An Enhanced NSGA-II Driven by Deep Reinforcement Learning to Mixed Flow Assembly Workshop Scheduling System with Constraints of Continuous Processing and Mold Changing
Abstract
1. Introduction
2. Problem Description and Modeling
2.1. Problem Description
- The quantity of workpieces to be produced in each period is known in advance.
- Each assembly task at any workstation involves installing only one component.
- A workstation may handle one or multiple types of components.
- If a workstation consecutively assembles components exceeding a predefined difficulty threshold, overload time is needed, and production must wait until completion.
- If the component type changes at a workstation, a mold-changing time is incurred before the next assembly task can begin.
- Owing to uncertainties in order placement and delivery timelines, the customer waiting time in the model considers only scheduling-related delays (not external factors).
- No stochastic disruptions are considered; the completion time of each order depends solely on the production schedule.
2.2. Notations
2.3. Model formulation
3. Deep-Reinforcement Learning-Driven Enhanced NSGA-II
Algorithm 1 The pseudocode of improved NSGA-II process driven by deep reinforcement learning |
Input: Demand data, processing technology data for various workpieces, and processing capability data for workstations. Output: Production scheduling scheme. 1 Initialize relevant parameters. 2 Load the deep reinforcement learning driven operator. 3 Initialize population. 4 Calculate the objective function of the initial population. 5 While the iteration limit is not reached, do: 6 Use crossover operations to search the current population. 7 Use mutation operations to search the current population. 8 Perform local search on the population using RL-VNS and update the gradient. 9 Calculate the crowding degree and perform fast non-dominated sorting. 10 Select the next generation individuals based on the elite retention strategy. 11 Output results and obtain statistical indicators. |
3.1. Encoding Scheme
3.2. Variable Neighborhood Search Operators
- Operator 1: (1) randomly select a production cycle with surplus production; (2) identify workpieces with production deficits in subsequent cycles; (3) swap their production positions.
- Operator 2: (1) Identify workpieces with surplus production; (2) locate cycles where the same workpiece shows production deficits; (3) exchange positions with other workpieces having production surplus in those cycles.
3.3. Deep-Reinforcement Learning-Driven Operator
Algorithm 2 The pseudocode of deep reinforcement learning-driven local search operator |
Input: Population to be searched, deep reinforcement learning agent Output: Population after search 1 While the search task for all search objectives is not completed, do: 2 if the number of iterations < the upper limit of iterations, do: 3 Update 4 Calculate the return values of various actions under by the deep reinforcement learning operator. 5 Select the corresponding local search operator for local search based on the return values. 6 Check if the new population has improvements? Determine the reward value . 7 Determine . 8 Use to update the gradient of deep reinforcement learning. |
4. Simulation Experiments
4.1. Test preparation and parameter settings
4.2. Algorithm Comparison Experiments
5. Conclusion
- Research on scheduling strategies considering inventory constraints in line-side warehouses in mixed-flow assembly scenarios is conducted.
- A bilevel programming model considering AGV routing in mixed-flow assembly scenarios is established, and scheduling strategies are studied.
