Situation dependent decision selector for production control in testing and rework cell

This paper presents a production control approach for the testing and rework cell in a dynamic and stochastic computer integrated manufacturing (CIM) system. The proposed situation dependent decision selector (SDDS) observes the status of the system and jobs at every decision point, and makes its three different decisions on dispatching, pre-emption, and dispatching within pre-emption in real time. This paper also develops a new feature selection algorithm for the SDDS. Experimental results show that the SDDS is better than other static control rules and another dynamic control approach with respect to the number of tardy jobs.     

Architecture of a dynamic production controller in CIM enterprise environments

This paper deals with a production control problem for the testing and rework cell in a dynamic and stochastic computer integrated manufacturing (CIM) system. A dynamic controller is designed to handle three different production control decisions; dispatching, pre-emption, and dispatching within pre-emption. The model of this system also helps users deal with a large number of input features in real situations. This paper describes a technical architecture of dynamic production control and design issues of a dynamic production controller. It explains the overall development processes of the system and functional diagrams in order to give readers a design guideline for the dynamic production controller.     

A deep learning approach for the dynamic dispatching of unreliable machines in re-entrant production systems

This research combines deep neural network (DNN) and Markov decision processes (MDP) for the dynamic dispatching of re-entrant production systems. In re-entrant production systems, jobs enter the same workstation multiple times and dynamic dispatching oftentimes aims to dynamically assign different priorities to various job groups to minimise weighted cycle time or maximise throughput. MDP is an effective tool for dynamic production control, but it suffers from two major challenges in dynamic control problems. First, the curse of dimensionality limits the computational performance of solving large MDP problems. Second, a different model should be built and solved after system configuration is changed. DNN is used to overcome both challenges by learning directly from optimal dispatching policies generated by MDP. Results suggest that a properly trained DNN model can instantly generate near-optimal dynamic control policies for large problems. The quality of the DNN solution is compared with the optimal dynamic control policies through the standard K-fold cross-validation test and discrete event simulation. On average, the performance of the DNN policy is within 2% of optimal in both tests. The proposed artificial intelligence algorithm illustrates the potential of machine learning methods in manufacturing applications.     

Aligning workload control theory and practice: lot splitting and operation overlapping issues

This paper addresses the problem of lot splitting in the context of workload control (WLC). Past studies on WLC assumed that jobs released to the shop floor proceed through the different stages of processing without being split. However, in practice, large jobs are often split into smaller transfer sublots so that they can move more quickly and independently through the production process and allow operations overlapping relating to the same job. This paper assesses the performance of different lot splitting policies for job release and dispatching strategies under lot splitting. A new dispatching rule was designed to specifically take advantage of lot splitting and operations overlapping in the context of WLC. Discrete-event simulation is used to assess system performance in relation to the ability to provide shorter delivery times and on time deliveries. Results highlight the importance of releasing the sublots of the same job together and demonstrate that combining an effective lot splitting policy with an appropriate dispatching rule can enhance the performance of production systems.     

Coordinating batching decisions in manufacturing networks

Family-based dispatching heuristics aim for improving job flow times by reducing time spent on set-ups. They realise set-up efficiencies by batching similar types of jobs. By their intuitiveness and the simplicity of their decision logic, they may contribute to an easy to implement and viable strategy in many practical settings. Similar to common dispatching rules most existing family-based dispatching heuristics are myopic, i.e. their decision scope is restricted to a single manufacturing stage. Hence, they neglect opportunities for improving shop performance by coordinating batching decisions with other manufacturing stages. Case examples from industry underpin the need for exploring these opportunities. We do so by studying a simple two-stage flow shop, entailing a serial and a batch stage. To facilitate shop coordination we propose extensions to existing family-based dispatching heuristics. Extended heuristics seek to further increase set-up efficiencies by allowing for upstream job re-sequencing, and pro-active set-ups, i.e. set-ups that may be initiated prior to the arrival of a job. Outcomes of an extensive simulation study indicate significant performance gains for extended heuristics vs. existing heuristics. Performance gains are largest for moderate and high set-up to run-time ratios.     

An investigation of cooperative dispatching for minimising mean flowtime in a finite-buffer-capacity dynamic flowshop

Scheduling in a dynamic flowshop that receives jobs at random and unforeseen points in time has traditionally been done by using dispatching rules. This study compares the performances of leading dispatching rules with a cooperative dispatching approach, for the objective of minimising mean flowtime in a flowshop, in which the buffers that hold in-process jobs between machines have finite capacities. Cooperative dispatching employs a consultative and consensus-seeking methodology for deciding which job to dispatch next on a machine. Computational experiments using randomly generated test problems for three different utilisation (congestion) levels are carried out for 5- and 10-machine flowshops, under a wide range of buffer capacities. The results highlight the sensitivity of some of the popular dispatching rules to buffer size. In contrast, cooperative dispatching emerges as a robust method that performs consistently well across the range of buffer sizes and machine utilisations tested. The reductions in mean flowtime obtained by cooperative dispatching, in comparison to the other dispatching rules, are particularly large in flowshops that operate with very tight buffer capacities and elevated levels of congestion     

