Evolving CBR and data segmentation by SOM for flow time prediction in semiconductor manufacturing factory

Flow time of semiconductor manufacturing factory is highly related to the shop floor status; however, the processes are highly complicated and involve more than 100 production steps. Therefore, a simulation model with the production process of a real wafer fab located in Hsin-Chu Science-based Park of Taiwan is built for further studying of the relationship between the flow time and the various input variables. In this research, a hybrid approach by combining Self-Organizing Map (SOM) and Case-Based Reasoning (CBR) for flow time prediction in semiconductor manufacturing factory is developed. And Genetic Algorithm (GA) is applied to fine-tune the weights of features in the CBR model. The flow time and related shop floor status are collected and fed into the SOM for clustering. Then, a corresponding SGA-CBR method is selected and applied for flow time prediction. Finally, using the simulated data, the effectiveness of the proposed method (SGA-CBR) is shown by comparing with other approaches.     

The design of an automated material handling system for a job shop

This paper describes the application of techniques developed for the management and control of Flexible Manufacturing Systems (FMS's) to a traditional job shop manufacturing sheet metal parts on typical sheet fabricating machines. The problem was the design and evaluation of an automated material handling subsystem (MHS) to support the shop. The MHS controlled the shop loading and floor control by its choice of orders to transport and enter the shop. Detailed simulation models were used to simulate the existing shop, the shop with a MHS, and several algorithms for loading and sequencing orders through the shop. This paper describes the shop, the models developed, the algorithms tested, and the simulation results. A highlight of the study was the ease and speed with which multiple models were developed using an interactive simulation generator, CAPS. Of particular interest is the procedure for shop scheduling, downstream pull, which uses heavily loaded machines to provide sequencing alternatives with the objective of “pulling” work through them to reach more lightly loaded machines. Shop output was estimated to increase from 26 orders per day to 45, a 73% increase, and turnaround time to decrease from ten days to two days.     

The development of a hybrid hierarchical/heterarchical shop floor control system applying bidding method in job dispatching

Advancement in computer technology has brought shop floor control into a new era. Hierarchical and heterarchical control structures are two of the most popular control frameworks applied by industries and academics. However, the loading on the central controller may be very heavy in a hierarchical control structure, as the number of equipment in the shop floor increases. On the other hand, in a heterarchical configuration, the inter-cell communication will be very busy making control strategies to achieve. In this research, a hybrid hierarchical/heterarchical structure is proposed. This approach reduces the loading on the shop floor controller by empowering each cell controller to determine the production schedules of its corresponding equipment. However, the shop floor controller still is responsible for the global conditions of the shop such as deciding the job finishing time and balancing the utilization rate of equipment. A bidding method based on the required production costs is also proposed and implemented under the developed control structure.     

Real-time scheduling for hybrid flowshop in ubiquitous manufacturing environment

This paper discusses the implementation of RFID technologies, which enable the shop floor visibility and reduce uncertainties in the real-time scheduling for hybrid flowshop (HFS) production. In the real-time HFS environment, the arriving of new jobs is dynamic, while the processes in work stages are not continuous. The decision makers in shop floor level and stage level have different objectives. Therefore, classical off-line HFS scheduling approaches cannot be used under these situations. In this research, two major measures are taken to deal with these specific real-time features. Firstly, a ubiquitous manufacturing (UM) environment is created by deploying advanced wireless devices into value-adding points for the collection and synchronization of real-time shop floor data. Secondly, a multi-period hierarchical scheduling (MPHS) mechanism is developed to divide the planning time horizon into multiple shorter periods. The shop floor manager and stage managers can hierarchically make decisions for their own objectives. Finally, the proposed MPHS mechanism is illustrated by a numerical case study.     

