In the coming Industrie 4.0 age, managemt such as business model and business system is required rather than just factroy systems to be built. This study suggested perspective combined Hardware with Software architecure is useful to make "make or buy" and "open and close" strategy. By comparative analysis of present German and Japanse FA system makers' businesses, it is considered how Japanese manufacturing companies will establish competitive advantages for the future.

View full abstract

[in Japanese]

View full abstract

This paper discusses the advantages of manufacturing facility simulation linked with physical systems in the industrial automation. The activities on manufacturing preparation and implementation processes of manufacturing systems have been improved using real and virtual manufacturing models. The standard reference model of enterprise and control integration is described and simulation layers in the hierarchy model are defined. The manufacturing facility simulation is linked with both physical controllers and their simulators by standard communication middleware connecting various communication protocols of the controllers and fieldbus. The applied case studies of improved processes and their advantages are explained.

View full abstract

TokushimaU-UTeM Academic Center (TMAC) has been successfully established in 2014 to promote the joint collaboration between the two institutions in terms of both education and research. One of the critical factors of this success was a good relationship between the two institutions based on the academic collaboration with alumni network in the past one decade. Presenting the overview of TMAC with its history and some of the critical on-going activities, this paper shows the feasibility of manufacturing system globalization through academic collaboration, and emphasized the importance of global engineer education.

View full abstract

A380 was reported on missing target in new production and leads to delayed market entry due to their weak product life cycle management (PLM). PLM is a strategic approach in the product development process. As a computer environment has evolved, the PLM also takes the change to be evolved. Accelerating the time in production process become important nowadays in order to capture the market demand and trend. In order to achieve the better glance, footwear manufacturing is taken to be discussed details. Accelerating time in the product development process and global information are two of the benefits of PLM in the cloud environment.

View full abstract

Due to the heterogeneous and anisotropy composition of the CFRP composite materials, trimming procedures can damage the material in ways that directly affect its mechanical properties. Currently a costly try and error approach is employed, to determine the optimal process conditions. To overcome the traditional approach, a numerical modelling of trimming CFRP composites using cross-nick cutter is proposed to simulate the process. A set of machining tests were performed in order to validate the accuracy of the model. Good agreement between simulation and experimental result shows the validity of the model in handling real-field problems.

View full abstract

A study on the relationship between wire electrical discharge machine (WEDM) parameters and surface finish are presented. This study analyzed the relationship between cutting parameter as variable input and surface roughness as output. To optimize the surface roughness which was measured both in horizontal and vertical direction, a Box Behnken of response surface methodology (RSM) with three process parameters; voltage, current and feed rate are used. Analysis of variance (ANOVA) was used to study the identified parameters that affecting the machined surface finish. With the results of P-Value and F-Value, machining voltage is known as the most contributing factor, followed by current and feed rate. The direction of measurement was found to be less significant. The linear prediction models were developed with average error of 3.67% and 1.74% horizontal and vertical surfaced roughness measurement direction respectively. The result of the validation also shows a good agreement with predicted values. It is beneficial as to assist machinist in improving the quality of the WEDM surface.

View full abstract

Customer demand is often crucial in manufacturing industry. There was research conducted to relate customer demand with safety stocks and lot sizes whereas others had included customer demand in Overall equipment effectiveness (OEE) with some assumptions on the characteristics of customer demand. This research proposes the procedure to observe the impact of different customer demand in affecting equipment utilization. Simulation is used to mimic different customer arrival distributions to production line and their corresponding impact on capacity and relationship between effectiveness are both examined. The research outcome is expected to provide new insight on the production planning and capacity expenditure in future.

View full abstract

Job Shop Scheduling (JSP) is widely applied in Industry especially for supply chain scheduling.Since JSP is a typical NP-Hard problem, a lot of researchers from various research disciplines,use algorithms to try to get the minimization of makespan(The time between the start and finish of a sequence of Jobs).However,the last makespan varies from each other due to different in travelling time from machine to machine. This paper shows how to determine the machine layout,and finally test the effectiveness of proposals by using a simulation model.

