Flexible Job Shop Scheduling Research Articles

Multi-agent model-based intensification-driven tabu search for solving the dual-resource constrained flexible job shop scheduling

ABSTRACT In today's rapidly evolving manufacturing landscape, the efficient allocation of human resources is crucial for optimizing production systems. This paper addresses the dual resource-constrained flexible job shop scheduling problem (DRCFJSP) by implementing a Multi-Start Tabu Search-based Multi-Agent Model (MuSTAM). MuSTAM considers a set of initial solutions that are run in parallel using an intensification technique. The main objective of MuSTAM is to reduce the completion time (makespan), which is an important metric in workshops. The proposed model is composed of two classes of agents: a supervisor agent named MainAgent and a set of tabu search agents called TabuAgents. The MainAgent launches the system, collects inputs, generates the initial population, creates TabuAgents based on the number of solutions in the initial population (PopSize), and displays the best solution. Each TabuAgent receives a solution from the initial population and then conducts a concentrated search around its neighbourhood space by applying the intensification technique of tabu search. TabuAgents cooperate and communicate with each other to enhance search quality. Numerical tests are performed during the experimental phase to compare MuSTAM with ITS based on well-known benchmark instances of the DRCFJSP. The results demonstrate that MuSTAM is effective in terms of makespan and CPU time.

Minimum-Energy Scheduling of Flexible Job-Shop Through Optimization and Comprehensive Heuristic

This study considers a flexible job-shop scheduling problem where energy cost savings are the primary objective and where the classical objective of the minimization of the make-span is replaced by the satisfaction of due times for each job. An original two-level mixed-integer formulation of this optimization problem is proposed, where the processed flows of material and their timing are explicitly considered. Its exact solution is discussed, and, considering its computational complexity, a comprehensive heuristic, balancing energy performance and due time constraint satisfaction, is developed to provide acceptable solutions in polynomial time to the minimum-energy flexible job-shop scheduling problem, even when considering its dynamic environment. The proposed approach is illustrated through a small-scale example.

A Two-Individual-Based Evolutionary Algorithm for Flexible Assembly Job Shop Scheduling Problem with Uncertain Interval Processing Times

The assembly job shop is a prevalent production organization mode in manufacturing enterprises. During the processing and assembly of products, operation processing times are influenced by numerous factors, leading to significant uncertainty. This paper investigates the flexible assembly job shop scheduling problem (FAJSP) with uncertain processing times, where processing times are represented as variable interval numbers. We develop a robust optimization model for the FAJSP, utilizing confidence level estimation to determine the ranges of processing times and reformulating the model based on the chance-constrained method. A two-individual-based master–apprentice evolutionary (MAE) algorithm is proposed. Two effective encoding schemes are designed to prevent the generation of infeasible solutions under assembly sequence constraints. Additionally, a decoding method based on interval scheduling theory is devised to accurately represent interval processing times. Case studies are conducted to validate the effectiveness of the proposed robust optimization model and demonstrate the superiority of the MAE algorithm.

Fuzzy flexible job shop dynamic scheduling considering machine remaining processing capacity

Dynamic fuzzy flexible job shop scheduling (DFFJSP) is significant in the medical equipment industry. Generally the dynamic flexible job shop scheduling problem considers a single dynamic factor or ignores resources such as machine itself also plays a huge role in dealing dynamic issues. Thus fuzzy processing time and machine breakdowns are considered in this paper. In order to copy with two problems mentioned above, the model of DFFJSP containing two objectives of makespan and energy consumption is established. To solve this problem, the remaining processing capacity of broken machine is applied to weaken the impacts of machine breakdowns and a multi-objective evolutionary algorithm (NG-WOA) is proposed, which integrating NSGA-II and whale optimal algorithm (WOA). Furthermore, the joining of variable neighborhood search makes NG-WOA skip the local optimum solution in time. Through experiments, the validity of remaining processing capacity is verified in rescheduling scheme. Then NG-WOA algorithm is proven can find better solutions when compared with other two algorithms according to the distribution of pareto forefront solutions in benchmark examples. The results of this paper show that the remaining processing capacity has greater advantages in guaranteeing the stability of dynamic scheduling problems, because the machine’s processing capacity can be utilized fully compared to previous strategies for fault disturbances.

