under the condition of sampling inspection and replacement in the step S11, thereby simulating the quantity B of the part cache region produced on the split line at time tj(t);
S13: combining with the step S12, the probability p of the main line production without line break is calculated through simulation1And cache area non-explosion probability p2And optimizing production line production scheduling.
Preferably, the data collected in step S11 includes the history data of the tact on each of the split lines and the number of historical spot checks and returns in the factory, and the tact distribution is simulated based on the history data of the tact.
Preferably, the tact distribution conforms to a normal distribution N (μ, σ)
2) Wherein the mean value is mu, and the fluctuation rate is sigma; then, given that the probability of the number n of parts produced per unit time T is p (n | μ, σ), the formula for the probability is derived
Wherein the parameters
k is the intermediate variable of the integral.
Preferably, the number of spot checks in step S11 satisfies a poisson distribution per unit time
The number of said setbacks also satisfies the poisson distribution
The piecemeal casual inspection probability is
The probability of putting back the sub-split is
Wherein j represents different types of vehicle models, aj and bj are respectively the sampling number and the replacement number on the split line, j and theta
jIs a parameter; the probability of dominant line spot check is
The dominant line has been put back to
Where a0 and b0 are the number of spot checks and returns on the main line, λ 0 and θ, respectively
0Is a parameter; the calculation of the parameters here is solved by the maximum likelihood method.
Preferably, the actual production quantity Qj-n-a on the split line at time t
j+b
j-a
0+b
0Wherein n is the production quantity of the split lines in unit time; in step S12
Wherein j represents different vehicle types; bj (t0) represents the number of parts in the cache region on the patched line at initial time t 0; qj0 is the total number of parts of the type required on the part line for a production sequence that is preset artificially, and the preset value is different according to different production sequences.
Preferably, the probability of the main line production being uninterrupted in step S13 is p1=pr{max[Dj(T0T) -Bjt is not more than delta, namely the mathematical model of the main line production is not broken; the probability of the buffer area not exploding is p2 ═ prmaxjBjt-DjT0, t is less than or equal to Z, namely the mathematical model of the buffer area not exploding; wherein, the 'delta' and the 'Z' are different values set manually according to different main line production sequences, and the Z is the highest cache part quantity.
Preferably, step S2 is: definition p ═ p1×α1+p2×α2+α3×QjSetting different delta values, searching out the maximum p and the corresponding Qi(ii) a Here α 1, α 2, α 3 are parameters set for normalization.
Preferably, the system for optimizing the scheduling method based on disturbance prediction comprises a split line cache non-explosion probability construction module, a main line production non-breakage probability construction module, a production line scheduling optimization module and a global scheduling optimization module; the sub-patched line cache non-explosion probability construction module and the mainline production non-interruption probability construction module construct a sub-patched line cache non-explosion probability model and a mainline production non-interruption probability model; the production line scheduling optimization module optimizes production line scheduling based on a normalization algorithm; and the global scheduling optimization module obtains the combination of the waiting time and the actual production quantity on the split line to perform global optimization.
The invention also discloses a computer readable storage medium, on which a computer program is stored, which, when executed, performs the optimal scheduling method based on disturbance prediction according to any one of claims 1 to 8.
The invention has the beneficial effects that:
the method comprises the steps of predicting various disturbance factors on a production line, and obtaining the line non-breaking probability of main line production and the non-explosion probability of a cache region according to the simulation of production beats, the probability simulation of production quantity in unit time T, the probability simulation of selective inspection and replacement in unit time on a main line and a sub-assembly line, the simulation of actual production quantity on the sub-assembly line and the quantity simulation of the part cache region produced on the sub-assembly line.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, an optimized scheduling method based on disturbance prediction includes the following steps:
s1: constructing a probability model of non-explosion of a split line cache and non-broken line of main line production;
s2: optimizing production line production scheduling based on a normalization algorithm;
s3: repeating S2 according to the different waiting time between two adjacent production sequences, obtaining the combination of the waiting time and the actual production quantity on the split line, thereby carrying out global optimization.