View full abstract

[in Japanese]

View full abstract

Petri net can represent discrete event systems graphically and mathematically. For the purpose of analysis of 0-1 integer programming problems, we propose the convertion algorithm for 0-1 integer programming problems into an optimal firing sequence problem of Petri net. We compare the solution of 0-1 integer programming problems with that of the optimal firing sequence problem of Petri net. The results show that the solutions for both problems are the same for a variety of instances. Moreover, we propose the algorithm to extract decomposable structure of Petri net into two subnets. The performance of the decomposition is evaluated in order to extract effective decomposable structure. The validity of this algorithm is shown by using Petri net decomposition algorithm.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

The Social Force Model introduced is to represent physical interactions and congestions among the people and the environment.The Social Force Model computes a force impact from the pedestrian's self-consciousness while the other two illustrate interaction forces from other pedestrians and the walls.The interation force consists of a pspchological force resulting from distance between each other,and a physical force inspired by counteracting body compression and sliding friction.The Social force model is applied to the autonomous distributed AGV in order to avoid the collision and achieve the effective transportation process, in this research. Simulation model is constructed based on the social force model by using multi method simulator AnyLogic.

View full abstract

Recently, "Industrial Internet" and "Industry 4.0", which refer to the integration of physical industrial machinery with software, interne, along with network sensors, actuators, and other industrial components, are discussed intensively. These technologies provide decantation of manufacturing with cloud-based manufacturing system (CBMS). Since the CBMS enables cross sourcing between connected companies, utilization of manufacturing facilities in these companies. As a result, these companies can reduce production cost. We have developed new concept of CBMS which implements dynamical generation of common manufacturing-bill of materials (M-BOM) from engineering-bill of materials (E-BOM). Simulation study on the effectiveness of the CBMS is summarized.

View full abstract

In sheet metal processing, reducing the waste of raw materials and managing the suitable production schedule are the important factors for the efficient manufacturing. In order to reduce the waste of sheet metal, the optimal cutting layout is determined by nesting. On the other hand, the manufacturing sequence for all manufacturing processes is decided through scheduling. Nesting and scheduling are affected with each other and there sometimes occur a certain trade-off between them. Therefore, those problems should be considered simultaneously in order to increase the efficiency of entire manufacturing activity. We have proposed the integrated method of nesting and scheduling. We adopted EDD based dispatching rule for scheduling in this method. In the manufacturing activity of sheet metal processing, bending process sometimes become bottleneck. In bending process, all parts are processed individually. On the other hand, in welding and assembling processes those are the post process of bending, some parts are combined and processed simultaneously. In order to create more suitable schedule, the combination of parts at welding and assembling processes should be considered when the operation sequence of bending process is determined. Therefore, in this study, we propose a local search method for improving the initial schedule that is created by using EDD dispatching rule. Finally, we verify the effectiveness of the proposed method through some computational experiments.

View full abstract

Given a set I = {i|i=1,2, ... ,n} of current n items (for example, n green peppers) with their weights wi and priorities p_i, a lexicographic bi-criteria combinatorial food packing problem asks to find a subset I'(⊆I) so that the total weight Σ_<i∈I'>w_i is no less than a specified target amount t for each package, and it is minimized as the primary objective, and further the total priority Σ_<i∈I'>p_i is maximized as the second objective. The food packing problem has been known to be NP-hard, while it can be solved exactly in O(nt) time if all the input data are assumed to be integral. We have designed a heuristic algorithm such that for a given real ε>0, it delivers a feasible subset I' with the total weight at most (1+ε) times the optimum, and it runs in 0(n^2/ε) time. In this paper, we show a complementary property of an optimal solution in rounded instances to be solved by the heuristic algorithm.

View full abstract

This study aims at developing a high-efficiency lot assignment method in semiconductor device manufacturing. In the multiproduct-production variable-quantity manufacturing, materials (production lot) flows through the product line disorderly. Then the production lots and equipment resources have multi factors, the combinatorial problem is complicated and difficult. Therefore we propose a model that leads an optimal solution for the situation. Throughout the study, we propose a lot assignment model that applies grouping structure and Combinatorial Auction method.