Discrete state transfer algorithm based on deep reinforcement learning for solving flexible job shop scheduling problem

Flexible Job Shop Scheduling (FJSP) is a scheduling problem widely used in real life. The research on its intelligent algorithm has important academic significance and application value. In order to solve FJSP, this paper proposes a discrete state transfer algorithm based on proximal policy optimization (DSTA-PPO) with the optimization goal of minimizing the maximum completion time. DSTA-PPO has the following three characteristics: (1) Considering that FJSP requires simultaneous scheduling of process sorting and machine allocation, a state feature that can fully express the current scheduling problem is designed by combining process coding and machine coding. (2) A variety of critical path-based search operations are designed for process sorting and machine allocation. (3) Through reinforcement learning training, it can effectively guide the agent to select the correct search operation to optimize the current scheduling sequence. Through simulation experiments based on different data sets, the effectiveness of each link of the algorithm was verified. At the same time, it was compared with the existing algorithms on the same example with minimizing the maximum completion time. The comparison results show that the proposed algorithm can solve the examples in most examples with a shorter completion time, and effectively solve the flexible job shop scheduling problem.

A hybrid particle swarm optimisation for flexible casting job shop scheduling problem with batch processing machine

AbstractA flexible casting job shop scheduling problem (FCJSP) with batch processing machines is proposed based on the analysis of the flexible job shop scheduling problem (FJSP) and the study of the expendable casting process. Considering the makespan under the influence of the energy consumption, the authors apply the time execution window to the FCJSP model in conjunction with the characteristics of casting production. A hybrid particle swarm optimisation algorithm (HPSO) is developed to solve the FCJSP. The HPSO employs a block integration decoding rule to address scheduling integration. Particle swarm optimisation is used for global search, employing both discrete and continuous search strategies. Furthermore, the local search employs tabu search with neighbourhood operations based on knowledge‐driven techniques. Simulation experiments demonstrate the feasibility of the proposed optimisation model. In the end, the HPSO algorithm has been successfully applied to the real expendable casting scheduling. The results demonstrate that it is more efficient and robust than previously reported algorithms.

Deep reinforcement learning driven trajectory-based meta-heuristic for distributed heterogeneous flexible job shop scheduling problem

As the production environment evolves, distributed manufacturing exhibits heterogeneous characteristics, including diverse machines, workers, and production processes. This paper examines a distributed heterogeneous flexible job shop scheduling problem (DHFJSP) with varying processing times. A mixed integer linear programming (MILP) model of the DHFJSP is formulated with the objective of minimizing the makespan. To solve the DHFJSP, we propose a deep Q network-aided automatic design of a variable neighborhood search algorithm (DQN-VNS). By analyzing schedules, sixty-one types of scheduling features are extracted. These features, along with six shaking strategies, are used as states and actions. A DHFJSP environment simulator is developed to train the deep Q network. The well-trained DQN then generates the shaking procedure for VNS. Additionally, a greedy initialization method is proposed to enhance the quality of the initial solution. Seven efficient critical path-based neighborhood structures with three-vector encoding scheme are introduced to improve local search. Numerical experiments on various scales of instances validate the effectiveness of the MILP model and the DQN-VNS algorithm. The results show that the DQN-VNS algorithm achieves an average relative percentage deviation (ARPD) of 3.2%, which represents an approximately 88.45% reduction compared to the best-performing algorithm among the six compared, with an ARPD of 27.7%. This significant reduction in ARPD highlights the superior stability and performance of the proposed DQN-VNS algorithm.

Multi-objective scheduling for an energy-efficient flexible job shop problem with peak power constraint