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
DRL | deep reinforcement learning |
NP-hard | non-deterministic polynomial |
NSGA-II | non-dominated sorting genetic algorithm II |
VNS | variable neighborhood search |
HV | Hypervolume |
GD | Generation Distance |
IGD | Inverse Generation Distance |
NNS | number of non-dominated solutions |
References
- Zhang, Q.B.; Si, H.Q.; Qin, J.Y.; Duan, J.G.; Zhou, Y.; Shi, H.X.; Nie, L. Double Deep Q-Network-Based Solution to a Dynamic, Energy-Efficient Hybrid Flow Shop Scheduling System with the Transport Process. Systems 2025, 13, 170. [Google Scholar] [CrossRef]
- Li, R.; Wang, L.; Gong, W.Y.; Chen, J.F.; Pan, Z.x.; Wu, Y.T.; Yu, Y. Evolutionary computation and reinforcement learning integrated algorithm for distributed heterogeneous flowshop scheduling. Eng. Appl. Artif. Intell. 2024, 135, 108775. [Google Scholar] [CrossRef]
- Zhou, F.S.; Hu, R.; Qian, B.; Shang, Q.X.; Yang, Y.Y.; Yang, J.B. Spatial decomposition-based iterative greedy algorithm for the multi-resource constrained re-entrant hybrid flow shop scheduling problem in semiconductor wafer fabrication. Comput. Ind. Eng. 2025, 208, 111330. [Google Scholar] [CrossRef]
- Wang, Y.; Ma, H.S.; Yang, J.H.; Wang, K.S. Industry 4.0: A way from mass customization to mass personalization production. Adv. Manuf. 2017, 5, 311–320. [Google Scholar] [CrossRef]
- Wu, Y.T.; Wang, L.; Li, R.; Xu, Y.X.; Zheng, J. A learning-based dual-population optimization algorithm for hybrid seru system scheduling with assembly. Swarm Evol. Comput. 2025, 94, 101901. [Google Scholar] [CrossRef]
- Shen, Z.Y.; Tang, Q.; Huang, T.; Xiong, T.Y. Solution of multistage flow shop scheduling model for leveling production. Comput. Integr. Manuf. Syst. 2019, 25, 2743–2752. [Google Scholar]
- Zhang, T.R.; Wang, Y.K.; Xie, W.; Xu, J.N.; Wang, R.L. Improved backtracking search algorithm oriented the problem of mixed flow production line sequencing. Modul. Mach. Tool Autom. Manuf. Tech. 2021, 8, 44–47+51. [Google Scholar]
- Su, Z.S.; Deng, C.; Chiong, R.; Jiang, S.L.; Zhang, K. Hybrid flow shop scheduling with continuous processing and resource threshold constraints: A case of steel plant. Expert Syst. Appl. 2025, 284, 127247. [Google Scholar] [CrossRef]
- Wang, H.; Sheng, B.Y.; Lu, X.C.; Fu, G.C.; Luo, P.P. Task package division method for the integrated scheduling framework of mixed model car-sequencing problem. Comput. Ind. Eng. 2022, 169, 108144. [Google Scholar] [CrossRef]
- Guan, L.; Wang, Y.L.; Tan, X.J.; Liu, C.L.; Gui, W.H. Machine scheduling optimization via multi-strategy information-aware genetic algorithm in steelmaking continuous casting industrial process. Control Eng. Pract. 2025, 164, 106404. [Google Scholar] [CrossRef]
- Laili, Y.J.; Peng, C.; Chen, Z.L.; Ye, F.; Zhang, L. Concurrent local search for process planning and scheduling in the industrial Internet-of-Things environment. J. Ind. Inf. Integr. 2022, 28, 100364. [Google Scholar] [CrossRef]
- Zhang, Y.H.; Hu, X.F.; Cao, X.F.; Wu, C.X. An efficient hybrid integer and categorical particle swarm optimization algorithm for the multi-mode multi-project inverse scheduling problem in turbine assembly workshop. Comput. Ind. Eng. 2022, 169, 108148. [Google Scholar] [CrossRef]
- Liu, G.Q.; Xi, H.; Wang, Y.; Wang, Z.Y. Optimization of mixed-model automobile assembly line sequencing problem based on improved genetic algorithm. Comput. Appl. Softw. 2021, 38, 78–83+137. [Google Scholar]
