Dynamic Job Shop Environment Research Articles

A data-driven simulation-optimization framework for generating priority dispatching rules in dynamic job shop scheduling with uncertainties

Modeling and optimizing dynamic job shop scheduling problems (DJSSP) without ample assumptions is inherently challenging due to the increasing complexity and uncertainty arising from various disruptive events. Moreover, establishing job due dates in dynamic job shop environments has become increasing difficult due to unforeseen events, while it remains a pivotal factor in determining tardiness-related objectives. A research gap exists in exploring the potential synergy between the two aforementioned aspects and its resulting implications. In this study, we propose a novel framework that combines data-driven simulation, machine learning, and evolutionary algorithm to generate new Priority Dispatching Rules (PDRs) for optimizing makespan and total tardiness. The framework incorporates a data-exchange mechanism for real-time updates of the simulation model based on the shop floor. Moreover, an online due date prediction model using XGBoost is developed, replacing the previous rough estimation of the Total Work Content (TWK) rule. Simulation experiments of various scales demonstrate that XGBoost-assisted due date prediction leads to an average reduction of 1.6% in makespan and 65.0% in total tardiness compared to the TWK rule. The proposed method outperforms existing well-known dispatching rules, yielding considerable improvements with average reductions of 90.8%, 40.3%, and 44.6% in total tardiness for vary-sized experiments. Finally, the approach is validated through a real industrial case study.

Combining Reinforcement Learning Algorithms with Graph Neural Networks to Solve Dynamic Job Shop Scheduling Problems

Smart factories have attracted a lot of attention from scholars for intelligent scheduling problems due to the complexity and dynamics of their production processes. The dynamic job shop scheduling problem (DJSP), as one of the intelligent scheduling problems, aims to make an optimized scheduling decision sequence based on the real-time dynamic job shop environment. The traditional reinforcement learning (RL) method converts the scheduling problem with a Markov process and combines its own reward method to obtain scheduling sequences in different real-time shop states. However, the definition of shop states often relies on the scheduling experience of the model constructor, which undoubtedly affects the optimization capability of the reinforcement learning model. In this paper, we combine graph neural network (GNN) and deep reinforcement learning (DRL) algorithm to solve DJSP. An agent model from job shop state analysis graph to scheduling rules is constructed, thus avoiding the problem that traditional reinforcement learning methods rely on scheduling experience to artificially set the state feature vectors. In addition, a new reward function is defined, and the experimental results prove that our proposed reward method is more effective. The effectiveness and feasibility of our model is demonstrated by comparing with general deep reinforcement learning algorithms on minimizing the earlier and later completion time, which also lays the foundation for solving the DJSP later.

Research on a Real-time Job-shop Scheduling Method Based on Reinforcement Learning

At present, manufacturing models are characterized by multi-variety, small batch, and diversification. It is insufficient to use traditional scheduling methods for production management with high performance. A real-time production scheduling system based on reinforcement learning (RL) is suggested in an effort to address the aforementioned issues. A brand-new manufacturing neural network is created to learn the state-action values for production scheduling in real time using high-dimensional data as the input. The detailed setup of network inputs, neural network, action, and reward are also designed. Then, a policy-based reinforcement learning algorithm is proposed to achieve the optimum objective. Finally, By contrasting the proposed scheduling strategy with rule-based approaches in a smart manufacturing environment, its efficacy is demonstrated. according to experimental data, the suggested algorithm can successfully improve performance in the dynamic job-shop environment.

A new approach for performance assessment of parallel and non-bottleneck machines in a dynamic job shop environment

PurposeThe current study aims to propose a new analytical approach by considering energy consumption (EC), maximum tardiness and completion time as the primary objective functions to assess the performance of parallel, non-bottleneck and multitasking machines operating in dynamic job shops.Design/methodology/approachAn analytical and iterative method is presented to optimize a novel dynamic job shop under technical constraints. The machine’s performance is analyzed by considering the setup energy. An optimization model from initial processing until scheduling and planning is proposed, and data sets consisting of design parameters are fed into the model.FindingsSignificant variations of EC and tardiness are observed. The minimum EC was calculated to be 141.5 hp.s when the defined decision variables were constantly increasing. Analysis of the optimum completion time has shown that among all studied methods, first come first served (FCFS), earliest due date (EDD) and shortest processing time (SPT) have resulted in the least completion time with a value of 20 s.Originality/valueConsiderable amount of energy can be dissipated when parallel, non-bottleneck and multitasking machines operate in lower-power modes. Additionally, in a dynamic job shop, adjusting the trend and arrangement of decision variables plays a crucial role in enhancing the system’s reliability. Such issues have never caught the attention of scientists for addressing the aforementioned problems. Therefore, with these underlying goals, this paper presents a new approach for evaluating and optimizing the system’s performance, considering different objective functions and technical constraints.

