Scheduling Of Manufacturing Systems Research Articles

AB&B: An Anytime Branch and Bound Algorithm for Scheduling of Deadlock-Prone Flexible Manufacturing Systems

This work investigates a scheduling problem of deadlock-prone flexible manufacturing systems modeled by place-timed Petri nets. It proposes an anytime branch and bound (AB&B) algorithm for it to minimize system makespan based on the branch tree of a net model and a highly permissive deadlock controller. The proposed algorithm searches a sequence of transitions in the branch tree that evolves the model from the initial marking to the final one. In order to prune the branch tree and increase search speed, this work develops two pruning rules, a lower bound of makespan, and a novel branching strategy. Their usage ensures AB&B’s high search efficiency. Experimental results demonstrate that the proposed algorithm surpasses the state-of-the-art ones. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —In practice, scheduling is one of the most important issues for production managers to address. Existing approaches to deadlock-prone flexible manufacturing system (FMS) scheduling have multiparameters. Their settings can impact the scheduling results greatly. Since their turning is a challenging task, it is difficult to find the best parameter settings. This article proposes an anytime branch and bound (AB&B) algorithm for minimizing the makespan of deadlock-prone FMSs. It has only one parameter, i.e., maximal CPU time, which needs no turning. When it is highly tight, AB&B can output a feasible and decent schedule. If it is larger and larger, AB&B can output a better and better schedule. Comparison studies show that AB&B outclasses existing ones significantly. It is suitable for real-time scheduling cases in which satisfied schedules must be offered in a very short time.

Energy-Saving Production Scheduling in a Single-Machine Manufacturing System by Improved Particle Swarm Optimization

A single-machine scheduling problem that minimizes the total weighted tardiness with energy consumption constraints in the actual production environment is studied in this paper. Based on the properties of the problem, an improved particle swarm optimization (PSO) algorithm embedded with a local search strategy (PSO-LS) is designed to solve this problem. To evaluate the algorithm, some computational experiments are carried out using PSO-LS, basic PSO, and a genetic algorithm (GA). Before the comparison experiment, the Taguchi method is used to select appropriate parameter values for these three algorithms since heuristic algorithms rely heavily on their parameters. The experimental results show that the improved PSO-LS algorithm has considerable advantages over the basic PSO and GA, especially for large-scale problems.

Energy-efficient multi-objective flexible manufacturing scheduling

This paper presents a novel scheduling of a resource-constrained Flexible Manufacturing System (FMS) with consideration of the following sub-problems: (i) machine loading and unloading, (ii) manufacturing operation scheduling, (iii) machine assignment, and (iv) Automated Guided Vehicle (AGV) scheduling. In the proposed model, both the AGV and machinery are considered as the required resources. Energy efficiency of AGVs has been studied in order to improve environmental sustainability in terms of a linear function, which is based on load and distance, accordingly. Because of the NP-hard characteristics of the problem, a modified multi-objective particle swarm optimization (MMOPSO) has been developed for solving the model and compared with the classic version of the multi-objective particle swarm optimization (MOPSO) algorithm in terms of five performance metrics. Finally, the results are evaluated by the application of a multi-criteria decision-making (MCDM) algorithm according to which the MMOPSO outperforms the MOPSO.

A Sequential Search Method of Dispatching Rules for Scheduling of LCD Manufacturing Systems

A combination of dispatching rules to determine a sequence of jobs is often used for the scheduling of liquid crystal display or semiconductor manufacturing systems. Fab engineers determine weights ranging between 0 and 1 for dispatching rules, and the dispatching rules assign a value from 0 to 1 for each job. Then the weight given by the engineers and the value from each dispatching rule are multiplied and added for all rules, and the job that has the largest weighted sum is chosen to be processed on an idle machine. Hence, the weights of dispatching rules significantly affect the performance of fabs and should be determined carefully. This article proposes a sequential search method for finding a proper set of weights for dispatching rules to improve an objective value of the fab, such as increasing throughput or decreasing setup times. We use a decision tree approach and a hierarchical clustering method to efficiently search for weights in a short period of time by eliminating some sub-spaces that are less likely to have good objective values. The proposed method finds better weight sets than a random search method and a response optimization method in experiments.