As shown in fig. 2, the constructing of the probabilistic model of the tiled cache non-explosion and the mainline production non-break in step S1 includes the following steps:
s11, (1) collecting historical data of the production beats of each piecing line, simulating and predicting the distribution of the production beats, and simulating the probability distribution of the number n of the produced parts in known unit time T; the production beat refers to the time of continuously finishing another part on the split line at intervals in the midway and corresponds to the production speed of the part. The actual production cycle may be different in the production cycle and the expected setting on the split line due to different equipment, such as different equipment agesDeviation in degree. If its tact satisfies normal distribution N (mu, sigma)
2) Wherein the mean value is mu, and the fluctuation rate is sigma; then the probability of the number n of the produced parts per unit time T is known to be p (n | mu, sigma), and the formula for the probability is derived
Wherein the parameters
k is an integrated intermediate variable; calculating different probabilities of different numbers n of products produced in unit time T, and listing.
(2) Collecting historical spot inspection conditions in a factory, and respectively simulating spot inspection probability distribution on a split line and spot inspection probability distribution on a main line; and simulating the playback probability distribution on the split joint line and the playback probability distribution on the main line. It is assumed here that the different types of components are independently and identically distributed. Taking a single piecemeal line as an example, according to the actual vehicle spot-check arrangement in a workshop, it is assumed that the spot-check time is relatively uniformly distributed, and the number of spot-checks satisfies the poisson distribution within a certain unit time
Wherein lambda is a parameter, and a is the number of spot checks; the number of the spot checks in the known unit time on the split line meets the probability that
Wherein j represents different kinds of vehicle types, a
jFor sampling the number of pieces on the piecing line, lambda
jIs a parameter; the number satisfaction probability of the spot checks on the main line is
Wherein a is
0As number of spot checks on the main line, λ
0Is a parameter; the calculation of the parameters here is solved by the maximum likelihood method.
Meanwhile, random time points for replacing the car with the car are set. It is assumed here that the time put back does not affect the mainline production and the piecemeal production. Known noteThe number of replacement bits in bit time also satisfies the poisson distribution
Wherein θ is a parameter and b is a number of setbacks; the probability that the number of the sampling inspection returns in unit time on the splitting line meets the requirement is
Wherein j represents different types of vehicle types, and bj is the number put back on the split line; the number of the putting back on the main line satisfies the probability that
Wherein b is
0For number of main line returns, θ
0Is a parameter; the calculation of the parameters here is solved by the maximum likelihood method.
(3) Simulating T according to the checked production queuing condition of the main line1Production quantity D of j type vehicles in main line sequence at any momentj(T0T), where T0And the time of the production completion of the split line of the last order is shown.
Firstly, at time t, the number D of vehicles in the main line sequence type jj(T0T), the simulation can be predicted from the inspected production queuing case of the mainline. For example, assume that the production sequence on the main line is fixed as AAABB, i.e., one production sequence is to produce 3A and 2B in order, and the time for production of A is TAProduction of B is at time TBThen the time for producing a sequence is 3TA+2TBThen at time t, the number D of the type A vehicle models is producedA(T0T) and number of vehicles producing type B DB(T0And t) are obvious.
S12: simulating the number B of parts cache regions produced on the split line at time t under the condition of the selective inspection and the replacement in the step S11j(t),Bj(t)=Bj(t0)+min(Qj0,Qj(n|μ,σ,λj,θj) ); wherein j represents different vehicle types; b isj(t0) Indicates the initial time t0The number of parts in the cache region on the split line is increased; qj0The preset value is an artificial preset value, namely the total number of the parts of the type needed by a production sequence on the split line, and different values are required to be artificially preset according to different production sequences; qjThe number of the parts actually produced on the splitting line at the moment t; and t belongs to [0 ], the total time of the current piecing production + change-time + t0]Here, change-time is the switching time for switching different production vehicle types; and Qj(n|μ,σ,λj,θj)=n-aj+bj-a0+b0(ii) a Wherein n is the production number of the splicing lines in unit time.
