Process logistics scheduling method and system

Abstract

The present invention provides a process logistics scheduling method and system, including: using a multi-objective group intelligent algorithm based on trend neighborhood search for scheduling optimization, including: population initialization step: constructing the chromosome construction of individuals in the population, and performing initialization operations on all individuals in the population; mutation operation step: generating new individuals by randomly changing certain genes on the chromosome to form a progeny population, increasing the diversity of the entire population, and expanding the search space, including exchange mutation or reverse mutation; selection operation step: simulating the characteristics of survival of the fittest in natural selection, by calculating the objective function value of each individual in the solution population, obtaining the fitness value of the individual, and allowing individuals with good fitness to survive in iterations, while keeping the size of the population constant. The present invention can ensure the consistency of chromosome design with the process, and at the same time ensure the effectiveness of population evolution, simple coding, simple rules, and easy operation.

Description

Technical Field

[0001] The present invention relates to the technical field of process logistics coding and process optimization scheduling, and in particular to a process logistics scheduling method and system. Background Art

[0002] Process optimization scheduling is a complex scheduling optimization problem. Traditional process optimization scheduling often uses a single chromosome chain to encode process logistics, which cannot effectively represent the complexity of the process and the order of the process, or requires the design of complex constraints to ensure the effectiveness of population evolution. The rules are complex and not easy to operate.

[0003] The invention patent with publication number CN105096011A discloses a logistics transportation scheduling method based on improved chromosome coding, including the following steps: 1) encoding; 2) initializing the population; 3) calculating the fitness of each individual, and the fitness is an indicator value used to measure the quality of individuals in the population; 4) determining which individuals can enter the next step by judging the fitness of each individual; 5) performing a crossover operation according to probability Pc; 6) performing a mutation operation according to probability Pm; 7) judging whether the fitness of the optimal individual meets the given conditions or the fitness of the individual no longer improves after repeated crossover and mutation. If so, the algorithm iteration process converges and the algorithm ends; otherwise, go to step 3) and enter the iterative operation; 8) the algorithm outputs the optimal solution. Summary of the invention

[0004] In view of the defects in the prior art, the present invention provides a process logistics scheduling method and system.

[0005] According to a process logistics scheduling method and system provided by the present invention, the scheme is as follows:

[0006] In the first aspect, a process logistics scheduling method is provided, which uses a multi-objective swarm intelligence algorithm based on trendy neighborhood search to perform scheduling optimization, including:

[0007] Population initialization step: construct the chromosomes of individuals in the population and perform initialization operations on all individuals in the population;

[0008] The chromosomes include process-machine chromosomes and process-processing feature chromosomes; wherein the process-machine chromosomes represent the selection of machines and fixtures for each process; the process-processing feature chromosomes represent the order of processing features on the machine tool selected for each process, as well as the processing scheme and tool for the processing features;

[0009] Mutation operation steps: Generate new individuals by randomly changing certain genes on chromosomes to form offspring populations, increase the diversity of the entire population, and expand the search space;

[0010] Selection operation steps: simulate the characteristics of survival of the fittest in natural selection, obtain the fitness value of each individual in the population by calculating the objective function value of the individual, and make the individuals with good fitness survive in the iteration while keeping the size of the population constant.

[0011] Preferably, the method adopts minimization of total processing cost and total processing time as optimization objectives, and adopts the following search trend rules:

[0012] Trend rule 1: When selecting the processing machine for each process, the probability of selecting a low-priced machine is greater than the probability of selecting a high-priced machine;

[0013] Trend rule 2: When selecting the processing scheme for each feature, the probability of selecting the processing scheme with less time is greater than the probability of selecting the processing scheme with more time;

[0014] Trend rule 3: When selecting the machining tool for the current machining plan of each feature, the probability of a high-priced tool being selected is greater than the probability of a low-priced tool being selected.