- Geng, K.F.; Ye, C.M. Joint scheduling of machine and AGVs in green hybrid flow shop with missing operations. Control Decis. 2022, 37, 2723–2732. [Google Scholar]
- Wu, X.L.; Cao, Z. Re-entrant hybrid flow shop scheduling problem with continuous batch processing machines. Comput. Integr. Manuf. Syst. 2022, 28, 3365–3378. [Google Scholar]
- Wallrath, R.; Zondervan, E.; Franke, M.B. Modified benders-DES algorithm for real-world flow shop and job shop scheduling problems. Comput. Chem. Eng. 2025, 202, 109221. [Google Scholar] [CrossRef]
- Gao, Q.L.; Liu, J.H.; Li, H.T.; Zhuang, C.B.; Liu, Z.W. Digital twin-driven dynamic scheduling for the assembly workshop of complex products with workers allocation. Rob. Comput. Integr. Manuf. 2024, 89, 102786. [Google Scholar] [CrossRef]
- Ding, X.J.; Gong, Z.C.; Yang, Y.P.; Shi, X.; Peng, Z.K.; Cao, X.B.; Hu, S.T. Modeling and scheduling a triply-constrained flow shop in biomanufacturing systems. J. Manuf. Syst. 2024, 76, 333–350. [Google Scholar] [CrossRef]
- Wang, L.; Pan, Z.X.; Wang, J.J. A review of reinforcement learning based intelligent optimization for manufacturing scheduling. Complex Syst. Model. Simul. 2021, 1, 257–270. [Google Scholar] [CrossRef]
- Karimi-Mamaghan, M.; Mohammadi, M.; Pasdeloup, B.; Meyer, P. Learning to select operators in meta-heuristics: An integration of Q-learning into the iterated greedy algorithm for the permutation flowshop scheduling problem. Eur. J. Oper. Res. 2023, 304, 1296–1330. [Google Scholar] [CrossRef]
- Leng, J.W.; Guo, J.W.; Zhang, H.; Xu, K.L.; Qiao, Y.; Zheng, P.; Shen, W.M. Dual deep reinforcement learning agents-based integrated order acceptance and scheduling of mass individualized prototyping. J. Clean. Prod. 2023, 427, 139249. [Google Scholar] [CrossRef]
- Yüksel, D.; Kandiller, L.; Taşgetiren, M.F. Q-learning guided algorithms for bi-criteria minimization of total flow time and makespan in no-wait permutation flowshops. Swarm Evol. Comput. 2024, 89, 101617. [Google Scholar] [CrossRef]
- Li, Z.; Wang, Y.T.; Han, Y.Y.; Gao, K.Z.; Li, J.Q. Q-Learning-Driven Accelerated Iterated Greedy Algorithm for Multi-Scenario Group Scheduling in Distributed Blocking Flowshops. Knowl.-Based Syst. 2025, 317, 113424. [Google Scholar] [CrossRef]
- Li, W.; Liang, P.; Sun, B.; Sun, Y.F.; Huang, Y. Reinforcement learning-based particle swarm optimization with neighborhood differential mutation strategy. Swarm Evol. Comput. 2023, 78, 101274. [Google Scholar] [CrossRef]
- Ding, L.S.; Luo, D.; Mudassar, R.; Yue, L.; Meng, L.L. A novel deep self-learning method for flexible job-shop scheduling problems with multiplicity: Deep reinforcement learning assisted the fluid master-apprentice evolutionary algorithm. Swarm Evol. Comput. 2025, 94, 101907. [Google Scholar] [CrossRef]
- Zheng, X.; Chen, Z. An improved deep Q-learning algorithm for a trade-off between energy consumption and productivity in batch scheduling. Comput. Ind. Eng. 2024, 188, 109925. [Google Scholar] [CrossRef]
- Yuan, M.H.; Ye, Y.; Huang, H.Y.; Zhang, Z.; Pei, F.Q.; Gu, W.B. Multi-objective energy-efficient scheduling of distributed heterogeneous hybrid flow shops via multi-agent double deep Q-Network. Swarm Evol. Comput. 2025, 98, 102076. [Google Scholar] [CrossRef]
- Deb, K.; Pratap, A.; Agarwal, S.; Meyarivan, T. A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans. Evol. Comput. 2002, 6, 182–197. [Google Scholar] [CrossRef]
Literature | Model Building | Solution Approach | |||||||
---|---|---|---|---|---|---|---|---|---|
