An approach to support SMEs in manufacturing knowledge organization

Different kinds of technological data are available in manufacturing enterprises, concerning the resources available as well as the processes and the components needed for the production of specific products. These data usually are not stored in a centralized knowledge management system, thus one of the main problem of managers, especially in small enterprises, is to efficiently manage their data and reuse the knowledge deriving from previous products when a new product has to be produced. Starting form the analysis of the technological data available in manufacturing enterprises, we defined a formal model as set of matrices; their analysis allows the definition of a data model to structure the technological information. The model is at the basis of the proposed system, called manufacturing knowledge organization (MAKO) to support managers in structuring and reusing the technological knowledge available in their enterprise. A prototype of the MAKO system was implemented by using open-source software and its potentialities are shown in a case study.     

A bi-objective integrated procurement,production, and distribution problem of a multi-echelon supply chain network design: A new tuned MOEA

Efficient management of supply chain (SC) requires systematic considerations of miscellaneous issues in its comprehensive version. In this paper, a multi-periodic structure is developed for a supply chain network design (SCND) involving suppliers, factories, distribution centers (DCs), and retailers. The nature of the logistic decisions is tactical that encompasses procurement of raw materials from suppliers, production of finished product at factories, distribution of finished product to retailers via DCs, and the storage of raw materials and end product at factories and DCs. Besides, to make the structure more comprehensive, a flow-shop scheduling model in manufacturing part of the SC is integrated in order to obtain optimal delivery time of the product that consists of the makespan and the ship time of the product to DCs via factories. Moreover, to make the model more realistic, shortage in the form of backorder can occur in each period. The two objectives are minimizing the total SC costs as well as minimizing the average tardiness of product to DCs. The obtained model is a bi-objective mixed-integer non-linear programming (MINLP) model that is shown to belong to NP-Hard class of the optimization problems. Thus, a novel algorithm, called multi-objective biogeography based optimization (MOBBO) with tuned parameters is presented to find a near-optimum solution. As there is no benchmark available in the literature, the parameter-tuned multi-objective simulated annealing algorithm (MOSA) and the popular non-dominated sorting genetic algorithm (NSGA-II) are developed to validate the results obtained and to evaluate the performance of MOBBO using randomly generated test instances.     

RFID-enabled automation in support of factory integration

In recognition of the need for more effective and efficient factory system integration solutions, this paper improves the method proposed in our previous work, focusing on new mechanisms to bridge the gap between shop floor automation and factory information systems. As one of the mainstream identification technologies, radio frequency identification (RFID) technology makes possible all products uniquely identifiable across factories. By taking advantage of data capacity stored in an RFID tag, critical manufacturing information on a product can be locally stored with the product. RFID technology provides a means for a product to rapidly retrieve its needed information as it advances through shop floors. A corresponding framework to enable the instant delivery of pertinent data and information on a uniquely identifiable job/product at point-of-need across factories is proposed. A small-scale manufacturing application is presented to show how the proposed approach can be implemented on the shop floor.     

A heuristic algorithm for the distributed and flexible job-shop scheduling problem

The distributed manufacturing takes place in a multi-factory environment including several factories, which may be geographically distributed in different locations, or in a multi-cell environment including several independent manufacturing cells located in the same plant. Each factory/cell is capable of manufacturing a variety of product types. An important issue in dealing with the production in this decentralized manner is the scheduling of manufacturing operations of products (jobs) in the distributed manufacturing system. In this paper, we study the distributed and flexible job-shop scheduling problem (DFJSP) which involves the scheduling of jobs (products) in a distributed manufacturing environment, under the assumption that the shop floor of each factory/cell is configured as a flexible job shop. A fast heuristic algorithm based on a constructive procedure is developed to obtain good quality schedules very quickly. The algorithm is tested on benchmark instances from the literature in order to evaluate its performance. Computational results show that, despite its simplicity, the proposed heuristic is computationally efficient and promising for practical problems.     

Job-Shop Scheduling with Cost Indexes: A Probabilistic Model

A production complex for concurrently manufacturing several products is studied. Every product has its own technological route, which is a sequence of jobs implemented on particular machines. The duration of every job is a random variable distributed by a given law. A discrete production time is defined for every product. A schedule for starting the production of all products that minimizes the expectation of total expenses on penalties and storage, i.e., overhead charges, is determined. Such a schedule is random and is formed by choosing a product from a queue for a given period. A decision making model is designed for determining cost indexes.     

