Neural network-based adaptive production control system for a flexible manufacturing cell under a random environment

A Neural Network (NN)-based Production Control System (PCS) for a Flexible Manufacturing Cell (FMC), operating in a highly random produce-to-order environment is presented. The proposed PCS chooses periodically, on the basis of the current state of the system, the most appropriate scheduling rule, out of several predetermined ones. The proposed PCS is based on multi-layer NNs, one for each competing scheduling rule, that predict the FMC's performance. The NNs are retrained periodically. The performance of the proposed NN-based PCS was tested by simulation of two different FMC configurations. The NN-based PCS has performed significantly better than a decision-tree-based PCS and a single-rule-based PCS.     

On-line scheduling in a multi-cell flexible manufacturing system

This paper deals with on-line scheduling in a multi-cell flexible manufacturing system, operating in a produce-to-order environment. A two level distributed production control system (DPCS) is developed and tested through a simulation study. The DPCS allows autonomous and simultaneous operation of each cell-controller, utilizing only local and short term information as well as simple heuristic rules. Simulation experiments show that the proposed DPCS achieves good results in throughput, tardiness of orders and WIP inventory level and that it is robust to machine and handling device failures.     

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.     

Fuzzy inference-based multiple criteria FMS scheduling

This paper proposes a fuzzy inference-based scheduling decision for flexible manufacturing systems (FMS) with multiple objectives. The objectives have different and dynamic preference levels. It is inferred that the changes in the production environment may be sensed by environmental variables. The detected changes are input in a fuzzy inference mechanism, which outputs the current preference levels of all objectives. A multiple criteria scheduling decision is then made, using the partitioned combination of the preference levels. An example of application is presented. Simulation results show very good performance for the proposed system.     

Adaptive scheduling and tool flow control in flexible job shops

The recent manufacturing environment is characterized as having diverse products due to mass customization, short production lead-time, and ever-changing customer demand. Today, the need for flexibility, quick responsiveness, and robustness to system uncertainties in production scheduling decisions has dramatically increased. In traditional job shops, tooling is usually assumed as a fixed resource. However, when a tooling resource is shared among different machines, a greater product variety, routing flexibility with a smaller tool inventory can be realized. Such a strategy is usually enabled by an automatic tool changing mechanism and tool delivery system to reduce the time for tooling set-up, hence it allows parts to be processed in small batches. In this paper, a dynamic scheduling problem under flexible tooling resource constraints is studied and presented. An integrated approach is proposed to allow two levels of hierarchical, dynamic decision making for job scheduling and tool flow control in flexible job shops. It decomposes the overall problem into a series of static sub-problems for each scheduling horizon, handles random disruptions by updating job ready time, completion time, and machine status on a rolling horizon basis, and considers the machine availability explicitly in generating schedules. The effectiveness of the proposed dynamic scheduling approach is tested in simulation studies under a flexible job shop environment, where parts have alternative routings. The study results show that the proposed scheduling approach significantly outperforms other dispatching heuristics, including cost over time (COVERT), apparent tardiness cost (ATC), and bottleneck dynamics (BD), on due-date related performance measures. It is also found that the performance difference between the proposed scheduling approach and other heuristics tend to become more significant when the number of machines is increased. The more operation steps a system has, the better the proposed method performs, relative to the other heuristics.     

Methods for activating the decision-making process that is used for controlling flexible manufacturing systems

This paper deals with controlling flexible manufacturing systems (FMS) operating in volatile production environments. Shnits et al. (Shnits, B., Rubinovitz, J., and Sinreich, D., 2004. Multicriteria dynamic scheduling methodology for controlling a flexible manufacturing system. International Journal of Production Research, 42 (17), 3457–3472.) and Shnits and Sinreich (Shnits, B. and Sinreich, D., 2006. Controlling flexible manufacturing systems based on a dynamic selection of an appropriate operational criteria and scheduling policy. International Journal of Flexible Manufacturing Systems, 18 (1), 1–27.) developed a multi-criteria dynamic scheduling mechanism for controlling an FMS that can cope with such environments. An important part of this mechanism functioning, which impinges directly on its performance, is the activation of its decision-making process. This study continues the research presented in the above-mentioned papers and proposes different triggering methods for activating the control system decision-making process. The operational conditions for each suggested triggering method were analysed and a comparative analysis between these methods was performed. It was revealed that the highly dynamic triggering method, which activates the decision-making process right before a resource becomes available, outperformed the triggering methods that use a predefined scheduling period.     