Manufacturing flow time estimation using the model-tree induction approach in a dynamic job shop environment

We describe an application of a supervised learning method, namely, model-tree induction, for the estimation of flow time in a dynamic job shop. An attempt is made to incorporate transfer batching as order review and release policy in the simulation model, whereas much of previous research appears to have ignored the order-review and release policy. The computational evaluation performance shows that the prediction accuracy of the model-tree induction method is superior to traditional methods such as total-work-content method and the dynamic-total-work-content method in estimating the flow time. This research work has been carried out with an open-source software in order to demonstrate the use of machine learning tools by the resource-constrained small- and medium-enterprises in their decision making processes.

Manufacturing flow time estimation using the model-tree induction approach in a dynamic job shop environment

Manufacturing flow time estimation using the model-tree induction approach in a dynamic job shop environment

An Efficient Approach for Bottleneck Resource(s) Detection Problem in the Multi-objective Dynamic Job Shop Environments

Nowadays energy saving is one of the crucial aspects in decisions. One of the approaches in this case is efficient use of resources in the industrial systems. Studies in real manufacturing systems indicating that one or more machines may also act as the Bottleneck Resource/ Resources (BR). On the other hand according to the Theory of Constraints (TOC), the efficient use of resources in manufacturing systems is limited by the capacity of the BR(s). Hence, in order to improveing such systems performance, the BR(s) should be identified and assessed and improved the using capacity of such resources to the greatest extent possible. Studies indicating that Bottleneck Resource Detection (BRD) problem in the “Multi-Objective and the Dynamic conditions” of job-shop is an important issue which has not been studied in the previous literature due to its computational complexity. Hence the development of an efficient approach to identify and assess BRs in Multi-objective Dynamic Job Shop (MODJS) has been considered as the subject of this paper. In this article, a BRD method based on the Taguchi method for MODJS (TM-MODJS) has been developed. The mentioned method takes the objectives of the MODJS as estimated indices and carries out typical and finite number of experiments by combining different suitable dispatching rules to detect BR(s) which have the greatest effect on the estimated index. Comparing the results indicates effectiveness of the developed method especially in scheduling results in a reasonable time.

Evaluation of Strategies for the Coupling of Central Planning and Autonomous Control in Dynamic Job Shop Environments

Autonomous control is a promising approach to cope with increasing dynamics in production environments. However, an important issue for the acceptance of autonomous control is the provision of a sufficient planning accuracy, e.g., regarding order sequences or workstation assignments. Therefore, transferring autonomous control into practical application requires the coupling of autonomous control with commonly used central planning methods. This paper introduces coupling strategies with different degrees of planning adherence. The introduced strategies are applied to the example of a job shop environment based on real data. The simulation-based comparison considers performance indicators concerning the logistic performance and planning adherence. Besides varying the level of dynamic influences, the simulation study also analyses the impact of different method combinations onto the coupled methods’ performance.

Dynamic job shop scheduling with fixed interval deliveries

Most classical scheduling models assume a job is delivered to a customer immediately after job processing is complete. In numerous practical situations, however, multiple delivery dates exist, and the time interval between any two consecutive delivery dates is constant. A finished job is supplied to a customer by truck on the earliest date in a series of fixed delivery dates, typically at or after processing is complete. This fixed delivery strategy results in substantial cost savings when delivery is expensive or complex. The goal of this study is to minimize the sum of due-date cost and earliness penalty associated with jobs scheduling in a dynamic job shop environment. The due date cost for a job is incurred for time spent delivering a job to a customer. Earliness penalty is incurred if a job is completed before the delivery date. This study identifies three dispatching rules, and proposes nine new rules by explicitly considering different due date costs per time unit and the earliness penalty per time unit of a job. The proposed rules are simple and easily implemented without preliminary runs for parameter estimation. Simulation results show that the proposed dispatching rules are significantly superior to their counterparts.

Analyzing the Influence of Capacity Adjustments on Performance Robustness in Dynamic Job-shop Environments

To keep up a high performance and to stay profitable, manufacturing systems need to be robust against fluctuations and disturbances. In this paper we present a brief overview on robustness measures in manufacturing and investigate the trade-off between robustness of operational performance and cost-efficiency. We further conduct a simulation study on a real world manufacturing system to analyze the influence of capacity adjustments on the performance robustness of different operational key figures.