Multi-objective scheduling in labor-intensive manufacturing systems

In a manufacturing facility, the decision maker considers customer satisfaction as an essential priority in which customer due dates should be met. The cell loading, scheduling and the total manpower are considered among the main decisions in a manufacturing system. Three performance measures are studied in this paper and they are the number of tardy jobs, total manpower, and average flow time. Two approaches are proposed to handle such performance measures. In the first approach, lexicographical analysis is proposed to achieve the objectives, when the performance measures are not equally important. A three-phase methodology is introduced. In the first phase, a mathematical model has been used to maximize the output rate by allocating manpower to operations for several manpower levels. In the second phase, another mathematical model is developed based on the results of the first phase to minimize total manpower such that no job is tardy (i.e., the number of tardy jobs (NT = 0)). In the third phase, a mathematical model is also proposed to minimize the average flow time based on the results obtained in the first and the second phases. The second approach allows varying weights for minimizing average flow time and total manpower subject to zero tardy jobs due to considerations for customer satisfaction. Thus, a two-phase methodology is followed, and a bi-objective fuzzy mathematical model is developed. Two-phase methodology produced similar results to three-phase methodology for certain values of weights assigned to performance measures. Three-phase methodology was efficient for lexicographical analysis. Furthermore, the number of cells in the system is affected by the nature of due dates. The number of cells and total manpower increase when the due dates are tight and decrease when the due dates are relaxed.

Incorporating order acceptance, pricing and equity considerations in the scheduling of cloud manufacturing systems: matheuristic methods

ABSTRACT Rooted from the Industry 4.0 principles, Cloud Manufacturing (CMfg) is a novel customer-oriented manufacturing norm, which can assist enterprises to withstand in the nowadays highly volatile and competitive market. CMfg systems comprise two separate parties, namely, customers and factories, with independent individuals. In this regard, considering the utilities of both customers and factories and establishing the equity amongst their individuals are of particular importance for the survival and flourishment of CMfg systems. Furthermore, due to the limited capacity of resources, tightness of due dates, and customers’ cost expectations, all orders may not be accepted in CMfg systems. Accordingly, this paper aims to explore a scheduling problem in a CMfg system. A multi-objective mathematical model is presented for the problem, which can determine the acceptance or rejection of orders, set prices, and schedule them in an integrated manner to maximise the customers and factories’ utilities, and enhance the equity among their members. Due to the high complexity of the problem, two matheuristic methods based on the Multi-Objective Grey Wolf Optimizer (MOGWO) and Non-dominated Sorting Genetic Algorithm II (NSGA-II) are developed. An extensive computational experiment is carried out to validate the proposed matheuristic methods and evaluate their performance. Moreover, some guidance is presented for managers by conducting a sensitivity analysis.

A discrete firefly algorithm for solving the flexible job-shop scheduling problem in a make-to-order manufacturing system

This work presents a multi-objective discrete firefly algorithm (MO-DFFA) for solving the flexible job-shop scheduling problem (FJSP) in a make-to-order production. Three different objectives are minimised simultaneously, being these objectives the weighted sum of the completion times of the orders, the workload of the critical machine and the total workload of all machines. Customer orders are ranked by priority according to the variables that the company considers the most relevant for its classification. Then, this priority is included in the FJSP model giving more preference in the scheduling phase to client requests with higher priority while, at the same, the volume of work of the resources is balanced to avoid machines saturation or under-utilization. With this approach both customer and company requirements can be satisfied and balanced. Furthermore, in the proposed framework customers can customize their orders choosing between some eligible product characteristics, which are considered as the different manufacturing operations which constitute each job. For solving the two sub-problems (i.e. operations assignment and sequencing) required for the scheduling of a flexible manufacturing system, we implemented a discrete version of the firefly algorithm metaheuristic. The computational results of several problem instances show that the presented MO-DFFA is a promising and efficient alternative to solve the FJSP in a customer-centric production system.