Controlling electricity costs, which are closely related to the peak power reached, while simultaneously reducing completion time and energy consumption, is crucial for achieving sustainability in the manufacturing industry. However, most existing research on energy-efficient flexible job shop scheduling problems primarily focuses on optimizing completion time and energy consumption, often neglecting the critical aspect of controlling electricity costs. With decreasing resources and increasing energy prices, it is necessary to control electricity costs by limiting the peak power reached during production. To address this gap, this paper investigates an energy-efficient flexible job shop scheduling problem with peak power constraint considering setup and transportation time (EEFJSSP-PPST). We establish a mathematical model for EEFJSSP-PPST and propose an improved non-dominated sorting genetic algorithm II (INSGA-II) incorporating several key improvements. We first introduce three scheduling rules in decoding to optimize objectives and employ heuristic rules to generate a high-quality initial population. Then, we propose an innovative cluster crossover method to accelerate convergence and design an adaptive-based local optimization strategy to further enhance performance. Finally, our experiments explore the impacts of different degrees of the peak power constraint on objectives and evaluate the effectiveness of the three scheduling rules. Additionally, the performance of INSGA-II is tested using 135 benchmark instances. The results indicate that INSGA-II outperformed its variations without improvements, achieving the best hypervolume (HV) indicator in 80% of the instances and the best inverted generational distance (IGD) indicator in 68.89% of the instances. Compared to other state-of-the-art algorithms, INSGA-II achieves the best HV and IGD indicators in 82.2% and 81.5% of the instances, respectively, and demonstrates better convergence and a superior Pareto front. Therefore, we conclude that the improvements in INSGA-II are effective and enhance the algorithm’s performance, making it outstanding in solving EEFJSSP-PPST.

Genetic Programming With Lexicase Selection for Large-Scale Dynamic Flexible Job Shop Scheduling

Dynamic flexible job shop scheduling is a prominent combinatorial optimisation problem with many real-world applications. Genetic programming has been widely used to automatically evolve effective scheduling heuristics for dynamic flexible job shop scheduling. A limitation of genetic programming is the premature convergence due to the loss of population diversity. To overcome this limitation, this work considers using lexicase selection to improve population diversity, which has achieved success on regression and program synthesis problems. However, it is not trivial to apply lexicase selection to genetic programming for dynamic flexible job shop scheduling, since a fitness case (training scheduling simulation) is often large-scale, making the fitness evaluation very time-consuming. To address this issue, we propose a new multi-case fitness scheme, which creates multiple cases from a single scheduling simulation. Based on the multi-case fitness, we develop a new genetic programming algorithm with lexicase selection, which uses a single simulation for fitness evaluation, thus achieving a better balance between the number of cases for lexicase selection and evaluation efficiency. The experiments on a wide range of dynamic scheduling scenarios show that the proposed algorithm can achieve better population diversity and final performance than the current genetic programming parent selection methods and a state-of-the-art deep reinforcement learning method.

Dynamic production scheduling and maintenance planning under opportunistic grouping

Maintenance and production scheduling are intertwined activities that should be addressed simultaneously to uphold production systems’ reliability and production efficiency. The digital twin is an emerging technology that advances industrial digitalization, as it embeds a “virtual” image of reality that runs in line with the real system, enabling evaluation, optimization, and prediction of the physical system’ state. On the other hand, deep reinforcement learning (DRL) can provide real-time decision-making by leveraging real-time data from the digital twin such as condition monitoring and production progress data. This research addresses the joint flexible job shop scheduling and maintenance planning problem, considering new job insertions and multi-component machines with economic dependencies among components. The objective is to minimize both the expected total tardiness and maintenance cost, considering opportunistic grouping of maintenance activities on components and breakdown costs associated with failure risk. To achieve this joint decision-making, we develop a hierarchical architecture with two interconnected hierarchies. The upper-level hierarchy determines whether to switch the decision-making process between production scheduling and maintenance planning. Subsequently, the lower-level hierarchy selects an action through the corresponding agent: a multi-head Deep Q-Network (DQN) if the maintenance option is chosen, and a DQN with seven dispatching rules for the production option. The computational experiment results reveal that the proposed scheduling method can learn a high-quality dispatching policy, outperforming the non-hierarchical DRL approach and individual dispatching rules in solution quality, CPLEX in runtime efficiency, and the genetic algorithm and particle swarm optimization in both solution quality and runtime efficiency.

A Bi-Population Competition Adaptive Interior Search Algorithm Based on Reinforcement Learning for Flexible Job Shop Scheduling Problem

