CN-122022353-A - Smart system-based intelligent optimization automatic scheduling method for multiple varieties of railway vehicle products

CN122022353ACN 122022353 ACN122022353 ACN 122022353ACN-122022353-A

Abstract

The invention relates to a SMART optimization automatic scheduling method for multiple varieties of railway vehicle products based on a SMART system, relates to the technical field of railway vehicle manufacturing, and solves the technical problems of low scheduling efficiency, unreasonable resource allocation, poor accuracy, weak strain capacity and insufficient visualization and traceability of the multiple varieties of railway vehicle production in the prior art. The automatic scheduling method comprises the steps of modeling scheduling basic data based on SMART, constructing a multi-constraint intelligent scheduling model, improving scheduling optimization solution of a genetic algorithm, dynamically adjusting and monitoring a mechanism in real time, designing a visual man-machine interaction interface, and issuing and executing a work order to close a loop. The automatic scheduling method can improve scheduling efficiency, optimize resource allocation, ensure scheduling accuracy, enhance dynamic strain capacity and realize visual management and control.

Inventors

Sui Yonggun
JIANG WENTAO
LIU XIANGFU
CAO YAPING

Assignees

中车长春轨道客车股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260205

Claims (7)

1. The SMART optimization automatic scheduling method for the rail vehicle products of multiple varieties based on the SMART system is characterized by comprising the following steps of: step S1, modeling based on scheduling basic data of a SMART system; Building a standardized data model covering sites, production lines, personnel, components and orders, and providing accurate data support for scheduling: s2, constructing a multi-constraint intelligent scheduling model; The core constraint conditions of the production of the railway vehicles are combined, a multi-constraint intelligent scheduling model adapting to the production of multiple varieties is constructed, and feasibility and optimality of a scheduling scheme are ensured; step S3, improving the scheduling optimization solution of the genetic algorithm; Solving the multi-constraint intelligent scheduling model by adopting an improved genetic algorithm, and generating an optimal scheduling scheme through iterative optimization; step S4, a dynamic adjustment and real-time monitoring mechanism; Based on the real-time data interaction capability of the SMART system, a dynamic adjustment mechanism of a scheduling scheme is established, and various changes in the production process are responded; S5, designing a visual man-machine interaction interface; Designing a man-machine interaction interface, and supporting visual control and manual intervention of the whole scheduling process; Step S6, the work order is issued and closed loop is executed; And converting the optimized scheduling scheme into an executable production work order, and realizing closed-loop management of scheduling and execution.
2. The SMART system-based rail vehicle product multi-variety intelligent optimization automatic scheduling method of claim 1, wherein the scheduling basis data modeling in step S1 specifically comprises: (1) Modeling site resources; Defining the upper limit of productivity and time constraint of each site; (2) Modeling a production line configuration; recording the number, priority and effective state of production lines corresponding to each site, and establishing an association mapping between the sites and the production lines; (3) Modeling a component process; Determining the codes, names, models and repair requirements of all the components and corresponding item codes, and combing the standard operation flow and process dependency relationship of all the components; (4) Modeling order requirements; order information is input, the order information is disassembled into executable process tasks, and time thresholds and quality requirements of each process are set.
3. The SMART system-based rail vehicle product multi-variety intelligent optimization automatic scheduling method of claim 1, wherein the multi-constraint intelligent scheduling model construction in step S2 specifically comprises: (1) Core constraints: Site constraint, namely strictly matching the site type with the product model, and adapting the repair grade with the site repair grade; line constraint, namely, based on the priority of the production line and the upper limit of productivity, avoiding overload operation of a single production line; time constraint, namely, following working time and rest configuration to ensure that the process beats conform to production practice; Process constraint, namely ensuring that the sequence of the working procedures is consistent with the repair level requirement of the component, and avoiding cross-flow scheduling; (2) Model objective function, namely multi-constraint intelligent scheduling model, namely constructing a multi-objective optimized model objective function by taking the maximized equipment utilization rate, the shortest order delivery cycle and the minimum process waiting time as core targets The method is characterized by comprising the following steps: Wherein, the Respectively, weight coefficients.