Development of a hybrid negotiation scheme for multi-agent manufacturing systems

A hybrid inter-agent negotiation mechanism based on currency and a pre-emption control scheme is proposed to improve the performance of multi-agent manufacturing systems. The multi-agent system considered consists mainly of four types of agents: machine, clone, part and mediator. The machine agent controls the scheduling and the execution of a task. The clone agent aims to maximize the utilization rate by attracting relevant work to the machine. The part agent communicates with the machine agent or clone agent to acquire necessary production resources in order to get the required processing done, and the mediator agent contains the status of the part that will be processed by the subcontracting machine agent. The primary objective is to design decentralized control protocols for discrete part manufacturing systems to enhance the efficiency of the system and to allocate dynamically the resources to critical jobs based on the dynamic search tree. This research incorporates both the currency and the pre-emption schemes within a common framework. Currency functions are used to help the agents meet their individual objectives, whereas the pre-emption scheme is used to expedite the processing of parts based on their due dates. A dynamic search algorithm for the best route selection of different operations based on the job completion time is also proposed and it is implemented on a small manufacturing unit.     

A methodology for planning and controlling workload in a job-shop: a four-way decision-making problem

There has been extensive research on workload and input–output control with the objective of improving manufacturing operations in job-shops. In this paper, a multiple decision-making scheme is proposed to plan and control operations in a general job-shop, and to improve delivery and workload related performance measures. The job-shop characteristics reinforce the need for designing a global system that controls both the jobs entering (order acceptance, due date setting and job release) and the work-in-process (dispatching), leading to an improvement of operational measures. Previous research has concentrated on scheduling a set of orders through the shop floor, according to some decision mechanism, in order to optimise some measure of performance (usually total lead time). This means that, since only a part of the decision-making system is being optimised, the resulting decision may be sub-optimal. In this paper it is shown that the performance of the different decision rules changes when they are considered simultaneously. Hence, a higher level approach, where the four decisions (order acceptance, due date setting, job release and dispatching) are considered at the same time, should be adopted to improve job-shop operational performance.     

API-based two-dimensional dispatching decision-making approach for semiconductor wafer fabrication with operation due date–related objectives

This paper presents an adjacent pairwise interchanges (API)-based two-dimensional dispatching decision-making approach for semiconductor wafer fabrication with operation due date-related objectives. Each time when a machine becomes idle, the proposed dispatcher chooses a target processing job from the competing jobs and assigns it a start time. Giving the operation due date information of each competing job, we formulate this dispatcher as the mean absolute deviation problem to keep the jobs finished around their operation due dates in a proactive way. Dominance properties of this problem are established using proof by APIs. Then, a heuristic comprised of job selection within candidate set, movement of job cluster and local search is designed to solve this problem more efficiently. Numerical experiments validate the efficiency of the proposed heuristic in a single-machine environment as well as in a simulated wafer fab abstracted from practice. In comparison with four most referenced due date-related dispatching rules, the simulation study reveals the benefits brought by the two-dimensional dispatching decision with different due date tightness taken into account.     

Multi-criteria optimisation-based dynamic scheduling for controlling FMS

This study deals with controlling flexible manufacturing systems (FMS) operating in volatile production environments. Most studies that address this issue use some sort of adaptive scheduling that enables the FMS to cope with the randomness and variability efficiently. The methods presented in the literature are usually based on heuristics and use simple dispatching rules. They do not consider changing the decision criteria dynamically as the system conditions change. In contrast to previous studies, the present study focuses on developing a control mechanism for dynamic scheduling that is based on incremental optimisation. This means that each time a scheduling decision is made, the local optimisation problem is solved such that the next jobs to be processed on machines are selected. The objective function (dominant decision criterion) for this optimisation problem is selected dynamically based on production order requirements, actual shop-floor status and system priorities. The proposed multi-criteria optimisation-based dynamic scheduling methodology was evaluated and compared with some known scheduling rules/policies. The results obtained demonstrate the superiority of the suggested methodology as well as its capability to cope with a multi-criteria environment.     