Personalized Order | Multiple Production Cycles | Change Mold Constraints | Continuous Processing Constraints | Multiple Objectives | Heuristic Algorithm | Experience Guidance | Reinforcement Learning | Deep Reinforcement Learning | |
Wu et al. [5] | ✔ | ✔ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ |
Shen et al. [6] | ✔ | ✔ | ✔ | ✖ | ✖ | ✔ | ✖ | ✖ | ✖ |
Zhang et al. [7] | ✔ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Su et al. [8] | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ | ✖ | ✖ |
Guan et al. [10] | ✔ | ✖ | ✖ | ✔ | ✔ | ✔ | ✔ | ✖ | ✖ |
Laili et al. [11] | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Zhang et al. [12] | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Liu et al. [13] | ✔ | ✔ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Geng et al. [14] | ✔ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Wu et al. [15] | ✔ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Wallrath et al. [16] | ✖ | ✖ | ✖ | ✖ | ✖ | ✔ | ✖ | ✖ | ✖ |
Gao et al. [17] | ✔ | ✖ | ✖ | ✖ | ✔ | ✔ | ✖ | ✖ | ✖ |
Ding et al. [18] | ✖ | ✖ | ✖ | ✖ | ✖ | ✔ | ✖ | ✖ | ✖ |
Yüksel et al. [22] | ✖ | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✔ | ✖ |
Li et al. [23] | ✖ | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✔ | ✖ |
Li et al. [24] | ✖ | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✔ | ✖ |
Ding et al. [25] | ✔ | ✔ | ✖ | ✖ | ✖ | ✔ | ✔ | ✔ | ✔ |
Zheng et al. [26] | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✔ | ✔ | ✔ |
Yuan et al. [27] | ✖ | ✖ | ✖ | ✖ | ✔ | ✔ | ✔ | ✔ | ✔ |
this paper | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
Symbols | Explanations | Symbols | Explanations |
---|---|---|---|
Decision variable, type and configuration number of the workpiece in median position of the production cycle | 0–1 variable, indicating whether the component can be assembled at station | ||
Upper limit of workpiece type number | 0–1 variable, where 0 indicates that the workpiece at position in period has no overload at station , and vice versa. | ||
Upper limit of the production cycle | 0–1 variable, indicating whether the workpiece type contains the accessory | ||
Accessory number limit | The load limit of the workstation assembly accessories | ||
Workstation number upper limit | The waiting time for mold change at station for workpiece at position produced in cycle | ||
The maximum number of workpieces produced in each production batch | The time when the workpiece at position is processed at station during cycle | ||
Accessory number, | The processing time of the workpiece at position produced in cycle at workstation | ||
Workpiece Type Number, | The output of type configured within cycle | ||
Workstation number, | The time for the normal assembly of part | ||
Workstation set, | The time when the workpiece at position is produced in cycle and leaves workstation | ||
A sufficiently large positive integer | The configuration set of workpieces produced by the mixed flow assembly line, | ||
Extra assembly time caused by overload | Total demand quantity of configuration of type , | ||
The quantity of configuration for workpiece type | The time for the workpiece produced at position during cycle to enter station . | ||
The total demand quantity of type configuration at time | Load statistical parameters of assembly parts at workstation |
Example | ||||
---|---|---|---|---|
1 | 18 | 60 | 37.59 | 2 |
2 | 30 | 60 | 38.55 | 2 |
3 | 30 | 66 | 41.96 | 5 |
4 | 30 | 66 | 41.96 | 5 |
5 | 15 | 33 | 20.94 | 10 |
6 | 15 | 33 | 21.11 | 10 |
7 | 5 | 11 | 7.0364 | 10 |
Example | RLVNS-NSGA-II | VNS-NSGA-II | SVNS-NSGA-II | ||||||