Maintenance scheduling for flexible multistage manufacturing systems with uncertain demands

ABSTRACT This paper proposes a maintenance scheduling method for the flexible multistage manufacturing system in multi-specification and small-batch production. The bi-directional interactions between the production and the deterioration of the station, and the uncertainty of the future production demands are mainly involved. The workload of each station changes with the dynamic production schedule, which will have a marked impact on the deterioration of the station. Meanwhile, the deterioration state of the station will in turn influence the selection of the station to complete the production tasks. Based on this interaction, a load integrated deterioration model is established, and then a cost-effective maintenance scheduling model is proposed for the system. Because of the uncertainty of the future production demands, the optimal preventive maintenance scheme for the system is obtained by minimizing the expected total maintenance cost per unit time within the next uncertain production period. To simplify the solving process, a greedy constraint algorithm is developed, with the duration of preventive maintenance being as the constraint. Numerical comparisons show that the expected total maintenance cost under the proposed maintenance scheduling model is always lower than the one under the full load model and the average load model.

New integration of preventive maintenance and production planning with cell formation and group scheduling for dynamic cellular manufacturing systems

This paper proposes a new integration method for cell formation, group scheduling, production, and preventive maintenance (PM) planning problems in a dynamic cellular manufacturing system (CMS). The cell formation sub-problem aims to form part families and machine groups, which minimizes the inter-cell material handling, under-utilization, and relocation costs. The production planning aspect is a multi-item capacitated lot-sizing problem accompanied by sub-contracting decisions, while the group scheduling problem deals with the decisions on the sequential order of the parts and their corresponding completion times. The purpose of the maintenance sub-problem is to determine the availability of the system and the time when the noncyclical perfect PM must be implemented to reduce the number of corrective actions. Numerical examples are generated and solved by Bender’s decomposition pack in GAMS to evaluate the interactions of the proposed model. Statistical analysis, based on a nonparametric method, is also used to study the behavior of the model’s cost components in two different situations. It is shown that by adding the PM planning decisions to the tactical decisions of the dynamic CMS, the optimal configuration and production plans of the system are heavily affected. The results indicate that omitting the PM actions increases the number of sudden failures, which leads to a higher total cost. Finally, it is concluded that the boost in the total availability of the dynamic CMS is one of the main advantages of the proposed integrated method.

A Hybrid Ant Colony Optimization and Simulated Annealing Algorithm for Multi-Objective Scheduling of Cellular Manufacturing Systems

During the last 2 decades, there have been many manufacturing companies in various industries that used the advantages of cellular manufacturing layouts. However, determining the best schedule for cellular layouts considering uncertain product demands is a big concern for scientists. In this research, a multi-objective decision-making model is proposed in the process of dynamic cellular production planning where the market demands are uncertain. In this regard, a non-linear mixed integer programming model is developed. The complexity of the model is high to consider the model as NP-hard. Therefore, a hybrid Ant colony Optimization and Simulated Annealing Algorithms are proposed to solve the problem. Then, the Taguchi method is used to estimate appropriate sets of parameters of the proposed algorithm. The results demonstrated that the proposed algorithm can generate the best part-routes of products in terms of time, cost and load variance in a reasonable time. The algorithm is then used for a cellular production plant which is the producer of heavy vehicles parts.

A dynamic decision model for energy-efficient scheduling of manufacturing system with renewable energy supply

The climate mitigation and the reduction of energy cost in manufacturing processes drive to expand the electricity generation from renewable sources. Nonetheless, intermittency of renewable energies, especially solar and wind energy, represents one on the main challenge, typically overcome by the installation of electricity storage systems. This issue can be addressed by a new and original approach, consisting in energy-flexibility of the production, in which manufacturing parameters are selected to optimize and to align production planning to the renewable energy availability. The paper deals with a time dependent theoretical and numerical model developed to calculate the time evolution of the electric power required by a manufacturing system, self-consistently coupled with a renewable plant. The aim of the model is to align the power required by the manufacturing system with the renewable energy supply in order to obtain the maximum monthly profit. The model has been applied to a single work center powered by the electric grid and by a photo-voltaic system, performing the machining process over one year of production. The model includes the tool cost, the stocked units, the energy cost and the penalty for the unsatisfied demand. The maximum profit has been calculated with a hourly adaption of manufacturing parameters, i.e. the cutting speed, to the renewable time dependent power profile. The model presents general features and can be applied when production processes are fully characterized.In order to find the maximum profit, the model, inherently nonlinear, has been solved by recurring to the Trust-Region Method. Different scenarios characterized by fluctuations of product demand are considered in order to investigate the sensitivity of the manufacturing system to the uncertainty of the forecast demand. The influence of the photo-voltaic supply has been investigated, comparing results obtained in the case of manufacturing systems powered only by the electric grid. Numerical results show how the proposed method allows to select an optimized production planning, reducing the energy costs and CO2 emissions and finding the maximum profit with the best compromise between the market demand and energy costs.