4. The SMART system-based rail vehicle product multi-variety intelligent optimization automatic scheduling method of claim 3, wherein step S3 specifically comprises: (1) The coding design adopts a four-dimensional coding mode of order-part-production line-time, and chromosome segments comprise order numbers, part codes, distribution production line numbers and procedure starting/ending time, so that the codes are ensured to be in one-to-one correspondence with production reality; (2) The method comprises the following steps of initializing a population, namely randomly generating 50-100 groups of initial scheduling schemes based on site and production line configuration, and ensuring that the initial population covers different resource allocation combinations; (3) The fitness calculation, namely taking the model objective function in the step S2 as a fitness function, calculating the fitness value of each group of scheduling schemes, and screening effective schemes meeting constraint conditions; (4) Genetic manipulation: Selecting individuals with the first 30% of fitness values by adopting a roulette method to enter the next generation, and reserving an optimal scheme; crossing, namely adopting a two-point crossing method to exchange production line distribution-time node segments of different chromosomes, and ensuring that the process and time constraint are still met after crossing; The variation is that the production line distribution or the working procedure time of partial individuals is randomly adjusted according to the probability of 0.05-0.1, so as to avoid the algorithm from sinking into local optimum; (5) And stopping iteration when the fluctuation of the optimal fitness value of 10 continuous iterations is less than 1%, and outputting an optimal scheduling scheme.
5. The SMART system-based rail vehicle product multi-variety intelligent optimization automatic scheduling method of claim 1, wherein step S4 specifically comprises: (1) And acquiring real-time data, namely acquiring production field data through an internet of things sensor and a SMART system, wherein the real-time data comprises the following steps of: The resource state is the equipment running state, namely normal/fault/maintenance and personnel on duty; production progress, namely actual completion time of each procedure and assembly progress of the components; External changes, order change, material supply delay, quality detection results; (2) Setting a deviation threshold, automatically triggering the early warning by the system when the acquired data exceeds the threshold, and analyzing the deviation reason; (3) Automatic adjustment, namely, for different deviation types, calling an improved genetic algorithm to quickly re-optimize: equipment failure, namely distributing unfinished tasks of a failure production line to an idle production line in the same site; material delay, namely adjusting working procedure starting time of related components, and preferentially executing tasks with sufficient materials; Updating the order data model, regenerating a scheduling scheme, and ensuring that the resource allocation of the new order and the original order is not in conflict; (4) And (3) adjusting synchronization, namely synchronizing the adjusted scheduling scheme to the interface of the production field terminal and the manager in real time, and automatically updating the work order information.
6. The SMART system-based rail vehicle product multi-variety intelligent optimization automatic scheduling method according to claim 1, wherein the man-machine interaction interface in step S5 is provided with: (1) The parameter configuration module supports inputting/modifying basic data of site planning, production line configuration, rest time and component process flow, and self-defining a Cheng Quanchong coefficient and a deviation threshold; (2) The scheduling visualization module displays the scheduling result, presents the process arrangement, the starting/ending time and the beat allocation of each component in different production lines, and supports screening and viewing according to the production lines, orders and component types; (3) And the adjusting operation module is used for providing two adjusting modes: Manual fine tuning, namely supporting a time node of a dragging procedure, modifying distribution of a production line, and automatically following and adjusting a subsequent related procedure; batch adjustment, namely supporting to modify the time of working and working on the day and the rest period, and automatically adapting the time plan of all relevant procedures by a system; (4) And the data tracing module records the generation log of the scheduling scheme, adjusts the record and the production execution data, supports the inquiry of the whole flow information according to the order number and the component code, and generates a scheduling effect analysis report.
7. The SMART system-based rail vehicle product multi-variety intelligent optimization automatic scheduling method of claim 1, wherein step S6 specifically comprises: (1) Generating a work order, namely automatically generating a production work order according to a scheduling scheme; (2) Pushing the work order to an operation terminal of a corresponding production line through a SMART system, and enabling an operator to carry out operation according to the work order; (3) And after the worker completes the working procedure, recording the actual completion time and the quality state at the terminal, and updating the production progress in real time by the system to form an execution closed loop.