Simulation-based planning and control: From shop floor to top floor

This paper illustrates how simulation-based shop-floor planning and control can be extended to enterprise-level activities (top floor). First, the general planning and control concept are discussed, followed by an overview of simulation-based shop-floor planning and control. Analogies between the shop floor and top floor are discussed in terms of the components required to construct simulation-based planning and control systems. Analogies are developed for resource models, coordination models, physical entities, and simulation models. Differences between the shop floor and top floor are also discussed in order to identify new challenges faced for top-floor planning and control. A major difference between the top floor and the shop floor is the way a simulation model is constructed for use in planning, depending on whether time synchronization among member simulations becomes an issue or not. Another difference is in the distributed communication/computing platform. This work uses a distributed computing platform using Web services technology to integrate heterogeneous simulations and systems in a distributed top-floor control environment. The research results reveal that simulation-based planning and control is extensible to the top-floor environment’s evolving new research challenges.     

Conceptual design of a shop floor control system from IDEF0

An approach for designing a shop floor control system is proposed in this paper. In this approach, the IDEF₀ model, Biemans' model, and the Phillips' CAM model are three major technologies used for determining the control request, control classification, and the control allocation. This approach has further been applied to propose a real shop floor control system in Taiwan, whose discussion is presented at the end.     

求解考虑机器调整时间的并行机分批优化调度问题

基于目前车间调度问题是以单个或整批进行生产加工的并行机调度模型已不再符合实际工况下的车间生产。提出以最小化最大完工时间为优化目标,对遗传差分进化混合算法,灰狼差分进化混合算法进行了比较。为提高加工工件进行分批及分批之后子批的分配与排序效率,该问题是对不同规模的经典并行机调度问题进行求解并展示两种算法的求解,证明了灰狼差分进化混合算法在寻优性能上优于遗传差分进化混合算法,不仅具有更好的解的稳定性,而且具有更高的寻优精度。     

Ant colony optimization approach to a fuzzy goal programming model for a machine tool selection and operation allocation problem in an FMS

Due to the global competition in manufacturing environment, firms are forced to consider increasing the quality and responsiveness to customization, while decreasing costs. The evolution of flexible manufacturing systems (FMSs) offers great potential for increasing flexibility and changing the basis of competition by ensuring both cost effective and customized manufacturing at the same time. Some of the important planning problems that need realistic modelling and quicker solution especially in automated manufacturing systems have assumed greater significance in the recent past. The language used by the industrial workers is fuzzy in nature, which results in failure of the models considering deterministic situations. The situation in the real life shop floor demands to adopt fuzzy-based multi-objective goals to express the target set by the management. This paper presents a fuzzy goal programming approach to model the machine tool selection and operation allocation problem of FMS. An ant colony optimization (ACO)-based approach is applied to optimize the model and the results of the computational experiments are reported.     

PLC定时计数功能在住宅小区自动控制中的应用研究

以某住宅小区的自动控制为例,利用三菱PLC的定时计数功能组成电子时钟,按预先设定的时段定时启动各控制程序,完成相应的控制功能,取得了很好的效果,对其它自动控制也有一定的借鉴作用。     

A closed-loop control architecture for CAM accounting for shop floor uncertainties

The primary focus of this paper is to demonstrate an enhanced closed-loop computer-aided manufacturing (CAM) architecture using a Petri net based analysis mechanism, which incorporates shop floor uncertainties to generate process plans and alternative processing sequences under dynamic conditions such as machine failures. Modern CAM software provide efficient, error-free, and user-friendly interfaces to develop processing sequences for manufacturing tasks. However, their effective scope is still limited from the post-process planning stage to the pre-machining stage. These types of software fail to incorporate the variability encountered in an actual shop floor to develop effective process plans. This is caused due to CAM software’s limitations in input handling mechanisms, limited simulation functionality, and inability to handle complex process variability from the shop floor. In this paper, CAM functionality is augmented with Petri net models to capture alternative process plans and shop floor variability arising due to equipment failures, repairs, and setups. The output from the analysis of the Petri net model provides a better understanding of the process characteristics and provides information to the CAM system to adapt to shop floor variability. The resulting information from the Petri net model allows the CAM system to compare alternative process plans while taking shop floor information into account. This mechanism can result in better decisions, reduced processing times, and better overall resource utilization, and provide alternative process plans based on current resource information.     