Corrective maintenance through dynamic work allocation and pre-emption: case study and application

This case study develops an innovative management and scheduling system for corrective maintenance of machines in a manufacturing facility. The study also involves a comparative evaluation of the proposed and the existing systems under a spectrum of operating conditions. A comprehensive simulation is used to evaluate system performances under a variety of settings which include reliability, service level, and cost consequences. The analysis is based on a full factorial experimental design. In summary, the developed self-regulating management system which involves dynamic work allocation and pre-emption is shown to yield higher machine availability and higher mechanic utilisation even with fewer mechanics. The study also finds that the new system is more streamlined, agile, and robust although it is subject to more-constrained machine reliability and mechanic service time environments. Further, a major reduction of current manpower can still achieve at least 95% machine availability, illustrating the cost effectiveness and efficacy of the developed system. This rule-based corrective maintenance system can be operated in uncertain environments on a real-time basis without additional reformatting costs and provides a competitive measure to deal with managerial issues such as low retention rate for skilled mechanics, highly uneven training levels and pay scales. The financial consequences and gains in strategic advantage with respect to the facility's operational structure are promising after implementation. Moreover, the system developed in this case study represents a meaningful starting point for a more vigorous theoretical research on the bucket brigade system to different functions in industrial and operations management.     

Real time selection of scheduling rules and knowledge extraction via dynamically controlled data mining

A new scheduling system for selecting dispatching rules in real time is developed by combining the techniques of simulation, data mining, and statistical process control charts. The proposed scheduling system extracts knowledge from data coming from the manufacturing environment by constructing a decision tree, and selects a dispatching rule from the tree for each scheduling period. In addition, the system utilises the process control charts to monitor the performance of the decision tree and dynamically updates this decision tree whenever the manufacturing conditions change. This gives the proposed system the ability to adapt itself to changes in the manufacturing environment and improve the quality of its decisions. We implement the proposed system on a job shop problem, with the objective of minimising average tardiness, to evaluate its performance. Simulation results indicate that the performance of the proposed system is considerably better than other simulation-based single-pass and multi-pass scheduling algorithms available in the literature. We also illustrate knowledge extraction by presenting a sample decision tree from our experiments.     

A hybrid dynamic pre-emptive and competitive neural-network approach in solving the multi-objective dispatching problem for TFT-LCD manufacturing

This research addresses a hybrid dynamic pre-emptive and competitive neural-network approach in solving the multi-objective dispatching problem. It optimises three performance criteria simultaneously, namely: cycle time, slack time, and throughput. A case study is adopted to illustrate the performance of applying the methodology. Thin film transistor-liquid crystal display (TFT-LCD) is a high-technology industry, with a growing market. The manufacturing process is complex. It involves multi-products, sequence-dependent set-ups, random breakdowns, and multiple-objectives, with bias-weighted optimisation problems. To determine appropriate dispatching strategies, under various system conditions, is a non-trivial challenge to control the complex systems. There has been little research on these problems aimed at solving them simultaneously. This paper presents an event-triggered dynamic dispatching system that combines artificial intelligence methods to archive optimum dispatching strategies under diverse shop-floor conditions. Results show this system to be superior to previous researches.     

Setup-oriented dispatching rules – a survey

This paper provides a survey of dispatching rules that explicitly take into account setup times in their decision making. Rules are classified into the categories of purely setup-oriented, composite and family-based rules, and the most promising rules from the three categories are identified from the literature. These rules are then compared empirically on various job shop problems with sequence-dependent setup times for their performance regarding mean setup time, mean flow time, mean tardiness and proportion of tardy jobs. The setup times are modelled using setup time matrices, and five different types of matrices are applied to assess the influence of this factor on the relative performance of a setup-oriented dispatching rule. Experimental results indicate that the choice of the best rule is often dependent on the setup time matrix structure. While good family-based rules exist for reducing the mean setup time and mean flow time, they are clearly outperformed by effective composite rules for due date-related criteria. Moreover, the better rules all seem to rely on queue information rather than only job attributes.     

Simulation-based optimisation of logistics distribution system for an assembly line with path constraints

The design of logistics distribution system for an assembly line with given layout is usually constrained by various factors such as the vehicles for the distribution of assembly components and the paths on the shop floor for the vehicle movement. Since design optimisation of these production systems with such constraints is hard to solve by mathematic approaches, simulation-based approach is mostly used. In this paper, simulation-based analysis and optimisation of a logistic distribution system is performed combined with a heuristic algorithm to obtain a solution that most suits the practical requirements of an assembly line. The proposed simulation-based method builds a model that satisfied various complicated real-world problems at the work site of an investigated factory. Meanwhile, a dynamic dispatching method for vehicle movement is also presented which can adjust the control strategy and decision parameters dynamically in the process of simulation and contributes significantly to form a more fruitful design scheme. Experimental results show that proposed method performs better than the analytical model-based approach and some other methods, and the dynamic vehicle dispatching method presented performs better than some existing common strategies used in literature.     