Description

Smart system-based intelligent optimization automatic scheduling method for multiple varieties of railway vehicle products Technical Field The invention relates to the technical field of railway vehicle manufacturing, in particular to a railway vehicle product multi-variety intelligent optimization automatic scheduling method based on a SMART system. Background In the production process of rail vehicles, production tasks of various products are faced. The rail vehicles with different models and different configurations have differences in the aspects of production procedures, production time, resource requirements and the like. The traditional scheduling method depends on manual experience, has low efficiency and is easy to make mistakes, and the production requirements of multiple varieties and small batches are difficult to meet. The competition in the rail vehicle market is increasing and the demands on product lead time and quality by customers are increasing. The production enterprises are required to arrange production plans more efficiently, resources are reasonably allocated, production efficiency is improved, and production cost is reduced. The existing system has limitation on the scheduling function, lacks an intelligent optimization algorithm and effective support for production of multiple varieties of products, cannot fully utilize production data to carry out deep analysis and decision, and is difficult to realize efficient scheduling of the production of the multiple varieties of railway vehicle products. In general, the prior art rail vehicle product scheduling methods suffer from the following drawbacks: (1) The scheduling efficiency is low, the manual scheduling needs 2-3 days to complete the planning and the making of single-batch multi-variety orders, and the work procedure connection confusion is easy to occur for multi-work-procedure parts such as a bogie, so that the production waiting time is prolonged; (2) Unbalanced resource allocation, namely, due to insufficient production line capacity, site constraint and rest time allocation, the condition that part of production lines run in overload and part of production lines are idle often occurs, and according to statistics, the utilization rate of special equipment in a traditional scheduling mode is only 60% -70%; (3) The scheduling accuracy is poor, the matching between the detail of the process flow of the easily neglected part and the repair grade requirement is judged by manual experience, the process is mismatched, the production is reworked, and the product delivery delay rate is up to more than 15 percent; (4) The strain capacity is weak, when order addition, material supply delay or equipment failure are faced, the whole flow plan is required to be manually adjusted again, the time is long, the chain reaction is easy to be caused, and the dynamic requirements of multi-variety production cannot be met; (5) The visualization and traceability are insufficient, the scheduling results are presented in a form of a table, an intuitive flow visualization interface is lacked, the production progress is difficult to monitor in real time, and the scheduling deviation reasons cannot be traced quickly. Disclosure of Invention The invention provides a SMART optimization automatic scheduling method for multiple varieties of rail vehicle products based on a SMART system, which aims to solve the technical problems of low scheduling efficiency, unreasonable resource allocation, poor accuracy, weak strain capacity and insufficient visualization and traceability of the rail vehicle multi-variety production in the prior art. In order to solve the technical problems, the technical scheme of the invention is as follows: A SMART rail vehicle product multi-variety intelligent optimization automatic scheduling method based on a SMART system comprises the following steps: step S1, modeling based on scheduling basic data of a SMART system; Building a standardized data model covering sites, production lines, personnel, components and orders, and providing accurate data support for scheduling: s2, constructing a multi-constraint intelligent scheduling model; The core constraint conditions of the production of the railway vehicles are combined, a multi-constraint intelligent scheduling model adapting to the production of multiple varieties is constructed, and feasibility and optimality of a scheduling scheme are ensured; step S3, improving the scheduling optimization solution of the genetic algorithm; Solving the multi-constraint intelligent scheduling model by adopting an improved genetic algorithm, and generating an optimal scheduling scheme through iterative optimization; step S4, a dynamic adjustment and real-time monitoring mechanism; Based on the real-time data interaction capability of the SMART system, a dynamic adjustment mechanism of a scheduling scheme is established, and various changes in the production process are responded; S5, designing a vi