An integrated scheduling and material-handling approach for complex job shops: a computational study

We suggest an extension of the shifting bottleneck heuristic for complex job shops that takes the operations of automated material-handling systems (AMHS) into account. The heuristic is used within a rolling horizon approach. The job-shop environment contains parallel batching machines, machines with sequence-dependent setup times, and re-entrant process flows. Jobs are transported by an AMHS. Semiconductor wafer fabrication facilities (wafer fabs) are typical examples for manufacturing systems with these characteristics. Our primary performance measure is total weighted tardiness (TWT). The shifting bottleneck heuristic (SBH) uses a disjunctive graph to decompose the overall scheduling problem into scheduling problems for single machine groups and for transport operations. The scheduling algorithms for these scheduling problems are called subproblem solution procedures (SSPs). We consider SSPs based on dispatching rules. In this paper, we are also interested in how much we can gain in terms of TWT if we apply more sophisticated SSPs for scheduling the transport operations. We suggest a Variable Neighbourhood Search (VNS) based SSP for this situation. We conduct simulation experiments in a dynamic job-shop environment in order to assess the performance of the suggested algorithms. The integrated SBH outperforms common dispatching rules in many situations. Using near to optimal SSPs leads to improved results compared with dispatching based SSPs for the transport operations.

Reinforcement Learning Based Job Shop Scheduling with Machine Choice

Job shop scheduling is a key technology in modern manufacturing. Scheduling performance will decide the enterprises’ core competitiveness. In this paper, improved reinforcement learning with cohesion is used in dynamic job shop environment, and it eased the contradiction of precocious and slow convergence. Also the machine choice is considered. So the dual scheduling which included job and machine is achieved in this system. And it obtains better results through the experiments. The utilization of equipments and the emergency handling capacity can be improved in the dynamic environment.

213 動的なジョブショップ環境下におけるマッチアップ・スケジューリング(OS2-3 動的不確実環境に強いシステム)

213 動的なジョブショップ環境下におけるマッチアップ・スケジューリング(OS2-3 動的不確実環境に強いシステム)

Genetic algorithm-based subproblem solution procedures for a modified shifting bottleneck heuristic for complex job shops

In this paper, we consider a modified shifting bottleneck heuristic for complex job shops. The considered job shop environment contains parallel batching machines, machines with sequence-dependent setup times and reentrant process flows. Semiconductor wafer fabrication facilities (Wafer Fabs) are typical examples for manufacturing systems with these characteristics. Our primary performance measure is total weighted tardiness (TWT). The shifting bottleneck heuristic uses a disjunctive graph to decompose the overall scheduling into scheduling problems for single tool groups. The scheduling algorithms for these scheduling problems are called subproblem solution procedures (SSPs). In previous research, only subproblem solution procedures based on dispatching rules have been considered. In this paper, we are interested in how much we can gain in terms of TWT if we apply more sophisticated subproblem solution procedures like genetic algorithms for parallel machine scheduling. We conduct simulation experiments in a dynamic job shop environment in order to assess the performance of the suggested subproblem solution procedures. It turns out that using near to optimal subproblem solution procedures leads in many situations to improved results compared to dispatching-based subproblem solution procedures.

ESTIMATION OF JOB COMPLETION TIMES IN A DYNAMIC SHOP JOB BASED ON STATIC JOB SHOP SOLUTION

We investigate the temporal behaviour of completion time estimates derived from scheduling algorithms that are based on the set of already known and accepted orders (static job shop problem). In a dynamic job shop environment, these estimates change over time, typically increasing. This leads to a systematic error in the completion time estimation based on static job shop algorithms.We propose an approach for calculating this systematic error. Furthermore, a method for compensating this error is derived. For some simple examples, the theory is worked out and compared with simulation results. In all cases, the estimates based on the static algorithm can substantially be improved.

Development and evaluation of a tree-indexing approach to improve case-based reasoning: illustrated using the due date assignment problem

In this study a novel case indexing approach is proposed for case-based reasoning (CBR). This new approach, called the tree-indexing approach, is a modified form of the inductive learning-indexing (IL-indexing) approach and is especially applied to assist CBR in numeric prediction. The tree-indexing approach organizes the cases in the memory by inducting a tree-shaped structure, in order to improve the efficiency and effectiveness of case retrieval. The experiments, using three real world problems from the UCI repository, show that the CBR with the tree-indexing approach (T-CBR) is superior to the conventional CBR. This study also applies T-CBR for solving the due date assignment problem in a dynamic job shop environment in order to investigate whether T-CBR's expected benefits can be observed in practice. The results of the experiments show that our proposed T-CBR can indeed more accurately predict the job due date than the other methods presently in use.