Practical application of product design method robust to disturbances

Contemporary designed and manufactured products should meet a wide range of client requirements and expectations regarding quality for conditions of use in a life cycle. Therefore, design methods, which enable an optimal choice of product parameters by decreasing product sensivity to occurrence of some disturbances in a manufacturing process, become more and more significant in industrial practice. Such a behavior probably allows getting the lowest quality loss while keeping planned production costs. In this article, a new method of products design robust to disturbances exsisting in a technological process is presented. The application of mathematical modeling and methods of parameter estimation enabled formulating an approach in which parameters and acceptable tolerance of manufacturing product parameters are calculated. The efficiency of the proposed method was verified in the machine‐building industry with the example of designing parameters of a brushless direct current motor. © 2010 Wiley Periodicals, Inc.     

汽车制造企业实施PLM系统的信息建模

针对汽车制造业的特点,对其实施PLM系统信息建模技术和方法进行了研究,提出了集成用户管理模型、产品分类模型、产品对象的版本模型、产品结构管理模型、产品配置管理模型和工作流程模型的综合信息建模技术,为汽车制造企业成功实施PLM系统提供科学、实用的方法论。     

汽车制造企业实施PLM系统的信息建模

针对汽车制造业的特点,对其实施PLM系统信息建模技术和方法进行了研究,提出了集成用户管理模型、产品分类模型、产品对象的版本模型、产品结构管理模型、产品配置管理模型和工作流程模型的综合信息建模技术,为汽车制造企业成功实施PLM系统提供科学、实用的方法论。     

Predictive model-based quality inspection using Machine Learning and Edge Cloud Computing

The supply of defect-free, high-quality products is an important success factor for the long-term competitiveness of manufacturing companies. Despite the increasing challenges of rising product variety and complexity and the necessity of economic manufacturing, a comprehensive and reliable quality inspection is often indispensable. In consequence, high inspection volumes turn inspection processes into manufacturing bottlenecks.In this contribution, we investigate a new integrated solution of predictive model-based quality inspection in industrial manufacturing by utilizing Machine Learning techniques and Edge Cloud Computing technology. In contrast to state-of-the-art contributions, we propose a holistic approach comprising the target-oriented data acquisition and processing, modelling and model deployment as well as the technological implementation in the existing IT plant infrastructure. A real industrial use case in SMT manufacturing is presented to underline the procedure and benefits of the proposed method. The results show that by employing the proposed method, inspection volumes can be reduced significantly and thus economic advantages can be generated.     

Automatic machine status prediction in the era of Industry 4.0: Case study of machines in a spring factory

Recent technological developments have fueled a shift toward the computerization and automation of factories; i.e., Industry 4.0. Unfortunately, many small- and medium-sized factories cannot afford the sensor-embedded machines, cloud system, or high-performance computers required for Industry 4.0. Furthermore, the simple production processes in smaller factories do not require the level of precision found in large factories. In this study, we explored the idea of using inexpensive add-on triaxial sensors for the monitoring of machinery. We developed a dimensionality reduction method with low computational overhead to extract key information from the collected data as well as a neural network to enable automatic analysis of the obtained data. Finally, we performed an experiment at an actual spring factory to demonstrate the validity of the proposed algorithm. The system outlined in this work is meant to bring Industry 4.0 implementations within grasp of small to medium sized factories, by eliminating the need for sensors-embedded machines and high-performance computers.     

手机零部件自动化测试质量控制系统

在国内制造业行业里,大部分还只是人工测试,不仅效率低下,还没有很好地将工控机网络与产品质量的控制这两者的概念有效地结合在一起。质量已经成为客户衡量产品以及选择供应商的一个最大的标杆,在日益激烈的竞争中,企业会把更多的重心放在产品的质量上,将努力做到出去的产品没有质量上问题。这个课题正是对此进行研究,针对手机壳制造型企业大批量生产装配过程中工人出现的漏测、误流等情况而开发的整套装配测试质量系统。该系统可以防止不良品或未检测产品流到下一工序以至于客户端,将产品的各个测试设备,通过局域网网络和数据库的结合来更好地控制产品的测试过程以及方便地对产品测试数据进行分析。所以,在测试领域,更多的制造业企业会以此为标准,进行广泛地应用,不但提高了效率还保证了产品质量。     

RMS design methodology for automotive framing systems BIW

Today, markets increasingly require more customized products, with shorter life cycles. In response, manufacturing systems have evolved from mass production techniques, through flexible automation and mass customization, to produce at mass production costs. Manufacturing facilities must incorporate more flexibility and intelligence, evolving toward reconfigurable manufacturing systems (RMS). RMS are amid to posses such flexibility and responsiveness and said to be the next generation of world class systems. RMS are designed for rapid change in structure and for a quickly adjustable production capacity. This paper proposes a new methodology (high level process) of framework using flexible and reconfigurable manufacturing systems principles for automotive framing systems as well as to provide a guideline to support the structure of different stages of the design methodology. The proposed methodology is presented through a case study using data based on actual production systems of three different styles; (process and design data) which supports the hypothesis of the research.     