S13: combining with step S12, calculating the probability p of no disconnection in the main line production by simulation1And cache region non-explosion probability p2Optimizing production line production scheduling; wherein, the probability model of the continuous line of the main line production is p1=pr{max[Dj(T0,t)-Bj(t)]Delta is less than or equal to }; the probability model of the cache region not exploding is p2=pr{max∑j[Bj(t)-Dj(T0,t)]Less than or equal to Z }; wherein, the 'delta' and the 'Z' are different values which are artificially set according to different main line production sequences, and the Z is the highest cache part number.
In step S2, the production line is optimized, and p is defined as p1×α1+p2×α2+α3×Qj(n|μ,σ,λj,θj) Setting different delta values, searching out the maximum p and the corresponding Qj(n|μ,σ,λj,θj) (ii) a Here α 1, α 2, α 3 are parameters set for normalization.
In step S3, global optimization is performed, and it is necessary to obtain station information on each split line to obtain first station information of a next sequence produced on the split line, where a time when a previous production sequence is finished is indicated by a time when the first station finishes feeding. Artificially setting a waiting time T according to the informationwaitAfter that, the air conditioner is started to work,and sending a command to the splitting line, and carrying out the production of the next sequence by the production line according to the command. Step S2 is repeated according to different waiting times, resulting in a combination of waiting time and production quantity, thereby performing global optimization. Search for the largest p and corresponding Q in step S3jFor the production of the next sequence of commands to the production line. p and QjThe command of the next production sequence is started as soon as the command is acquired, and if the command is issued as soon as the command is acquired, namely the command of the next production sequence is started as soon as the production sequence is produced, the main line speed of the next production sequence can not keep up, and the buffer area can be burst. If a waiting time T is setwaitThat is, the waiting time T is passedwaitAnd then, the command is sent to the production line, so that the phenomenon of burst in the cache area can be avoided.
About TwaitIs related to the speed difference on the main line and the split line and the computing power of the computer. (1) For a production sequence, because the production speed of the split line is greater than that of the main line, the split line needs to be stopped for a period of time after being produced for a period of time, namely TwaitIf the main line production speed is fast, namely the main line consumes the sub-assembly parts fast, namely the speed difference between the main line and the sub-assembly line is small, the waiting time T is neededwaitIt is a little shorter; if the main line production speed slow point, i.e. the main line consumes the piecing parts slow point, i.e. the speed difference between the main line and the piecing line is a little greater, the waiting time T is neededwaitIt is relatively long. (2) If the computer has good computing power, a plurality of groups of production sequences can be computed, and the more the production sequences are computed at one time, the corresponding waiting time T iswaitThe longer. Based on the calculation power, T can be searchedwait0,60s,120s,180s and the like. And (6) solving an optimal result through global optimization.
In addition, when the signal is broken in the middle of the production line, the production beat is disturbed, and the production beat satisfies the normal distribution
Then the probability of the number n of parts produced per unit time T being known is satisfied
Referring to fig. 3, the present invention further discloses an optimized scheduling system based on disturbance prediction, which includes: the system comprises a split line cache non-explosion probability construction module, a main line production non-breakage probability construction module, a production line scheduling optimization module and a global scheduling optimization module; the sub-patched line cache non-explosion probability construction module and the mainline production non-interruption probability construction module construct a sub-patched line cache non-explosion probability model and a mainline production non-interruption probability model; the production line scheduling optimization module optimizes production line scheduling based on a normalization algorithm; and the global scheduling optimization module is used for obtaining the combination of the waiting time and the actual production quantity on the split line and carrying out global optimization.