---|---|---|---|---|---|---|---|---|---|
1 | 1 | 20 | 12.95 | 3 | 20 | 13.25 | 1 | 20 | 14.425 |
2 | 3 | 20 | 15.675 | 0 | 20 | 11.05 | 1 | 20 | 12.375 |
3 | 4 | 20 | 14.475 | 8 | 20 | 16.425 | 8 | 20 | 15.75 |
4 | 8 | 20 | 16.575 | 6 | 20 | 15.275 | 7 | 20 | 15.375 |
5 | 8 | 20 | 17 | 8 | 20 | 15.675 | 7 | 20 | 15.925 |
6 | 6 | 20 | 16.9 | 6 | 20 | 16.6 | 6 | 20 | 16.6 |
7 | 10 | 20 | 17.1 | 6 | 20 | 17 | 6 | 20 | 17 |
Example | RLVNS-NSGA-II | VNS-NSGA-II | SVNS-NSGA-II | ||||||
---|---|---|---|---|---|---|---|---|---|
1 | 6.19 × 10−4 | 1.73 × 10−1 | 3.03 × 10−2 | 2.17 × 10−2 | 4.55 × 10−1 | 1.00 × 10−1 | 1.05 × 10−2 | 6.69 × 10−1 | 6.91 × 10−2 |
2 | 1.83 × 10−8 | 2.08 × 10−1 | 3.16 × 10−2 | 6.11 × 10−3 | 5.78 × 10−1 | 9.50 × 10−2 | 4.94 × 10−3 | 2.77 × 10−1 | 8.21 × 10−2 |
3 | 1.75 × 10−4 | 1.36 × 10−1 | 2.17 × 10−2 | 1.21 × 10−2 | 1.96 × 10−1 | 6.81 × 10−2 | 1.29 × 10−2 | 2.92 × 10−1 | 6.25 × 10−2 |
4 | 4.59 × 10−6 | 2.01 × 10−1 | 3.93 × 10−2 | 2.87 × 10−2 | 8.25 × 10−1 | 2.16 × 10−1 | 4.61 × 10−3 | 5.35 × 10−1 | 1.57 × 10−1 |
5 | 6.47 × 10−4 | 1.39 × 10−1 | 3.45 × 10−2 | 3.63 × 10−2 | 4.04 × 10−1 | 1.50 × 10−1 | 3.53 × 10−2 | 2.88 × 10−1 | 1.10 × 10−1 |
6 | 1.27 × 10−3 | 7.96 × 10−2 | 3.05 × 10−2 | 1.65 × 10−2 | 2.53 × 10−1 | 9.56 × 10−2 | 1.17 × 10−2 | 3.52 × 10−1 | 9.58 × 10−2 |
7 | 0.00 | 1.36 × 10−1 | 4.04 × 10−2 | 7.52 × 10−3 | 2.21 × 10−1 | 9.72 × 10−2 | 2.03 × 10−3 | 2.21 × 10−1 | 6.13 × 10−2 |
Example | RLVNS-NSGA-II | VNS-NSGA-II | SVNS-NSGA-II | ||||||
---|---|---|---|---|---|---|---|---|---|
1 | 7.23 × 10−1 | 1.21 | 1.08 | 5.93 × 10−1 | 1.02 | 8.80 × 10−1 | 7.94 × 10−1 | 1.10 | 9.63 × 10−1 |
2 | 8.12 × 10−1 | 1.20 | 1.06 | 6.00 × 10−1 | 1.12 | 8.82 × 10−1 | 5.44 × 10−1 | 1.10 | 8.67 × 10−1 |
3 | 9.13 × 10−1 | 1.16 | 1.05 | 4.66 × 10−1 | 9.54 × 10−1 | 7.86 × 10−1 | 3.60 × 10−1 | 1.01 | 8.01 × 10−1 |
4 | 8.46 × 10−1 | 1.19 | 1.03 | 6.30 × 10−1 | 9.35 × 10−1 | 7.61 × 10−1 | 6.37 × 10−1 | 9.58 × 10−1 | 8.13 × 10−1 |
5 | 7.98 × 10−1 | 1.17 | 1.01 | 4.14 × 10−1 | 9.18 × 10−1 | 7.06 × 10−1 | 6.49 × 10−1 | 1.01 | 7.80 × 10−1 |
6 | 7.83 × 10−1 | 1.09 | 9.56 × 10−1 | 4.92 × 10−1 | 9.38 × 10−1 | 7.27 × 10−1 | 4.98 × 10−1 | 9.82 × 10−1 | 7.53 × 10−1 |
7 | 8.27 × 10−1 | 1.20 | 1.01 | 4.21 × 10−1 | 1.02 | 7.76 × 10−1 | 5.30 × 10−1 | 1.15 | 8.66 × 10−1 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Yang, B.; Chen, J.; Xiao, X.; Li, S.; Ren, T. An Enhanced NSGA-II Driven by Deep Reinforcement Learning to Mixed Flow Assembly Workshop Scheduling System with Constraints of Continuous Processing and Mold Changing. Systems 2025, 13, 659. https://doi.org/10.3390/systems13080659
Yang B, Chen J, Xiao X, Li S, Ren T. An Enhanced NSGA-II Driven by Deep Reinforcement Learning to Mixed Flow Assembly Workshop Scheduling System with Constraints of Continuous Processing and Mold Changing. Systems. 2025; 13(8):659. https://doi.org/10.3390/systems13080659
Chicago/Turabian StyleYang, Bihao, Jie Chen, Xiongxin Xiao, Sidi Li, and Teng Ren. 2025. "An Enhanced NSGA-II Driven by Deep Reinforcement Learning to Mixed Flow Assembly Workshop Scheduling System with Constraints of Continuous Processing and Mold Changing" Systems 13, no. 8: 659. https://doi.org/10.3390/systems13080659
APA StyleYang, B., Chen, J., Xiao, X., Li, S., & Ren, T. (2025). An Enhanced NSGA-II Driven by Deep Reinforcement Learning to Mixed Flow Assembly Workshop Scheduling System with Constraints of Continuous Processing and Mold Changing. Systems, 13(8), 659. https://doi.org/10.3390/systems13080659