Utilization of a reinforcement learning algorithm for the accurate alignment of a robotic arm in a complete soft fabric shoe tongues automation process

As usher in Industry 4.0, there has been much interest in the development and research that combine artificial intelligence with automation. The control and operation of equipment in a traditional automated shoemaking production line require a preliminary subjective judgment of relevant manufacturing processes, to determine the exact procedure and corresponding control settings. However, with the manual control setting, it is difficult to achieve an accurate quality assessment of an automated process characterized by high uncertainty and intricacy. It is challenging to replace handicrafts and the versatility of manual product customization with automation techniques. Hence, the current study has developed an automatic production line with a cyber-physical system artificial intelligence (CPS-AI) architecture for the complete manufacturing of soft fabric shoe tongues. The Deep-Q reinforcement learning (RL) method is proposed as a means of achieving better control over the manufacturing process, while the convolutional and long short-term memory artificial neural network (CNN + LSTM) is developed to enhance action speed. This technology allows a robotic arm to learn the specific image feature points of a shoe tongue through repeated training to improve its manufacturing accuracy. For validation, different parameters of the network architecture are tested, and the test convergence accuracy was found to be as high as 95.9 %. During its actual implementation, the production line completed 509 finished products, of which 349 products were acceptable due to the anticipated measurement error. This showed that the production line system was capable of achieving optimum product accuracy and quality with respect to the performance of repeated computations, parameter updates, and action evaluations.     

Lead time control in multi-class multi-stage assembly systems with finite capacity

In this paper, we model multi-class multi-stage assembly systems with finite capacity as queueing networks. It is assumed that different classes (types) of products are produced by the production system and products’ orders for different classes are received according to independent Poisson processes. Each service station of the queueing network specifies a manufacturing or assembly operation, in that processing times for different types of products are independent and exponentially distributed random variables with service rates, which are controllable, and the queueing discipline is First Come First Served (FCFS). Different types of products may be different in their routing sequences of manufacturing and assembly operations. For modeling multi-class multi-stage assembly systems, we first consider every class separately and convert the queueing network of each class into an appropriate stochastic network. Then, by using the concept of continuous-time Markov processes, a system of differential equations is created to obtain the distribution function of manufacturing lead time for any type of product, which is actually the time between receiving the order and the delivery of finished product. Furthermore, we develop a multi-objective model with three conflicting objectives to optimally control the service rates, and use goal attainment method to solve a discrete-time approximation of the original multi-objective continuous-time problem.     

An online collaborative semiconductor yield forecasting system

Yield forecasting is a very important task to a semiconductor manufacturing factory which is a typical group-decision-making environment. Namely, many experts will gather to predict the yields of products collaboratively. To enhance both the precision and accuracy of collaborative semiconductor yield forecasting, an online expert system is constructed in this study. The collaborative semiconductor yield forecasting system adopts the client–server architecture, and therefore the necessity for all experts to gather at the same place is relaxed, which is especially meaningful for a multiple-factory case. To demonstrate the applicability of the collaborative semiconductor yield forecasting system, an experimental system has been constructed and applied to two random-access-memory products in a real semiconductor manufacturing factory. Both the precision and accuracy of forecasting the yields of the two products were significantly improved. Besides, the collaborative semiconductor yield forecasting system was also considered as a convenient platform for the product engineers or quality control staff from different factories to share their opinions about the yield improvement process of a product being manufacturing with the same technology in multiple factories.     

A generative neural network model for the quality prediction of work in progress products

One of the key challenges in manufacturing processes is improving the accuracy of quality monitoring and prediction. This paper proposes a generative neural network model for automatically predicting work-in-progress product quality. Our approach combines an unsupervised feature-extraction step with a supervised learning method. An autoencoding neural network is trained using raw manufacturing process data to extract rich information from production line recordings. Then, the extracted features are reformed as time-series and are fed into a multi-layer perceptron for predicting product quality. Finally, the outputs are decoded into a forecast quality measure. We evaluate the performance of the generative model on a case study from a powder metallurgy process. Our experimental results suggest that our method can precisely capture the defective products.     