In this paper, a bi-population competition adaptive interior search algorithm (BCAISA) based on a reinforcement learning strategy is proposed for the classical flexible job shop scheduling problem (FJSP) to optimize the makespan. First, the scheduling solution is represented using a machine-job-based two-segment integer encoding method, and various heuristic rules are then applied to generate the initial population. Secondly, a bi-population mechanism is introduced to partition the population into two distinct sub-populations. These sub-populations are specifically tailored for machine assignment and operation permutation, employing different search strategies respectively, aiming to facilitate an efficient implementation of parallel search. A competition mechanism is introduced to facilitate the information exchange between the two sub-populations. Thirdly, the ISA is adapted for the discrete scheduling problem by discretizing a series of search operators, which include composition optimization, mirror search, and random walk. A Q-learning-based approach is proposed to dynamically adjust a key parameter, aiming to strike a balance between the capacity for global exploration and local exploitation. Finally, extensive experiments are conducted based on 10 well-known benchmark instances of the FJSP. The design of the experiment (DOE) method is employed to determine the algorithm’s parameters. Based on the computational results, the effectiveness of four improvement strategies is first validated. The BCAISA is then compared with fifteen published algorithms. The comparative data demonstrate that our algorithm outperforms other algorithms in 50% of benchmark instances. Additionally, according to the relative percentage deviation (RPD) from the state-of-the-art results, the BCAISA also exhibits superior performance. This highlights the effectiveness of our algorithm for solving the classical FJSP. To enhance the practical application, the scope of the ISA will be broadened in future work to more complex problems in real-world scenarios.

An energy-saving distributed flexible job shop scheduling with machine breakdowns

Due to the change of consumption pattern, distributed flexible production is now becoming a major manufacturing mode. The machine breakdown seriously affects the production in workshops. How to efficiently adjust the scheduling scheme after machine breakdowns in distributed flexible job shop is an important concern for enterprises. Meanwhile, energy consumption is attracting more and more attention as environmental issues become increasingly serious. Thus, the energy-saving scheduling on distributed flexible job shop considering machine breakdowns is studied, and a mixed-integer programming model is established to optimize makespan, total energy consumption and machine load difference, simultaneously. Based on the problem characteristics, an improved memetic algorithm is designed. To improve the quality of initial solutions, an initialization strategy incorporating multiple rules is designed. An active decoding method with an adjustment strategy is adopted to fully utilize machine idleness. To expand the search range, the idea of dual-population collaborative optimization is used. Multi-level intelligent mutation operation strategies are introduced to further enhance the quality of solutions. In addition, an adaptive parameter rule is designed to adjust to appropriate search capabilities at different evolutionary stages. To verify the effectiveness of the proposed strategies and algorithm, the memetic algorithm, the artificial bee colony algorithm, the differential evolution algorithm and the estimation of distribution algorithm are chosen as comparison algorithms. The same iteration number and same running time are used as the termination criteria of all algorithms, and 60 instances of different scales are solved. The experimental results demonstrate that the effectiveness of the strategies and algorithm designed in this paper. Notably, our algorithm consistently achieved good results even when the same running time is used as the termination criterion.

A Deep Reinforcement Learning Method Based on a Transformer Model for the Flexible Job Shop Scheduling Problem

The flexible job shop scheduling problem (FJSSP), which can significantly enhance production efficiency, is a mathematical optimization problem widely applied in modern manufacturing industries. However, due to its NP-hard nature, finding an optimal solution for all scenarios within a reasonable time frame faces serious challenges. This paper proposes a solution that transforms the FJSSP into a Markov Decision Process (MDP) and employs deep reinforcement learning (DRL) techniques for resolution. First, we represent the state features of the scheduling environment using seven feature vectors and utilize a transformer encoder as a feature extraction module to effectively capture the relationships between state features and enhance representation capability. Second, based on the features of the jobs and machines, we design 16 composite dispatching rules from multiple dimensions, including the job completion rate, processing time, waiting time, and manufacturing resource utilization, to achieve flexible and efficient scheduling decisions. Furthermore, we project an intuitive and dense reward function with the objective of minimizing the total idle time of machines. Finally, to verify the performance and feasibility of the algorithm, we evaluate the proposed policy model on the Brandimarte, Hurink, and Dauzere datasets. Our experimental results demonstrate that the proposed framework consistently outperforms traditional dispatching rules, surpasses metaheuristic methods on larger-scale instances, and exceeds the performance of existing DRL-based scheduling methods across most datasets.