A multi-objective particle swarm optimisation for integrated configuration design and scheduling in reconfigurable manufacturing system

To provide accurate capacity and functionality needed for each demand period (DP), a reconfigurable manufacturing system (RMS) is able to change its configuration with time. For the RMS with multi-part flow line configuration that concurrently produces multiple parts within the same family, the cost and delivery time are dependent on its configuration and relating scheduling for any DP. So far, the study on solution method for the integrated optimisation problem of configuration design and scheduling for RMS is scarce. To efficiently find solutions with tradeoffs between total cost and tardiness, a multi-objective particle swarm optimisation (MoPSO) based on crowding distance and external Pareto solution archive is presented to solve practical-sized problems. The devised encoding and decoding methods along with the particle updating mechanism of MoPSO ensure any particle a feasible solution. The comparison between MoPSO and ε-constraint method versus small-sized cases illustrates the effectiveness of MoPSO. The comparative results between MoPSO and nondominated sorting genetic algorithm II (NSGA-II) against eight problems show that the MoPSO outperforms the NSGA-II in both solution quality and computation efficiency for the integrated optimisation problem.

Machine learning for predictive scheduling and resource allocation in large scale manufacturing systems

The digitalization processes in manufacturing enterprises and the integration of increasingly smart shop floor devices and software control systems caused an explosion in the data points available in Manufacturing Execution Systems. The degree in which enterprises can capture value from big data processing and extract useful insights represents a differentiating factor in developing controls that optimize production and protect resources. Machine learning and Big Data technologies have gained increased traction being adopted in some critical areas of planning and control. Cloud manufacturing allows using these technologies in real time, lowering the cost of implementing and deployment. In this context, the paper offers a machine learning approach for reality awareness and optimization in cloud.Specifically, the paper focuses on predictive production planning (operation scheduling, resource allocation) and predictive maintenance. The main contribution of this research consists in developing a hybrid control solution that uses Big Data techniques and machine learning algorithms to process in real time information streams in large scale manufacturing systems, focusing on energy consumptions that are aggregated at various layers. The control architecture is distributed at the edge of the shop floor for data collecting and format transformation, and then centralized at the cloud computing platform for data aggregation, machine learning and intelligent decisions. The information is aggregated in logical streams and consolidated based on relevant metadata; a neural network is trained and used to determine possible anomalies or variations relative to the normal patterns of energy consumption at each layer. This novel approach allows for accurate forecasting of energy consumption patterns during production by using Long Short-term Memory neural networks and deep learning in real time to re-assign resources (for batch cost optimization) and detect anomalies (for robustness) based on predicted energy data.

Petri-net-based dynamic scheduling of flexible manufacturing system via deep reinforcement learning with graph convolutional network