[0015] Preferably, the population initialization comprises the following steps:

[0016] Step S1: construct a certain body chromosome in the initial solution from the process-machine chromosome; starting from the first process, select the processing machine that can be used in the current process, and follow the trend rule 1 when selecting; when a machine for a certain process is selected, the fixture of the machine is determined accordingly;

[0017] Step S2: Select and sort the features to be processed on this machine in the current process;

[0018] Step S3: Select a machining tool for the current feature machining plan, and follow trend rule 3 when selecting;

[0019] Step S4: After arranging all the features that can be processed in a process, create the next process and repeat steps S1 to S3 until all the processing features are sorted;

[0020] Step S5: When the process-machine chromosome and the process-processing feature chromosome are all constructed, the chromosome construction of an individual in the population is completed, and steps S1 to S4 are repeated until all individuals in the population complete the initialization operation.

[0021] Preferably, the step S2 specifically includes: randomly selecting a feature from the processing features that meet the sequence requirements. If the feature appears for the first time, it represents feature step 1, and at this time, a processing scheme for processing this feature needs to be selected; the processing scheme must be a processing scheme that can be used by the current machine;

[0022] If it is a feature that appears for the second time, there is no need to select a processing plan, and directly follow the processing plan of the first step of this feature; follow the tendency rule 2 when selecting a processing plan.

[0023] Preferably, the selection operation step adopts the roulette wheel selection method based on the multi-objective weighted sum, and the specific operation is as follows:

[0024] (1) After normalizing all objective function values to the interval [0, 1], assign weights with practical significance to each objective function, and then aggregate all objective functions into a single objective function, that is, convert the optimization problem of multiple objectives into a single objective optimization problem of the following form by the weighted sum method:

[0025]

[0026] where the weight coefficient λ i ≥0, representing the relative importance degree of each objective function, and Ω is the set of feasible solutions;

[0027] (2) Calculate the fitness Fit j (x) (j = 1, 2, 3,..., N) of each individual in the population according to the aggregated single objective function, where N is the population size; when the optimization objective of the problem is to minimize the objective function F j (x) (j = 1, 2, 3,..., N), the smaller the objective function, the larger the fitness value Fit i of the individual, and the easier the individual is to be selected;

[0028] (3) Calculate the probability P j that each individual in the population is inherited to the next generation population;

[0029]

[0030] (4) Calculate the cumulative probability q k of each individual in the population:

[0031]

[0032] (5) Generate a uniformly distributed pseudo-random number r1 in the interval [0, 1];

[0033] (6) If q[k - 1] < r1 ≤ q[k], then select individual k;

[0034] (7) Repeat steps (5) and (6) a total of N times, that is, form the next generation parent population C i+1 with size N.

[0035] Preferably, the process-machine chromosome and process-processing feature chromosome are aligned on the process arrangement sequence.

[0036] Preferably, the process-machine chromosomes are arranged in the order of the processes, including a machine chromosome chain and a fixture chromosome chain;

[0037] The machine chromosome chain represents the processing machine selected in a certain process, and uses the machine number in the form of an integer as the encoding format;

[0038] The fixture chromosome chain represents the tooling fixture used in the process, and the encoding method is in the form of machine number + fixture number in the form of an integer.

[0039] Preferably, the process-processing feature chromosome represents the sequential processing order of workpiece features, including a processing feature chromosome chain, a feature processing scheme chromosome chain and a tool chromosome chain;

[0040] The processing feature chromosome chain represents the chromosome code corresponding to the workpiece feature, and uses the workpiece number + feature number in the form of an integer as the coding format;

[0041] The characteristic processing scheme chromosome chain represents the processing scheme used under the selection result of the process-machine chromosome, and adopts the machine code + scheme code in the form of integer as the encoding format;

[0042] The tool chromosome chain represents the tool selected under the processing scheme, and is represented by a machine code + a tool code in the form of an integer.