A distributed shifting bottleneck heuristic for complex job shops

In this paper, we consider distributed versions of a modified shifting bottleneck heuristic for complex job shops. The considered job shop environment contains parallel batching machines, machines with sequence-dependent setup times and reentrant process flows. Semiconductor wafer fabrication facilities are typical examples for manufacturing systems with these characteristics. The used performance measure is total weighted tardiness (TWT). We suggest a two-layer hierarchical approach in order to decompose the overall scheduling problem. The upper (or top) layer works on an aggregated model. Based on appropriately aggregated routes it determines start dates and planned due dates for the jobs within each single work area, where a work area is defined as a set of parallel machine groups. The lower (or base) layer uses the start dates and planned due dates in order to apply shifting bottleneck heuristic type solution approaches for the jobs in each single work area. We conduct simulation experiments in a dynamic job shop environment in order to assess the performance of the heuristic. It turns out that the suggested approach outperforms a pure First In First Out (FIFO) dispatching scheme and provides a similar solution quality as the original modified shifting bottleneck heuristic.

Sequencing shop floor operations: a practical approach

PurposeThe work sequence indicated by a production schedule generated by existing scheduling techniques may not necessarily meet the need for efficient sequencing of day‐to‐day shop floor operations in a dynamic job shop environment. The purpose of this paper is to present a new practical approach to efficiently sequencing day‐to‐day shop floor operations of a job shop.Design/methodology/approachA color coding and priority system is introduced to identify production jobs that are to be processed at work centers which have a work sequence preference for performing various jobs. Based on a realistic production schedule, a three‐phase sequencing method is developed to create a truly feasible shop schedule, with the most efficient work sequence possible.FindingsSpecific features of the approach are presented and illustrated with a numerical example. It enables production jobs to be run in the best possible way at individual work centers. Various options of implementing the approach in practical applications are discussed.Research limitations/implicationsThe approach serves as an enhanced supplement to a job shop operations scheduling system for efficient sequencing of shop floor operations. It is to be implemented at the shop floor level.Practical implicationsThe approach can be implemented as a real‐time computer integrated shop dispatching system to ensure an efficient work sequence for shop floor operations of job shops in actual industrial settings.Originality/valueThe paper addresses the need for effectively coordinating production jobs of varying routings on the shop floor, which cannot be met by existing scheduling techniques or shop dispatching practice. It provides manufacturing practitioners with a structured approach for managing shop floor operations.

Using Data Mining for Due Date Assignment in a Dynamic Job Shop Environment

Due date assignment is an important task in shop floor control, affecting both timely delivery and customer satisfaction. Due date related performances are impacted by the quality of the due date assignment methods. Among the simple and easy to implement due date assignment methods, the total work content (TWK) method achieves the best performance for tardiness related performance criteria and is most widely used in practice and in study. The performance of the TWK method can be improved if the due date allowance factor k could render a more precise and accurate flowtime estimation of each individual job. In this study, in order to improve the performance of the TWK method, we have presented a model that incorporated a data mining tool – Decision Tree – for mining the knowledge of job scheduling about due date assignment in a dynamic job shop environment, which is represented by IF-THEN rules and is able to adjust an appropriate factor k according to the condition of the shop at the instant of job arrival, thereby reducing the due date prediction errors of the TWK method. Simulation results show that our proposed rule-based TWK due date assignment (RTWK) model is significantly better than its static and dynamic counterparts (i.e., TWK and Dynamic TWK methods). In addition, the RTWK model also extracted comprehensive scheduling knowledge about due date assignment, expressed in the form of IF-THEN rules, allowing production managers to easily understand the principles of due date assignment .

Dynamic rescheduling using a simulation‐based expert system

The development of a rule‐based expert system (ES), driven by a discrete event simulation model, that performs dynamic shop scheduling is described. Based on a flowtime prediction heuristic that has been developed and base‐line runs to establish the efficacy of scheduling strategies such as shortest processing time (SPT), critical ratio, total work, etc., a rescheduling‐based dispatching strategy is investigated in a dynamic job shop environment. The results are discussed and analyzed.