To benefit from the accurate simulation and high-throughput data contributed by advanced digital twin technologies in modern smart plants, the deep reinforcement learning (DRL) method is an appropriate choice to generate a self-optimizing scheduling policy. This study employs the deep Q-network (DQN), which is a successful DRL method, to solve the dynamic scheduling problem of flexible manufacturing systems (FMSs) involving shared resources, route flexibility, and stochastic arrivals of raw products. To model the system in consideration of both manufacturing efficiency and deadlock avoidance, we use a class of Petri nets combining timed-place Petri nets and a system of simple sequential processes with resources (S3PR), which is named as the timed S3PR. The dynamic scheduling problem of the timed S3PR is defined as a Markov decision process (MDP) that can be solved by the DQN. For constructing deep neural networks to approximate the DQN action-value function that maps the timed S3PR states to scheduling rewards, we innovatively employ a graph convolutional network (GCN) as the timed S3PR state approximator by proposing a novel graph convolution layer called a Petri-net convolution (PNC) layer. The PNC layer uses the input and output matrices of the timed S3PR to compute the propagation of features from places to transitions and from transitions to places, thereby reducing the number of parameters to be trained and ensuring robust convergence of the learning process. Experimental results verify that the proposed DQN with a PNC network can provide better solutions for dynamic scheduling problems in terms of manufacturing performance, computational efficiency, and adaptability compared with heuristic methods and a DQN with basic multilayer perceptrons.

Mixed integer programming models for concurrent configuration design and scheduling in a reconfigurable manufacturing system

A reconfigurable manufacturing system can evolve its configuration to offer exactly the capacity and functionality needed for every demand period. For the reconfigurable manufacturing system with multi-part flow-line configuration simultaneously producing multiple parts within the same family, the production cost and the delivery time are closely related to its configuration and corresponding scheduling for certain demand period. Although studies on multi-part flow-line configuration design are abundant, studies on concurrent optimization of configuration design and scheduling for reconfigurable manufacturing system are scarce. First, a generic mixed integer nonlinear programming model for concurrent configuration design and scheduling is established to relax the limitation of the existing model, and then a mixed integer linear programming model is derived. The decisions of the two generalized models are to decide the amount of stations, the amount of identical machines and machines’ configuration for every station, and assign parts to machines along the multi-part flow line together with sequencing assigned parts for each machine. Based on the mixed integer linear programming model, an exact ε-constraint method is developed to obtain the Pareto optimal solutions with tradeoffs between cost and tardiness. The validation of two models and the ε-constraint method is verified against two cases adapted from the literature.

A Framework for Adaptive Scheduling in Cellular Manufacturing Systems

The Fourth Industrial Revolution has made efficient and adaptive manufacturing a prominent research topic. The necessity for personalized products leads to complex production lines. Due to the increased complexity, planning and scheduling of manufacturing processes seeks to identify near-optimum solutions to ensure fast and precise decision making. This research paper aims to contribute to the field of adaptive scheduling by proposing an algorithm that enables near real-time cooperation among machines, workforce and the production manager. A framework for self-adaptive scheduling in a real-world manufacturing case deriving from an SME that produces solar panels will be presented.

Optimal Scheduling of Flexible Manufacturing System Using Improved Lion-Based Hybrid Machine Learning Approach

Dispatching rules are generally useful for scheduling jobs in flexible manufacturing systems (FMSs). However, the appropriateness of these rules relies heavily on the condition of the system; thus, there is no single rule that always outperforms others. In this state of affairs, diverse machine-learning technology offers an effective approach for dynamic scheduling, as it allows managers to identify the most suitable rule at each moment. Nonetheless, various machine-learning algorithms may provide diverse recommendations. The main objective of this study is to implement FMS scheduling using intelligent hybrid learning algorithms with metaheuristic improvements. The developed model involves three phases: feature extraction, optimal weighted feature extraction, and prediction. After the benchmark datasets for the FMS are gathered, feature extraction is performed using t-distributed stochastic neighbor embedding, linear discriminant analysis, linear square regression, and higher-order statistical features. Further, an optimal weighted feature extraction method is developed to select the optimal features with less correlation using the improved lion algorithm (LA), which is called the modified nomadic-based LA (MN-LA). Finally, the optimally selected weighted features are subjected to a hybrid learning algorithm with the integration of a fuzzy classifier and a deep belief network (DBN). For improving the prediction model, the membership function of the fuzzy classifier is optimized using the proposed MN-LA. Moreover, the activation function and the number of hidden neurons in the DBN are optimized using the MN-LA. The main objective of the optimized hybrid classifier is to enhance prediction accuracy. The experimental results indicate the effectiveness of the proposed heuristic-based scheduling method for FMSs.