[0043] In the second aspect, a process logistics scheduling system is provided, which uses a multi-objective group intelligence algorithm based on trend neighborhood search for scheduling optimization, including:

[0044] Population initialization module: constructs the chromosomes of individuals in the population and performs initialization operations on all individuals in the population;

[0045] The chromosomes include process-machine chromosomes and process-processing feature chromosomes; wherein the process-machine chromosomes represent the selection of machines and fixtures for each process; the process-processing feature chromosomes represent the order of processing features on the machine tool selected for each process, as well as the processing scheme and tool for the processing features;

[0046] Mutation operation module: Generate new individuals by randomly changing certain genes on chromosomes to form offspring populations, increase the diversity of the entire population, and expand the search space;

[0047] Selection operation module: simulates the characteristics of survival of the fittest in natural selection, obtains the fitness value of each individual in the population by calculating the objective function value of the individual, and allows individuals with good fitness to survive in the iteration while keeping the size of the population constant.

[0048] Preferably, the system adopts minimization of total processing cost and total processing time as optimization objectives, and adopts the following search trend rules:

[0049] Trend rule 1: When selecting the processing machine for each process, the probability of selecting a low-priced machine is greater than the probability of selecting a high-priced machine;

[0050] Trend rule 2: When selecting the processing scheme for each feature, the probability of selecting the processing scheme with less time is greater than the probability of selecting the processing scheme with more time;

[0051] Trend rule 3: When selecting the machining tool for the current machining plan of each feature, the probability of a high-priced tool being selected is greater than the probability of a low-priced tool being selected.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention ensures the consistency of chromosome design with the process through multi-layer coding design, and has the advantages of simple coding and easy population initialization;

[0054] 2. The present invention adopts search tendency rules to ensure the effectiveness of population evolution, and has the advantages of simple evolution rules, population offspring conforming to the process, and easy operation and implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0056] Figure 1 It is a schematic diagram of the process-machine chromosome coding;

[0057] Figure 2 It is a schematic diagram of the process-characteristic chromosome coding;

[0058] Figure 3 It is a flowchart of a multi-target swarm intelligence algorithm based on directional neighborhood search;

[0059] Figure 4 This is a diagram illustrating the cumulative probability of individuals in roulette. DETAILED DESCRIPTION

[0060] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0061] The embodiment of the present invention provides a process logistics scheduling method, referring to Figure 1 and Figure 2 As shown, the details are as follows:

[0062] Process-machine chromosome: indicates the selection of machines and fixtures for each process.

[0063] Process-processing feature chromosome: represents the order of processing features on the machine tool selected in each process, as well as the processing plan and tools for the processing features; the process-machine chromosome and the process-processing feature chromosome are aligned in the process arrangement sequence.

[0064] Among them, the process-machine chromosome is arranged in the order of the process, including the machine chromosome chain and the fixture chromosome chain; the machine chromosome chain represents the processing machine selected for a certain process, and preferably uses the machine number in the form of an integer as the encoding format; the fixture chromosome chain represents the tooling fixture used in the process, and the preferred encoding method is the form of an integer machine number + fixture number.

[0065] The process-processing feature chromosome represents the sequential processing order of the workpiece features, including the processing feature chromosome chain, the feature processing scheme chromosome chain and the tool chromosome chain; the processing feature chromosome chain represents the chromosome code corresponding to the workpiece feature, preferably using the workpiece number + feature number in the form of an integer as the encoding format; the feature processing scheme chromosome chain represents the processing scheme that can be used under the selection result of the process-machine chromosome, using the machine code + scheme code in the form of an integer as the encoding format; the tool chromosome chain represents the tool selected under the processing scheme, preferably using the machine code + tool code in the form of an integer.

[0066] The present invention provides a process logistics scheduling method, which uses a multi-objective group intelligent algorithm based on trend neighborhood search to perform scheduling optimization, specifically including a population initialization step, a mutation operation step and a selection operation step:

[0067] Population initialization step: constructing chromosomes of individuals in the population and performing initialization operations on all individuals in the population; the chromosomes include the above-mentioned process-machine chromosomes and process-processing feature chromosomes;

[0068] Mutation operation steps: Generate new individuals by randomly changing certain genes on chromosomes to form offspring populations, increase the diversity of the entire population, and expand the search space, including swap mutation or reverse mutation;

[0069] Selection operation steps: simulate the characteristics of survival of the fittest in natural selection, obtain the fitness value of each individual in the population by calculating the objective function value of the individual, and make the individuals with good fitness survive in the iteration while keeping the size of the population constant.