Evolution of the linkage structure of ICT industry and its role in the economic system: the case of Korea*

ABSTRACT

When assessing the structural role of an industry sector within an economic system, considering its relationship to other sectors is crucial. Among others, Information and Communication Technology (ICT) industry, one of the innovation accelerators or key engines of economic growth, is evaluated. Specifically, we analyze inter-industry production inducement linkages within a qualitative input–output analysis framework, since it is useful for understanding the key structure of an economic system. Our purpose is to understand the significant spillover structure of the Republic of Korea’s ICT industry within the national production system, as it has played an important role in the national economy and grown dramatically over the years. The findings from the structural analysis, focused on changes in links, are as follows: First, ICT manufacturing showed a higher degree of heterogeneity than ICT service sectors in its sensitivity effects structure, an indication that this sector needs to be utilized in various other industries. Second, the spectrum of industries having significant production inducement linkages with the ICT industry is limited and furthermore, the influence effects of the ICT manufacturing sector diminished considerably although the ICT industry’s sensitivity effects increased. Finally, intra-industry linkages within the ICT industry are gradually strengthened especially between ICT services and manufacturing. These findings call for sustained policy efforts to promote the virtuous circle in the overall inter-industry production inducement system, by increasing the utilization of products and services from other sectors by ICT sectors (especially ICT manufacturing) as well as the application of ICT in other sectors.     

An enhanced model for the integrated production and transportation problem in a multiple vehicles environment

Solving an integrated production and transportation problem (IPTP) is a very challenging task in semiconductor manufacturing with turnkey service. A wafer fabricator needs to coordinate with outsourcing factories in the processes including circuit probing testing, integrated circuit assembly, and final testing for buyers. The jobs are clustered by their product types, and they must be processed by groups of outsourcing factories in various stages in the manufacturing process. Furthermore, the job production cost depends on various product types and different outsourcing factories. Since the IPTP involves constraints on job clusters, job-cluster dependent production cost, factory setup cost, process capabilities, and transportation cost with multiple vehicles, it is very difficult to solve when the problem size becomes large. Therefore, heuristic tools may be necessary to solve the problem. In this paper, we first formulate the IPTP as a mixed integer linear programming problem to minimize the total production and transportation cost. An efficient genetic algorithm (GA) is proposed next to tackle the problem when it becomes too complicated. The objectives are to minimize total costs, where the costs include production cost and transportation cost, under the environment with backup capacities and multiple vehicles, and to determine an appropriate production and distribution plan. The results demonstrate that the proposed GA model is an effective and accurate tool.     

Clustering and group selection of interim product in shipbuilding

Cellular manufacturing of the interim products based on group technology (GT) is a useful way to increase the productivity and improving the process flows of shipbuilding. In this paper, we analyze the similarity relationship on interim product to form product families and establish requirement mode of similarity measurement based on shipbuilding GT. The classifier of ART2 artificial neural network is proposed on the basis of the adaptive resonance theory. It can classify and identify automatically the input data through analyzing the characteristics of interim product in shipbuilding. With the modified algorithm, the interim product can easily and efficiently be formed product families and controlled to group related interim product into families in reason by similar coefficient. Meanwhile, through analyzing the production constraint to form assembly cell, the set of evaluating indexes is founded whose weights are decided by the entropy weight method and the comprehensive assessment values of product families design schemes can be computed to judge the final optimization scheme through the grey correlation analysis. The formation of sub-assembly interim product family in a bulk carrier is taken as an example to verify the proposed method, and the results show that it is an effective method for solving the group manufacturing problem in shipbuilding activities.     

Risk reduction in new product launch: A hybrid approach combining direct digital and tool-based manufacturing

The research contributes to the body of knowledge on new product development by considering the potential use of a major emerging production technology in the early phase of final production. Direct digital manufacturing (DDM) methods such as additive manufacturing (AM) have been introduced as a production method for some small and complicated parts, mostly in the aerospace and medical industries (in batches of one or a few). However, it still is not viewed as a suitable method for producing numerous parts in small batch sizes. In this study, we will utilize scenario-modeling based on real-world case data to illustrate the potential of a novel production method which we call “hybrid production” in new product launch. This production method combines DDM with conventional production methods over the product life-cycle. Our case study data is on a toolless production method called Incremental Sheet Forming (ISF) which is theoretically a DDM method. The cases have been analyzed to understand the economic feasibility and benefits of DDM utilization throughout new product life-cycle. Results of our study suggest, while implementation of conventional production from the beginning does not present a significant cost savings over the hybrid production, when product succeeds in the market, conventional method yields a high cost when the success does not materialize on the first attempt. This directly translates to investment risks (related to the cost of tool modification or replacement and inventory obsolescence), in addition to loss of flexibility to respond to market feedback and consequently lower chance of market acceptance. Additionally, DDM at the beginning of our proposed hybrid production can shorten the products’ time to market which is considered to be an essential factor for success.