Enhanced Monte‐Carlo tree search for dynamic flexible job shop scheduling with transportation time constraints

AbstractWith the increasing uncertainty and complexity of industrial production, dynamic events are inevitably happened during the production process. This paper deals with the transportation time constrained dynamic flexible job shop scheduling problem with machine breakdown, new jobs arrival, jobs cancellation, and processing time change of operations. Considering the actual production situation, a makespan rescheduling model based on transportation time is established. A Monte‐Carlo tree search based algorithm is proposed to solve the dynamic flexible job shop scheduling problem such that the makespan is minimized. In order to improve the accuracy of the developed algorithm and the quality of the obtained scheduling schemes, subtree keeping policy, multi‐branching simulation, reinforcement learning, greedy strategy, and expectation‐best evaluation method are employed. Moreover, a subtree pruning policy is embedded to improve the search efficiency of the developed algorithm. Simulation experiments are conducted on a series of instances of different sizes to compare the developed approach with the GA‐based scheduling approach and the Monte‐Carlo tree search based approach. The simulation results indicate that the developed approach is superior to the existing scheduling approaches in terms of the qualities of the obtained solutions.

Dynamic flexible job shop scheduling based on deep reinforcement learning

The unpredictable dynamic events in smart factory seriously influence the scheduling schemes and production efficiency. To minimize the total tardiness of orders, this paper proposes a Deep Reinforcement Learning (DRL) method to solve the Dynamic Flexible Job Shop Scheduling Problem (DFJSP) with random job arrival. In the scheduling process, the intelligent agent can select the operations to be processed on the available machines according to the job shop state at each scheduling point by transforming DFJSP into a Markov Decision Process (MDP). For reflecting the job shop environment, eight state features are extracted, and six composite dispatching rules are designed as the action space. Moreover, a reward function is developed to enhance the learning efficiency and solution quality of the intelligent agent, the Soft-max selection strategy is applied to balance exploration and exploitation. The Dueling Double Deep Q-Network (Dueling DDQN) algorithm is introduced to train the intelligent agent, which effectively alleviates the overestimation problem. Numerical simulation results show that the proposed Dueling DDQN method effectively solves the DFJSP. Compared with existing scheduling strategies, the proposed approach not only exhibits superior performance but also maintains its advantages among different scale instances.

Flexible assembly job shop scheduling problem considering reconfigurable machine: A cooperative co-evolutionary matheuristic algorithm

With popularity of mass customization, enterprises urgently need to improve reconfigurability to handle rapidly changing market demands. Reconfigurable machines have been widely applied in the flexible assembly job shop. These reconfigurable machines can be equipped with various and limited auxiliary modules (AMs) to provide multiple alternative functions to manufacture numerous products. Therefore, this study addresses flexible assembly job shop scheduling problem considering reconfigurable machines with limited AMs to minimize makespan. First, a mixed-integer linear programming (MILP) model is formulated to define the problem. Then, a cooperative co-evolutionary matheuristic algorithm (CCMA) is presented to efficiently tackle large-scale FAJSP-RM problem. In this algorithm, a MILP-based evolution method, which relies on a decomposed MILP (D-MILP) with known sequence decisions, is proposed to explore the optimal machine and AM assignments. Meanwhile, a collaborative initialization and dual collaborative strategy are proposed to improve the performance of the proposed CCMA. Comprehensive experiments are conducted on 720 instances, extended from the well-known Fdata and BRdata benchmarks, to evaluate the proposed MILP model and CCMA algorithm. The results illustrate that the MILP model can produce optimal solutions for small-scale instances. The MILP-based evolution method improves the performance of the CCMA algorithm by 12.63, and the CCMA outperforms other well-known algorithms from numerical, statistical, differential and stable analysis.

Multi-guided population co-evolutionary algorithm based on multiple similarity decomposition for large-scale flexible job shop scheduling problem

As manufacturing shifts towards large-scale production, the size of the workshop increases, and its search space exponentially expands. It is difficult for existing algorithms to obtain an ideal scheduling solution in an acceptable time. For the large-scale flexible job shop scheduling problem (LSFJSP), a multi-guided population co-evolutionary algorithm based on multiple similarity decomposition (MPCSD) is designed. Faced with the problem of high dimensionality and complex solution space, a multiple similarity decomposition strategy is proposed. It proceeds to group based on similarity information at the dimension and population level. To obtain convergence-preferred and diversity-preferred dimension groupings, a training-set solution selection method is proposed. Inspired by the idea of divide-and-conquer, a multi-guided co-evolutionary strategy is proposed. It improves the exploration efficiency of the algorithm in the search space. To test effectiveness on more complex LSFJSP, a set of large-scale test problems including LS1-12 are designed. On LS1-12, MPCSD is compared with seven other algorithms to demonstrate its superiority. MPCSD performed best on 11, 7, and 10 of the 12 test problems on Inverted Generational Distance (IGD), Hypervolume (HV) and Schott’s Spacing Metric (SP), respectively. Meanwhile, the Relative Deviation (RD) results showed that MPCSD obtained the best fitness performance.