[0070] Minimizing the total processing cost and total processing time is used as the optimization goal, and the following search trend rules are adopted:

[0071] Trend rule 1: When selecting the processing machine for each process, the cheaper machine has a greater probability of being selected.

[0072] Trend rule 2: When selecting the processing plan for each feature, the processing plan that takes less time has a greater probability of being selected.

[0073] Trend rule 3: When selecting the machining tool for the current machining plan of each feature, the cheaper tool has a greater probability of being selected.

[0074] The population initialization steps are as follows:

[0075] Step S1: Construct a certain body chromosome in the initial solution from the process-machine chromosome. Starting from the first process, select the processing machine that can be used in the current process, and follow the trend rule 1 when selecting; when the machine of a certain process is selected, the fixture of the machine is determined accordingly.

[0076] Step S2: Select and sort the features processed on this machine in the current process. Randomly select a feature from the processing features that meet the order requirements. If it is the first time that the feature appears, it represents feature step 1. At this time, you need to select a processing plan for processing this feature. The processing plan must select a processing plan that can be used by the current machine. If it is the second time that the feature appears, there is no need to select a processing plan, and directly follow the processing plan of this feature step 1. Follow the trend rule 2 when selecting the processing plan.

[0077] Step S3: Select the machining tool for the current feature machining plan, and follow trend rule 3 when selecting.

[0078] Step S4: After arranging all the features that can be processed in a process, create the next process and repeat steps S1 to S3 until all the processing features are arranged.

[0079] Step S5: When the process-machine chromosome and the process-processing feature chromosome are all constructed, the chromosome construction of an individual in the population is completed, and steps S1 to S4 are repeated until all individuals in the population complete the initialization operation.

[0080] The mutation operation step generates new individuals by randomly changing certain genes on the chromosomes to form the offspring population, increase the diversity of the entire population, and thus expand the search space, including exchange mutation, reverse mutation, etc. In particular, the present invention adopts a layered chromosome mutation method, starting from the neighborhood of the current better result, to expand the search range, and follow the trend search rules during the search to ensure the order requirements of the processed features and the feasibility of the mutation results.

[0081] The selection operation step simulates the characteristics of survival of the fittest in natural selection. By calculating the objective function value of each individual in the solution population, the fitness value of the individual is obtained, and individuals with good fitness are retained in the iteration, while keeping the size of the population constant, thereby accelerating the global convergence speed.

[0082] Reference Figure 3 and Figure 4 As shown, the selection operation steps adopt the roulette selection method based on multi-objective weighted sum, and the specific operations are as follows:

[0083] (1) After standardizing all objective function values ​​to the interval [0, 1], assign a weight with practical significance to each objective function, and then aggregate all objective functions into a single objective function. That is, the optimization problem of multiple objectives is converted into a single objective optimization problem in the following form through the weighted sum method:

[0084]

[0085] Among them, the weight coefficient λ i ≥0, indicating the relative importance of each objective function, and Ω is the set of feasible solutions;

[0086] (2) Calculate the fitness of each individual in the group based on the aggregated single objective function: j (x)(j=1,2,3,…,N), N is the size of the population. When the optimization goal of the problem is to minimize the objective function F j (x)(j=1,2,3,…,N), the smaller the objective function is, the higher the individual fitness value Fit is. i The larger it is, the easier it is for an individual to be selected.

[0087] (3) Calculate the probability P of each individual in the population being inherited to the next generation j ;

[0088]

[0089] (4) Calculate the cumulative probability q of each individual in the population k :

[0090]

[0091] (5) Generate a pseudo-random number r1 uniformly distributed in the interval [0, 1];

[0092] (6) If q[k - 1] < r1 ≤ q[k], then select individual k;

[0093] (7) Repeat steps (5) and (6) a total of N times to form the next-generation parental population C of size N i+1 .

[0094] The embodiment of the present invention provides a process logistics scheduling method and system. Through multi-layer coding design, the consistency between chromosome design and the technological process is ensured. At the same time, the search tendency rule is adopted to ensure the effectiveness of population evolution in group evolution, and it has the advantages of simple coding, simple rules, and easy operation.