基于Petri网的PLC系统设计分析与远程监控

Petri网在离散事件的动态仿真中有着广泛的应用,而如何将Petri网应用到离散制造系统中具有重大的价值.以离散制造业中最常见的搬运机械手模型为基础,提出了它的Petri网模型设计方法,并且利用科学的转换方法将Petri网转化为梯形图,应用到在制造业中广泛使用的PLC系统中.利用这种设计方法,不仅能够轻松地将Petri网应用到实践当中,还能够对模型的死锁、生产周期和不变量进行定量的分析,给制造系统进行管理控制、故障诊断和现场监控带来方便.     

On scheduling in a GT environment

Traditional research in scheduling has concentrated on developing optimal and optional tending heuristic algorithms for n x m job shop and flow shop problems. Relatively no work has been conducted in the area of scheduling within a GT cell. This paper examines the GT cell environment from a scheduling perspective in relation to a job and flow shop, and presents a modified approach to scheduling within a GT cell that implicitly takes advantage of common setups and part family coding structures.     

Pheromone based emergent shop floor control system for flexible flow shops

New approaches for the development of shop floor control systems are needed to introduce better response to unanticipated disturbance situations and a better handling of ‘reconfiguration’ in production environments. In this paper, a development approach for agent-based shop floor control systems is presented, that uses co-ordination concepts as observable in insect colonies. In this scheme, agents operate within an information distribution environment, where information is made available in the form of ‘artificial pheromones’. Pheromone concepts from insect colonies and their mapping into a control system architecture are presented, a test bed implementation is discussed.     

From simple digital twin to complex digital twin part II: Multi-scenario applications of digital twin shop floor

The shop floor has always been an important application object for the digital twin. It is well known that production, process, and product are the core business of the shop floor. Therefore, the digital twin shop floor covers multi-dimensional information and multi-scale application scenarios. In this paper, the digital twin shop floor is constructed according to the modeling method of the complex digital twin proposed in Part I. The digital twin shop floor is firstly divided into several simple digital twins that focus on scenarios of different scales. Two simple application scenarios are constructed, including tool wear prediction and spindle temperature prediction. Main functions in different application scenarios, such as data acquisition, data processing, and data visualization, are implemented and encapsulated as components to construct simple digital twins. Secondly, ontology models, knowledge graphs, and message queues are used to assemble these simple digital twins into the complex digital twin shop floor. And two complex application scenarios are constructed, including machining geometry simulation considering spindle temperature and production scheduling considering tool wear. The implementation of the complex digital twin shop floor demonstrates the feasibility of the proposed modeling method.     

Dynamic planning model for determining cutting parameters using neural networks in feature-based process planning

Although feature-based computer-aided process planning plays a vital role in automating and integrating design and manufacturing for efficient production, its off-line properties prohibit the shop floor controllers from rapidly coping with unexpected production errors. The objective of the paper is to suggest a neural network-based dynamic planning model, by which the shop floor controllers determine cutting parameters in real-time based on shop floor status. At off-line is the dynamic planning model constructed as a neural network form, and then embedded into each removal feature. The dynamic planning model will be executed by the shop floor controllers to determine the cutting parameters. A prototype system is constructed to validate whether the dynamic planning model is capable of determining dynamically and efficiently the cutting parameters for a particular set of shop operating factors. Owing to the dynamic planning model, the shop floor controller will increase flexibility and robustness by rapidly and adaptively determining the cutting parameters in unexpected errors occurring.     

An AIS-based hybrid algorithm with PDRs for multi-objective dynamic online job shop scheduling problem

The dynamic online job shop scheduling problem (JSSP) is formulated based on the classical combinatorial optimization problem – JSSP with the assumption that new jobs continuously arrive at the job shop in a stochastic manner with the existence of unpredictable disturbances during the scheduling process. This problem is hard to solve due to its inherent uncertainty and complexity. This paper models this class of problem as a multi-objective problem and solves it by hybridizing the artificial intelligence method of artificial immune systems (AIS) and priority dispatching rules (PDRs). The immune network theory of AIS is applied to establish the idiotypic network model for priority dispatching rules to dynamically control the dispatching rule selection process for each operation under the dynamic environment. Based on the defined job shop situations, the dispatching rules that perform best under specific environment conditions are selected as antibodies, which are the key elements to construct the idiotypic network. Experiments are designed to demonstrate the efficiency and competitiveness of this model.     