View full abstract

[in Japanese]

View full abstract

This paper proposes a robust method for capacity requirements planning (CRP) that can generate a stable load plan against dynamic changes in manufacturing environment. The proposed method determines the processing periods of operation orders so as to minimize the probability that the resource requirement of each operation order exceeds the capacity of corresponding resource. In numerical experiments, the load plan generated by the proposed method is compared to those using the conventional method and the effectiveness of the proposed CRP method is demonstrated.

View full abstract

[in Japanese]

View full abstract

In this study, key decision for the creation of production plans are defined as production-planning attributes. Production-planning attributes correlate complexly in production-planning problems. Traditionally, the production-planning problem splits sub-problems based on experiences, because of the complexity. Such approaches make solution space over-restricted and make it difficult to find a better solution. We have proposed ZDDs' network for addressing larger production-planning problems. The network can be created by linkages of ZDDs that express constraints in individual sub-problems and between sub-problems. In this paper, we describe a solution searching method for the restricted solution space represented by the ZDDs' network.

View full abstract

This paper describes an optimizing process scheduling on Flexible Manufacturing System (FMS). In the case that each work cell takes different processing times to complete the same process, process scheduling becomes a complicated problem. In this case, process scheduling using a calculator is effective. Process scheduling using a tree search is effective method to obtain an optimizing schedule. As problems become large, so searching answers become difficult. For this reason, method for choosing an option is important. We adopt the methods is based on a method for evaluating a complete schedule and verify the effectiveness of the method through some computational experiments.

View full abstract

In our previous works, we proposed a multi-agent transport control system for a production system, which uses two types of information : path information obtained by communication with neighboring agents and pheromone information deposited on the tracks of products. In this paper, in order to improve the performance of the proposed method, we modify the update rule of path information, and also provide several type of information, deposited on the tracks of products, as candidates of pheromone information. Then, we estimate the effectiveness of the improved method embedded with one of the candidates information, through a simulation experiment.

View full abstract

In order to realize the efficient manufacturing thorough intensive manufacturing processes, 5-axis machine tools attract a wide attention. New scheduling systems are proposed by using 3DCAM aimed at high production efficiency for 5-axis machine tools, in this research. The scheduling systems consist of 3axis and 5axis CNC machine tools, 3DCAD database, Customized 3DCAM corresponding to types of machine tools, and the jobs. The proposed scheduling systems carry out following 3 steps. The individual machine tools refer the 3DCAD database to obtain the 3DCAD data of jobs which can be carried out next machining process in the first step. 3DCAM creates the NC data and estimates processing time of all jobs in second step. The individual machine tools and jobs carry out the autonomous distributed scheduling process in order to minimize the make-span in third step.

View full abstract

Process planning plays an important role of a bridge between product design and manufacturing. Computer-aided process planning (CAPP) has been a topic of discussion in this half century, and numerous researches have been performed. The recent diversification in customers' needs as been driving the development of agile manufacturing that can flexibly adapt to different manufacturing situations. CAPP should also be discussed from this point of view and a set of flexible process planning methods have been proposed. Those methods dealt with process planning only for one product and therefore generate the optimal process plan which achieves the shortest total machining time. However, in real manufacturing, multiple workpieces are machined using multiple machine tools. In this paper, those methods are enhanced from the point of view of total productivity. The selection problem of the optimal process plan in the conventional methods and scheduling problem are merged with each other, and selection a set of the optimal process plans for a set of formulated as a 0-1 integer programming problem.

View full abstract

This paper deals with job shop scheduling with alternative machines. Genetic algorithm incorporating heuristic rules for job selection and machine selection is used for searching schedules to minimize mean tardiness of jobs. As heuristic rules, SL/RPN + SPT rule is used for job selection and (WINQ+RPT+PT)×PT rule is used for machine selection. Numerical experiments are carried out to evaluate the performance of the proposed method in comparison with a genetic algorithm proposed by Moon et al, a Lagrangian relaxation method and a mixed integer programing method. Experimental results show that the proposed method outperforms the other methods in larger scale of problems.