[0095] Those skilled in the art know that in addition to implementing the system and its various devices, modules, and units provided by the present invention in the form of pure computer-readable program code, the method steps can be logically programmed to enable the system and its various devices, modules, and units provided by the present invention to be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers, etc., to achieve the same functions. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a kind of hardware component, and the devices, modules, and units included therein for implementing various functions can also be regarded as the structure within the hardware component; the devices, modules, and units for implementing various functions can also be regarded as either software modules for implementing the method or the structure within the hardware component.

[0096] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific implementation manners. Those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. Without conflict, the embodiments of the present application and the features in the embodiments can be combined arbitrarily with each other.

Claims

1. A process logistics scheduling method, characterized in that: A multi-objective swarm intelligence algorithm based on directional neighborhood search is used for scheduling optimization, including: Population initialization step: construct the chromosomes of individuals in the population and perform initialization operations on all individuals in the population; The chromosomes include process-machine chromosomes and process-processing feature chromosomes; wherein the process-machine chromosomes represent the selection of machines and fixtures for each process; the process-processing feature chromosomes represent the order of processing features on the machine tool selected for each process, as well as the processing scheme and tool for the processing features; Mutation operation steps: Generate new individuals by randomly changing certain genes on chromosomes to form offspring populations, increase the diversity of the entire population, and expand the search space; Selection operation steps: simulate the survival of the fittest in natural selection, calculate the objective function value of each individual in the population, obtain the fitness value of the individual, and make the individuals with good fitness survive in the iteration, while keeping the size of the population constant; The method adopts minimization of total processing cost and total processing time as the optimization goal, and adopts the following search trend rules: Trend rule 1: When selecting the processing machine for each process, the probability of selecting a low-priced machine is greater than the probability of selecting a high-priced machine; Trend rule 2: When selecting the processing scheme for each feature, the probability of selecting the processing scheme with less time is greater than the probability of selecting the processing scheme with more time; Trend rule 3: When selecting the machining tool for the current machining plan of each feature, the probability of selecting a tool with a high price is greater than the probability of selecting a tool with a low price; In the population initialization step, the process-machine chromosome and the process-processing feature chromosome are aligned on the process arrangement sequence; the process-machine chromosome is arranged in sequence according to the order of the processes, including a machine chromosome chain and a fixture chromosome chain; the machine chromosome chain represents the processing machine selected for a certain process, and uses an integer machine number as the encoding format; the fixture chromosome chain represents the tooling fixture used in the process, and the encoding method is an integer machine number + fixture number format; The process-processing feature chromosome represents the sequential processing order of the workpiece features, including a processing feature chromosome chain, a feature processing scheme chromosome chain and a tool chromosome chain; the processing feature chromosome chain represents the chromosome code corresponding to the workpiece feature, using an integer workpiece number + feature number as the encoding format; the feature processing scheme chromosome chain represents the processing scheme used under the selection result of the process-machine chromosome, using an integer machine code + scheme code as the encoding format; the tool chromosome chain represents the tool selected under the processing scheme, using an integer machine code + tool code for representation.

2. The process logistics scheduling method according to claim 1, characterized in that: The population initialization includes the following steps: Step S1: construct a certain body chromosome in the initial solution from the process-machine chromosome; starting from the first process, select the processing machine that can be used in the current process, and follow the trend rule 1 when selecting; when a machine for a certain process is selected, the fixture of the machine is determined accordingly; Step S2: Select and sort the features processed on this machine in the current process; Step S3: Select the processing tool for the current feature processing plan, and follow the tendency rule 3 when selecting; Step S4: After arranging all the features that can be processed within one process, create the next process, and repeat Steps S1 to S3 until all the processing features are sorted; Step S5: When the process-machine chromosome and the process-processing feature chromosome are all constructed, the chromosome construction of an individual in the population is completed. Repeat Steps S1 to S4 until the initialization operations of all individuals in the population are completed.