The split line cache non-explosion probability model and the probability model of the main line production non-broken line comprise: collecting historical production beat data, simulating probability distribution p (n) of the number n of the produced parts in unit time T, collecting historical data of historical sampling inspection and quantity of returned parts in a factory, and respectively simulating sub-assembly sampling inspection probability p (a)j) The probability p (b) of putting back in piecesj) Main line spot check probability p (a)0) And a main line put back probability p (b)0) And simulating the production quantity Dj (T) of the type j vehicles in the main line sequence at any time T0,t),T0Representing the time when the split line production of the last production sequence is completed; simulating the actual production quantity Q on the split joint line at the time t according to the simulated split joint line and the probability distribution on the main line which is put back by the selective inspectionjSo as to simulate the number B of part cache regions produced on the split line at the time tj(t) thereby obtaining said patchwork cache non-explosion probability model p1=pr{max[Dj(T0,t)-Bj(t)]Delta is not more than is equal to delta; the probability model of non-explosion of the cache region is p2=pr{max∑j[Bj(t)-Dj(T0,t)]Less than or equal to Z }; wherein, the 'delta' and the 'Z' are different values set manually according to different main line production sequences, and the Z is the highest cache part quantity.
The tact distribution conforms to a normal distribution N (mu, sigma)
2) Wherein the mean value is mu, and the fluctuation rate is sigma; then, given that the probability of the number n of parts produced per unit time T is p (n | μ, σ), the formula for the probability is derived
Wherein the parameters
k is the intermediate variable of the integral.
The number of the sampling inspection satisfies Poisson distribution in unit time
The number of said setbacks also satisfies the poisson distribution
The piecemeal casual inspection probability is
The probability of putting back the sub-split is
Wherein j represents different types of vehicle types, aj and bj are respectively the sampling number and the replacing number on the split line, and lambada j and theta
jIs a parameter; the probability of dominant line spot check is
The dominant line has been put back to
Wherein a is
0And b
0Are respectively on the main lineNumber of spot checks and number of returns, λ
0And theta
0Is a parameter; the calculation of the parameters here is solved by the maximum likelihood method.
Actual production quantity Q on split line at time tj=n-aj+bj-a0+b0And n is the production quantity of the split lines in unit time.
The number of the buffer areas Bj(t)=Bj(t0)+min(Qj0,Qj) (ii) a Wherein j represents different vehicle types; b isj(t0) Indicates the initial time t0The number of parts in the cache region on the split line is increased; qj0The total number of the parts required on the splicing line for a production sequence preset artificially needs to be preset with different values according to different production sequences.
The production line scheduling optimization module optimizes production line scheduling based on a normalization algorithm, wherein the optimization algorithm is defined as p ═ p1×α1+p2×α2+α3×QjSetting different delta values, searching out the maximum p and the corresponding Qj(ii) a Here α 1, α 2, α 3 are parameters set for normalization.
The global optimization module optimizes the global situation, station information on each split line needs to be obtained, first station information of a next sequence produced on the split lines is obtained, the time when the first station finishes feeding indicates the time when a previous production sequence finishes, and a waiting time T is set artificially according to the informationwaitThen, sending a command to the production line, and carrying out production of the next sequence by the production line according to the command; repeatedly calculating p as p according to different waiting time1×α1+p2×α2+α3×QjObtaining the waiting time TwaitAnd production quantity QjTo perform global optimization. Search for the largest p and corresponding QjFor the production of the next sequence of commands to the production line. p and QjIs ready to be acquired, but is not commanded to start the next production sequence as soon as it is acquired, i.e. is commanded immediately upon acquisitionIf a command for starting the next production sequence just after production in one production sequence causes the main line speed of the next production sequence to be unable to keep up, the cache area may burst. If a waiting time T is setwaitThat is, the waiting time T is passedwaitAnd then, the command is sent to the production line, so that the phenomenon of burst in the cache area can be avoided.
Various disturbance factors on the production line are predicted, and the scheduling of the production line is optimized according to the prediction result, so that the production of the production line is not broken, and the number of parts in the cache region is not exploded.
The invention also discloses a computer readable storage medium, wherein a computer program is stored on the medium, and after the computer program runs, the optimized scheduling method based on the disturbance prediction is executed.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.
In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a web site, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk (disk) and disc (disc), as used herein, includes Compact Disc (CD), laser disc, optical disc, Digital Versatile Disc (DVD), floppy disk and blu-ray disc where disks (disks) usually reproduce data magnetically, while discs (discs) reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.