Performance-based dynamic scheduling model for flexible manufacturing systems

A performance-based dynamic scheduling model for random flexible manufacturing systems (FMSs) is presented. The model is built on the mathematical background of supervisory control theory of discrete event systems. The dynamic FMS scheduling is based on the optimization of desired performance measures. A control theory-based system representation is coupled with a goal programming-based multi-criteria dynamic scheduling algorithm. An effectiveness function, representing a performance index, is formulated to enumerate the possible outputs of future schedules. Short-term job scheduling and dispatching decisions are made based on the values obtained by optimizing the effectiveness function. Preventive actions are taken to reduce the difference between actual and desired target values. To analyse the real-time performance of the proposed model, a software environment that included various Visual Basic Application® modules, simulation package Arena®, and Microsoft Access® database was developed. The experimentation was conducted (a) to determine the optimum look-ahead horizons for the proposed model and (b) to compare the model with conventional scheduling decision rules. The results showed that the proposed model outperformed well-known priority rules for most of the common performance measures.     

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.     

Multicriteria dynamic scheduling methodology for controlling a flexible manufacturing system

A methodology is presented for the dynamic scheduling of flexible manufacturing systems (FMSs). A two-level control hierarchy is suggested. The higher level is used for determining a dominant decision criterion and relevant scheduling rules, based on an analysis of the actual shop status. The lower level uses simulation for determining the best scheduling policy to be selected. Simulation is used to evaluate different control options, and once a control decision is made, it is operated in real time to serve as the FMS controller. The suggested scheduling and control scheme is being developed, implemented and tested in a physical computer integrated manufacturing (CIM)/FMS environment at the CIM and Robotics Lab of the Faculty of Industrial Engineering and Management, Technion. This will serve as a test-bed to study the performance of the FMS under different scheduling rules and control options, and to recommend the best combination of control policies and parameters for specific system conditions and global production objectives.     

Loading and control policies for a flexible manufacturing system

An experimental investigation of operating strategies for a computer-controlled flexible manufacturing system is reported. The system is a real one, consisting of nine machines, an inspection station and a centralized queueing area—all interconnected by an automatic material-handling mechanism. The operating strategies considered involve policies for loading (allocating operations and tooling to machines) and real-time flow control. A detailed simulation was employed to test alternatives. The results are different from those of classical job shop scheduling studies, showing the dependence of system performance on the loading and control strategies chosen to operate this flexible manufacturing system. Loading and control methods are defined that significantly improve the system's production rate when compared to methods which were previously applied to the system. Finally, some conclusions are presented concerning the control of these automated systems.     

Using Functional Virtual Population as assistance to learn scheduling knowledge in dynamic manufacturing environments

When a scheduling environment is static and system attributes are deterministic, a manufacturing schedule can be obtained by applying analytical tools such as mathematical modelling technology, dynamic programming, the branch- and-bound method or other developed searching algorithms. Unfortunately, a scheduling environment is usually dynamic in a real manufacturing world. A production system may vary with time and require production managers to change schedule repeatedly. Therefore, the main aim here was to find a scheduling method that could reduce the need for rescheduling. An approach called Functional Virtual Population was proposed as assistance to learn robust scheduling knowledge for manufacturing systems under rationally changing environments. The used techniques include machine learning with artificial neural networks and IF-THEN scheduling rules. To illustrate the study in detail, a simulated flexible manufacturing system consisting of four machines, four parts, one automatic guided vehicle and eight buffers was built as the foundation for learning the concept. Also, Pythia software (a back-propagation-based neural networks) was employed as the learning tool in the learning procedure.     

Learning-based adaptive controller for dynamic manufacturing cells

Due to the complexity, uncertainty and dynamics in the modern manufacturing environment, a flexible and adaptive cell controller is essential to achieve the system production goals. The paper proposes a learning-based approach for the adaptive controller, which will receive feedback on current performance from the cell, and fine-tune its knowledge base by using a cerebellar model articulation controller (CMAC) network. To examine the proposed controller's performance in manufacturing cells, several experiments are conducted based on simulation. The results show that the controller performs well under multiple (conflicting) performance measures. Furthermore, it is shown that the controller with feedback can learn and adapt to changing environments. Most interestingly, the paper also demonstrates that the proposed controller can adapt to changing system objectives (desired performance measures).     

A rule-based,object-oriented framework for operating flexible manufacturing systems

A methodology for operating a flexible manufacturing system (FMS) is presented which consists of a decomposition of the problem, where heuristics, rules, and simulation are integrated to make the control decisions needed for random FMS's. Expert systems are used to make operating decisions, to choose heuristics, and make operating policies. Multiple passes of discrete-event simulation are employed to evaluate decision alternatives. Object-oriented programming provides a suitable environment to implement the integrative framework     

Experimental investigation of flexible manufacturing system scheduling decision rules

This paper reports the results of an experimental investigation of scheduling decision rules for a dedicated flexible manufacturing system. A simulation model of an existing flexible manufacturing system (FMS) comprised of 16 computer numerical controlled machines (CNC) was constructed using actual operation routings and machining times to evaluate the performance of various part loading and routing procedures. The results indicate that FMS performance is significantly affected by the choice of heuristic parts scheduling rules.     