Pheromone-based coordination for manufacturing system control

A pheromone-based coordination approach, which comes from the collective behavior of ant colonies for food foraging, is applied to control manufacturing system in this paper, aiming at handling dynamic changes and disturbances. The pheromone quantum of manufacturing cell is calculated inversely proportional to the cost, which can guarantee a minimal cost to process the orders. This approach has the capacity for optimization model to automatically find efficient routing paths for processing orders and to reduce communication overhead which exists in contract net protocol in shop floor control system. A prototype system is developed, and experiments confirm that pheromone-based coordination approach has excellent control performance and adaptability to disturbances in shop floor.     

A simulation study comparing GT vs. job shop manufacturing systems

Two discrete simulation models were developed, one representing a typical job shop (JS) manufacturing system (process layout) and the other representing group technology (GT) manufacturing system (cellular layout). Hypothetical data generated by FORTRAN program and MICRO-CRAFT layout package were used to validate the two models. Simulations were performed using SLAM II. The two layouts were compared with one another under controlled experimental conditions using various combinations of four operating variables: batch size, demand rate, ratio of setup time to process time, and transporter speed. Six indicators of systems performance were used in the study. These were: average time in system, products completed, jobs in process, average machine utilization, average queue length, and average waiting time.

The study found that for environment as set for the models, GT system outperformed JS system as expected. However, using large batch size (more than 75) in Experiment 1 (changing batch size) allowed JS system to perform as well as GT system. Certain specific changes in batch size also allowed to establish optimum results in both systems. No individual change of demand rate (Experiment 2), ratio of setup time to process time (Experiment 3), or transporter speed (Experiment 4) allowed JS system to outperform GT system.     

Industry case studies in the use of immersive virtual assembly

In this paper, we report on two engineering case studies that have been conducted as part of a Virtual Assembly Technology Consortium. The objectives of the case studies were to determine if immersive virtual assembly capabilities allow industry assembly situations to be modelled and studied realistically, and to demonstrate the downstream value of the virtual assembly capabilities in areas such as ergonomics, assembly installation, process planning, installation, and serviceability. What is of special significance is that instead of modelling simplified problems or perceived representative situations, the case studies were constructed from actual assembly floor projects and situations encountered at industry member sites and with considerable participation from industry engineers and manufacturing shop floor personnel. Based on the success of the case studies, the consortium members inferred that virtual assembly methods are poised to move out of the realm of special projects and test scenarios to deployment in the actual design and manufacturing cycle. However, in order to be truly accepted in industry, there are still issues to be addressed in terms of ease of use, portability of the applications, and preparation of the models for the evaluations. Thus, the case studies added a new dimension to the exploration and understanding of how this new technology could be of practical value in industry.     

The order fulfillment planning problem considering multi-site order allocation and single-site shop floor scheduling

The order fulfillment planning process in the thin film transistor–liquid crystal display panel industry is analyzed in this study. A two-phase order fulfillment planning structure is proposed, including the multi-site order allocation among module factories and single-site shop floor scheduling in each factory. In the first phase, the order allocation problem is solved using a mathematical programming model considering practical characteristics, including product structures, customer preferences, alternative bill-of-material, and production constraints. In the second phase, a constraint-based simulation scheduling algorithm is developed to address the scheduling problem in each module factory for determining the ideal order release time. Since production planning and scheduling are dealt with different time scales, the major challenge for the integration lies in the large problem size of the optimization model and becomes intractable. Most of the time bucket-based planning methods in the past literature simplify their scheduling models, but in this paper the detailed shop floor operations and processing behaviors are considered, such as changeover time, processing sequence of orders, and machine characteristics. Finally, a practical case in Taiwan will be employed to testify the feasibility of the proposed order fulfillment planning process; meanwhile, through the analysis of experiments, the adaptability and comparison of different planning approaches in an environment of various market demands are discussed.