Robust scheduling of flexible manufacturing systems with unreliable operations and resources

This paper addresses the topic of robust scheduling for flexible manufacturing systems (FMS) with operation interruptions and unreliable resources. The proposed approach uses timed Petri nets as a model of the FMS in uncertain environments. This model includes controllable and uncontrollable transitions. The unexpected firings of the uncontrollable transitions represent operation and resource failures and the risk to deviate from the scheduled trajectories. This paper proposes an anytime graph search algorithm with a new objective function that combines performance and risk. In addition, the graph search incorporates a new filtering mechanism that learns from previous runs and two node expansion strategies. The performance of the proposed algorithm is compared with that of another already existing algorithm. A running example and a case study from a real company illustrate the efficiency of the proposed scheduling approach.

A Knowledge-Based Cuckoo Search Algorithm to Schedule a Flexible Job Shop With Sequencing Flexibility

Scheduling of complex manufacturing systems entails complicated constraints such as the mating operational one. Focusing on the real settings, this article considers an extended version of a flexible job shop problem that allows the precedence between the operations to be given by an arbitrary directed acyclic graph instead of a linear order. In order to obtain its reliable and high-performance schedule in a reasonable time, this article contributes a knowledge-based cuckoo search algorithm (KCSA) to the scheduling field. The proposed knowledge base is initially trained off-line on models before operations based on reinforcement learning and hybrid heuristics to store scheduling information and appropriate parameters. In its off-line training phase, the algorithm SARSA is used, for the first time, to build a self-adaptive parameter control scheme of the CS algorithm. In each iteration, the proposed knowledge base selects suitable parameters to ensure the desired diversification and intensification of population. It is then used to generate new solutions by probability sampling in a designed mutation phase. Moreover, it is updated via feedback information from a search process. Its influence on KCSA’s performance is investigated and the time complexity of the KCSA is analyzed. The KCSA is validated with the benchmark and randomly generated cases. Various simulation experiments and comparisons between it and several popular methods are performed to validate its effectiveness. Note to Practitioners —Complex manufacturing scheduling problems are usually solved via intelligent optimization algorithms. However, most of them are parameter-sensitive, and thus selecting their proper parameters is highly challenging. On the other hand, it is difficult to ensure their robustness since they heavily rely on some random mechanisms. In order to deal with the above obstacles, we design a knowledge-based intelligent optimization algorithm. In the proposed algorithm, a reinforcement learning algorithm is proposed to self-adjust its parameters to tackle the parameter selection issue. Two probability matrices for machine allocation and operation sequencing are built via hybrid heuristics as a guide for searching a new and efficient assignment scheme. To further improve the performance of our algorithm, a feedback control framework is constructed to ensure the desired state of population. As a result, our algorithm can obtain a high-quality schedule in a reasonable time to fulfill a real-time scheduling purpose. In addition, it possesses high robustness via the proposed feedback control technique. Simulation results show that the knowledge-based cuckoo search algorithm (KCSA) outperforms well some existing algorithms. Hence, it can be readily applied to real manufacturing facility scheduling problems.

A multi-agent architecture for scheduling in platform-based smart manufacturing systems

During the past years, a number of smart manufacturing concepts have been proposed, such as cloud manufacturing, Industry 4.0, and Industrial Internet. One of their common aims is to optimize the collaborative resource configuration across enterprises by establishing platforms that aggregate distributed resources. In all of these concepts, a complete manufacturing system consists of distributed physical manufacturing systems and a platform containing the virtual manufacturing systems mapped from the physical ones. We call such manufacturing systems platform-based smart manufacturing systems (PSMSs). A PSMS can therefore be regarded as a huge cyber-physical system with the cyber part being the platform and the physical part being the corresponding physical manufacturing system. A significant issue for a PSMS is how to optimally schedule the aggregated resources. Multi-agent technology provides an effective approach for solving this issue. In this paper we propose a multi-agent architecture for scheduling in PSMSs, which consists of a platform-level scheduling multi-agent system (MAS) and an enterprise-level scheduling MAS. Procedures, characteristics, and requirements of scheduling in PSMSs are presented. A model for scheduling in a PSMS based on the architecture is proposed. A case study is conducted to demonstrate the effectiveness of the proposed architecture and model.