A comparison of sequencing rules in static and dynamic hybrid flow systems

It is well known that efficient scheduling of jobs is essential for improving the economics of production in manufacturing organizations. As a result, extensive research has been conducted on scheduling, especially in job shop and flow shop settings. In contrast, little research has been done on hybrid flow systems, even though they are found in many industries, including beer processing, glass container production, pertroleum refining, plastic-coated cable production, and fertilizer production. Furthermore, the few studies that have dealt with hybrid systems have been limited by the assumptions made about their operating environments. Therefore, we conducted a study that extends the previous work on hybrid systems in two significant ways: (1) it included financially oriented scheduling rules and a new, related performance measure; and (2) the new rules were compared with the existing ones in a large simulation experiment under both static and dynamic (generally encountered in practice) hybrid flow shop environments. To date such comparisons have been made only under static environments. The results show that the relative performances of the scheduling rules differ as the assumptions regarding the operating environment are changed.     

Work sequencing in a manufacturing cell with limited labour constraints

This study focuses on the analysis of group scheduling heuristics in a dual-constrained, automated manufacturing cell, where labour utilization is limited to setups, tear-downs and loads/unloads. This scenario is realistic in today's automated manufacturing cells. The results indicate that policies for allocating labour to tasks have very little impact in such an environment. Furthermore, the performance of efficiency oriented, exhaustive, group scheduling heuristics deteriorated while the performance of the more complex, non-exhaustive heuristics improved. Thus, it is recommended that production managers use the simplest labour scheduling policy, and instead focus their efforts to activities such as job scheduling and production planning in such environments.     

Switching mode control strategy in manufacturing execution systems

The paper describes a semi-heterarchical control solution for mixed planning and scheduling, routing and job execution in flexible manufacturing systems based on the paradigms of holonic manufacturing and product-driven automation. The main feature of the control solution is the bidirectional switching of the operating mode (scheduling, routing) between centralised and decentralised in the presence of perturbations to ensure as long as possible both global optimisation and agility to changes in batch orders, while featuring robustness to disturbances in the production environment. At the theoretical level, the control solution is described in terms of generic structural and dynamic models. The implementation is done using a multi-agent Java Agent Development Framework framework.     

Production flow control in a cell with groups of identical machines

This paper addresses the production control problems created in the scheduling of a flexible manufacturing cell consisting of groups of identical machines. An actual manufacturing system, drawn from Sikorsky Aircraft Corporation, was used as a prototype for modeling the cell to study its optimal behavior. The model, stated in the form of optimal production flow control combines classical machine-part scheduling and scheduling of part fixtures. Such an extension induces three problem formulations comprising practical production issues important for many modern plants other than Sikorsky. These formulations are related to three production cases covering different degrees of fixture availability. With the aid of the maximum principle, a numerical scheduling method is constructed for all three problem formulations, and case studies are presented.     

Development of a real-time multi-objective scheduler for a semiconductor fabrication system

Semiconductor wafer fabrication involves possibly one of the most complex manufacturing processes ever used. This causes a number of decision problems. A successful system control strategy would assign appropriate decision rules for decision variables. Therefore, the goal of this study is to develop a scheduler for the selection of decision rules for decision variables in order to obtain the desired performance measures given by a user at the end of a certain production interval. In this proposed methodology, a system control strategy based on a simulation technique and a competitive neural network is suggested. A simulation experiment was conducted to collect the data containing the relationship between the change of decision rule set and current system status and the performance measures in the dynamic nature of semiconductor manufacturing fabrication. Then, a competitive neural network was applied to obtain the scheduling knowledge from the collected data. The results of the study indicate that applying this methodology to obtaining a control strategy is an effective method considering the complexity of semiconductor wafer fabrication systems.     

Experimental investigation of FMS machine and AGV scheduling rules against the mean flow-time criterion

Although a significant amount of research has been carried out in the scheduling of flexible manufacturing systems (FMSs), it has generally been focused on developing intelligent scheduling systems. Most of these systems use simple scheduling rules as a part of their decision process. While these scheduling rules have been investigated extensively for a job shop environment, there is little guidance in the literature as to their performance in an FMS environment. This paper attempts to investigate the performances of machine and AGV scheduling rules against the mean flow-time criterion. The scheduling rules are tested under a variety of experimental conditions by using an FMS simulation model.