3. The process logistics scheduling method according to claim 2, characterized in that: The specific content of Step S2 includes: randomly select a feature from the processing features that meet the order requirements. If it is a feature that appears for the first time, it represents Feature Step 1. At this time, it is necessary to select the processing plan for processing this feature; the processing plan must be selected from the processing plans that can be used by the current machine; If it is a feature that appears for the second time, there is no need to select a processing plan, and directly follow the processing plan of Feature Step 1; follow the tendency rule 2 when selecting the processing plan.

4. The process logistics scheduling method according to claim 1, characterized in that: The selection operation step adopts the roulette wheel selection method based on the multi-objective weighted sum. The specific operation is as follows: (1) After normalizing all the objective function values to the interval [0, 1], assign weights with practical significance to each objective function, and then aggregate all the objective functions into a single objective function, that is, convert the optimization problem of multiple objectives into a single objective optimization problem of the following form by the weighted sum method: Among them, the weight coefficient λ i ≥0, indicating the relative importance of each objective function, and Ω is the set of feasible solutions; (2) Calculate the fitness of each individual in the group based on the aggregated single objective function: j (x), j = 1, 2, 3, ..., N, N is the size of the population; when the optimization goal of the problem is to minimize the objective function F j (x), the smaller the objective function is, the higher the individual fitness value is. i The larger it is, the easier it is for an individual to be selected; (3) Calculate the probability P of each individual in the population being inherited to the next generation j ; (4) Calculate the cumulative probability q of each individual in the population k : (5) Generate a uniformly distributed pseudo-random number r1 in the interval [0, 1]; (6) If q[k - 1] < r1 ≤ q[k], then select individual k; (7) Repeat steps (5) and (6) N times to form the next generation parent population C of size N. i+1 .

5. A process logistics scheduling system, characterized in that: Adopt a multi-objective swarm intelligence algorithm based on tendency neighborhood search for scheduling optimization, including: Population initialization module: Construct the chromosomes of individuals in the population and perform initialization operations on all individuals in the population; The chromosome includes a process-machine chromosome and a process-processing feature chromosome; among them, the process-machine chromosome: represents the machine and fixture selected for each process; the process-processing feature chromosome: represents the order of the processing features on the machine selected for each process, as well as the processing plan and tool of the processing features; Mutation operation module: Generate new individuals by randomly changing some genes on the chromosome to form the offspring population, increase the diversity of the entire population, and expand the search space; Selection operation module: Simulate the survival of the fittest in natural selection. By calculating the objective function value of each individual in the population, obtain the fitness value of this individual, and keep the individuals with good fitness in the iteration, while keeping the size of the population constant; The system adopts minimizing the total processing cost and the total processing time as the optimization objectives, and at the same time adopts the following search tendency rules: Tendency rule 1: When selecting the processing machine for each process, the probability of selecting a machine with a lower price is greater than the probability of selecting a machine with a higher price; Tendency rule 2: When selecting the processing plan for each feature, the probability of selecting a processing plan with less time is greater than the probability of selecting a processing plan with more time; Trend rule 3: When selecting the machining tool for the current machining plan of each feature, the probability of selecting a tool with a high price is greater than the probability of selecting a tool with a low price; In the population initialization module, the process-machine chromosome and the process-processing feature chromosome are aligned on the process arrangement sequence; the process-machine chromosome is arranged in sequence according to the order of the processes, including a machine chromosome chain and a fixture chromosome chain; the machine chromosome chain represents the processing machine selected for a certain process, and uses an integer machine number as the encoding format; the fixture chromosome chain represents the tooling fixture used in the process, and the encoding method is an integer machine number + fixture number format; The process-processing feature chromosome represents the sequential processing order of the workpiece features, including a processing feature chromosome chain, a feature processing scheme chromosome chain and a tool chromosome chain; the processing feature chromosome chain represents the chromosome code corresponding to the workpiece feature, using an integer workpiece number + feature number as the encoding format; the feature processing scheme chromosome chain represents the processing scheme used under the selection result of the process-machine chromosome, using an integer machine code + scheme code as the encoding format; the tool chromosome chain represents the tool selected under the processing scheme, using an integer machine code + tool code for representation.

Images