Research on Multi-Objective Low-Carbon Flexible Job Shop Scheduling Based on Improved NSGA-II

To optimize the production scheduling of a flexible job shop, this paper, based on the NSGA-II algorithm, proposes an adaptive simulated annealing non-dominated sorting genetic algorithm II with enhanced elitism (ASA-NSGA-EE) that establishes a multi-objective flexible job shop scheduling model with the objective functions of minimizing the maximum completion time, processing cost, and carbon emissions generated from processing. The ASA-NSGA-EE algorithm adopts an adaptive crossover and mutation genetic strategy, which dynamically adjusts the crossover and mutation rates based on the evolutionary stage of the population, aiming to reduce the loss of optimal solutions. Additionally, it incorporates the simulated annealing algorithm to optimize the selection strategy by leveraging its cooling characteristics. Furthermore, it improves the elite strategy through incorporating elite selection criteria. Finally, by simulation experiments, the effectiveness of the improved NSGA-II algorithm is validated by comparing it with other algorithms.

Flexible Job Shop Scheduling Problem-Solving Using Apiary Organizational-Based Optimization Algorithm

Flexible job shop scheduling problem (FJSSP) is a complex and challenging problem that plays a crucial role in industrial and manufacturing production. FJSSP is an expansion of the standard job shop scheduling problem (JSSP). One of FJSSP’s objectives that the manufacturing system competing for is minimizing the makespan. This paper uses a new nature-inspired metaheuristic optimization algorithm called the Apiary Organizational-Based Optimization algorithm (AOOA) to solve the FJSSP. This Algorithm simulates the organizational behavior of honeybees inside the apiary and translates their activities and vital processes during their lifecycle into phases that can solve such NP-hard problems. Two benchmark datasets, Brandimarte and Hurink, with 10 MK instances and 24 (edata, rdata, and vdata) instances respectively, were used to demonstrate the ability of AOOA to solve FJSSP. Moreover, the results of AOOA were compared with a set of state-of-the-art algorithms and statistically measured using the paired samples t-test and p-value, RPD, and group-based superiority statistical analysis to test its performance. AOOA outperformed Elitism GA, Enhanced GA, Improved GA, and MOGWO in solving all 10 MK instances and HICSA in solving 9 MK instances out of 10. Moreover, AOOA overcame CS, CS-BNG, CS-ILF, CHA, and MCA in solving 24, 12, 12, 23, and 24 instances of edata, rdata, and vdata, respectively. AOOA proved its robustness, showing promising outcomes.

A machine learning-based simulation metamodeling method for dynamic scheduling in smart manufacturing systems

Conventional Digital Twins (DTs) in smart manufacturing rely on complex and time-intensive simulation models, hindering real-time DT-based decision-making. However, the availability of big data in Manufacturing Execution Systems (MES) enables training different Machine Learning (ML) models for fast and accurate predictions and decision assessments. Accordingly, this paper proposes an ML-Based Simulation Metamodeling Method (MLBSM) to facilitate DT-based decision-making for dynamic production scheduling in complex Stochastic Flexible Job Shop (SFJS) environments. The proposed MLBSM integrates three modules: a novel data vectorizing method (SPBM), multi-output Adaptive Boosting Regressor (ABR) models, and a new statistical risk evaluation method. SPBM converts unstructured production log data into numerical vectors for ABR training by calculating numeric penalty scores for each job based on the position of operations in the schedule queue. Each trained ABR predicts mean job completion times for various dynamic scenarios based on shift schedules. The risk evaluation method estimates the standard deviation of job completion times and calculates the delay probability scores for each job, aiding DT in promptly evaluating production schedules. Working seamlessly together, MLBSM modules present a novel way of using production log data for ML training and ultimately bypassing several computationally intensive simulation replications. In this research, a simulation model generates the synthetic MES data, focusing on the machining process at a photolithography workstation in the semiconductor manufacturing. Experiments demonstrate the MLBSM’s accuracy and efficiency, predicting high-risk jobs with over 80% recall and being at least 70 times faster than conventional simulation runs. Sensitivity analyses also confirm the MLBSM’s consistency under different workstation conditions.