Bottleneck-based heuristics to minimize tardy jobs in a flexible flow line with unrelated parallel machines

This paper develops new bottleneck-based heuristics with machine selection rules to solve the flexible flow line problem with unrelated parallel machines in each stage and a bottleneck stage in the flow line. The objective is to minimize the number of tardy jobs in the problem. The heuristics consist of three steps: (1) identifying the bottleneck stage; (2) scheduling jobs at the bottleneck stage and the upstream stages ahead of the bottleneck stage; (3) using dispatching rules to schedule jobs at the downstream stages behind the bottleneck stage. A new approach is developed to find the arrival times of the jobs at the bottleneck stage, and two decision rules are developed to schedule the jobs on the bottleneck stage. This new approach neatly overcomes the difficulty of determining feasible arrival times of jobs at the bottleneck stage. In order to evaluate the performance of the proposed heuristics, six well-known dispatching rules are examined for comparison purposes. Six factors are used to design 729 production scenarios, and ten test problems are generated for each scenario. Computational results show that the proposed heuristics significantly outperform all the well-known dispatching rules. An analysis of the experimental factors is also performed and several interesting insights into the heuristics are discovered.     

Integer programming-based real-time dispatching (i-RTD) heuristic for wet-etch station at wafer fabrication

A real-time dispatching heuristic derived from an optimal binary integer programming model (i-RTD) to minimise the makespan of the automatic wet-etch station (AWS) scheduling problem with a chemical draining constraint is presented. Each chemical bath is drained and refilled every several days. This draining process, known as chemical dumping in the industry, occurs during the middle of a process and interrupts production. This chemical draining event creates a need for intelligent flow shop scheduling so that all baths are optimally utilised and so that jobs can skip a bath being drained and move to the next bath without interruption. The production losses from these interruptions can be minimised by optimally sequencing jobs before they enter the AWS. This paper first creates the binary integer programming (BIP) model to exactly describe the AWS scheduling problem, which later forms the basis for the i-RTD model to generate an efficient solution in real time. This unique heuristic accounts for the chemical draining constraint, time window constraint, zero wait constraint, and real-time constraint and represents a new dispatching paradigm of considerable practical interest.     

Solving multi-objective rescheduling problems in dynamic permutation flow shop environments with disruptions

In multi-objective optimisation problems, optimal decisions need to be made in the presence of trade-offs among conflicting objectives which may sometimes be expressed in different units of measure. This makes it difficult to reduce the problem to a single-objective optimisation. Furthermore, when disruptive changes emerge in manufacturing environments, such as the arrival of new jobs or machine breakdowns, the scheduling system should be adapted by responding quickly. In this paper, we propose a rescheduling architecture for solving the problem based on a predictive-reactive strategy and a new method to calculate the reactive schedule in each rescheduling period. Additionally, we developed a methodology that allows the use of multi-objective performance metrics to evaluate dispatching rules. These rules are applied at a benchmark specifically designed for this paper considering three objective functions: makespan, total weighted tardiness and stability. Three types of disruptions are also considered: arrivals of new jobs, machine breakdowns and variations in job processing times. Results showed that the RANDOM rule provides a better behaviour compared to other evaluated rules and a lower ratio of non-dominated solutions compared to ATC (apparent tardiness cost) and FIFO (first-in-first-out) rules. Moreover, the behaviour of the hypervolume metric depends on the problem dimensions.     

A decision-making perspective on hybrid manufacturing system control

This paper presents a novel scheme for hybrid control of a manufacturing system where jobs visit workstations following a bidding process. An automated guided vehicle system is selected for handling material between workstations. The proposed hybrid architecture is appropriate particularly for circumstance in which vehicles deliver materials across the factory, the local control authority is emphasized, or when a simple, straightforward, quick response control at the lower level is desired. Because control activities are inherent in making decisions, three types of decisions (i.e. induction, dispatching, and routing) are used to present the manufacturing system controls. Unambiguously to represent the controls and capture the system behaviour, a mathematical formalism is applied. The discussed context includes the mathematical formalism of those control decisions, the system behaviour in terms of its component interactions, and a controller's internal passing messages. An illustrative example demonstrates the system control's effectiveness.     

Family based dispatching in manufacturing networks

Intrinsic to family based dispatching is the grouping of similar types of jobs in front of a machine for joint processing. Machine flow times may be improved in this way, as less time is spent on set-ups. Our observations in practice, however, suggest that family based dispatching may result in a bulky arrival pattern for successor manufacturing stages, thereby causing additional delay. So far, literature seems to neglect this effect. To explore this issue we develop queuing theoretical approximations of flow times for a simple two-stage shop. It appears that the optimal batch size for the shop is typically smaller than the optimal batch size for the batch machine. Furthermore, we propose extensions to existing dispatching rules by using information on successor stages. Existing and new extended rules are tested by an extensive simulation study. In line with the queuing theoretical analysis the outcomes indicate that exhaustive rules – assuming batch size to be equal to family queue length – are clearly outperformed by non-exhaustive rules – allowing for smaller batches. Moreover, results show that the inclusion of local information on successor stages in rule decision making improves shop flow times.