View full abstract

Recently, global supply chain management has an important role for globalization of marketplaces and global production in order to maximize the total profit. The noncooperative game between the supplier and the retailer is generally formulated as a bilevel programming problem. The bilevel programming problem has the nonlinear objective function with demand uncertainty. However, the nonlinear function has some difficulties to be solved and the study on the algorithm to solve the nonlinear bilevel programming has not been addressed so far. In this paper, we propose the algorithm to derive Stackelberg equilibrium in order to solve the nonlinear bilevel programming problem in which the supplier and the retailer make decisions respectively with the demand uncertainty.

View full abstract

[in Japanese]

View full abstract

In this paper, CF (carbon Fiber) and an Epoxy matrix are used to make a very strong non-corrosive light weight CFRP Coil spring. The CF-Epoxy coil spring is produced by Winding CF cloth/strands on a silicone mold transcribed/produced from a 3D printed ABS coil mold. CF-Epoxy coil spring is subjected to a series of experiments to determine the mechanical properties

View full abstract

[in Japanese]

View full abstract

Recently the development of the prototyping technology using 3D printers is remarkable. However, it is impossible to make a prototype using same resin for the mass production, so that in this case it is necessary to make a mold. It was problem that these methods need high cost and long delivery time. We developed mold core and cavity by using FDM-3D printer (AM) for prototype manufacturing system made of specific-grade of resin, and evaluated as intensity target which produce 10 pcs of prototype less than 8 hours from receiving 3D data by web. To begin with 3D-data was received by web on the assumption, subsequently commenced designing mold core and cavity by using 3D-CAD, manufactured the core and cavity by FDM-3D printer, and mounted them on mold base, and produced prototypes by the mold injection process. Two types of 3D-data as product were provided for this study, which the box shape of 35mm×20mm×10mm (t=2mm) and the wheel shape as a little complex shape(925), and evaluated prototypes of each product about measurements and appearance characteristics.

View full abstract

The additive manufacturing is very useful for prototyping in design analysis and functional testing. It can reduce lead-time and costs for product development. As one of open issues, it is difficult to make larger models than the machine size and take long time for making solid models because processing point is limited in a very small area. So, we propose an approach that we divide one object into some segments, fabricate them with hollow space by using 3D-printers, and then run polymer into the hollow space by a vacuum pomp to reinforce the model strength. In the present paper, some experiments are conducted to evaluate this approach.

View full abstract

Color marking of titanium by titanium oxide coatings are widely used in many fields such as medical implants because they are environment friendly and safe to human body due to no dyes used. Laser irradiation is an efficient approach to obtain oxide coatings. But it has not been used widely in the industry as it is regarded difficult and not repeatable. In this research, the coloring mechanism of titanium oxide layers generated by pulsed Nd:YAG laser (3rd harmonic: λ=355nm) treatment in air was discussed. A series of colors were obtained by varying the laser irradiation pulse energy. With the assumption of formation of uniform titanium dioxide layer on titanium surface, it was confirmed that observed colors calculated by light interference theory were in good accordance with the experimental ones. For investigating hues, CIE 1931 x-y color space and Surface Profile Meter were used. In conclusion, intensity decrease according to increasing pulse energy.

View full abstract

[in Japanese]

View full abstract

A new planning method is proposed to invest manufacturing equipment based on asset management in order to cope with fluctuations in demand for manufacturing systems, in this research. Some portfolios are set to consider candidates of the equipment investment. Returns and risks of the individual portfolios are required to select the best portfolio in the invest planning based on the asset management. The returns and the risks of the portfolios are obtained by using the returns and the risks of the investment of the individual manufacturing equipment. The returns and risks of the investment of the individual manufacturing equipment are calculated based on the averages and the variances of profit for the cases where the individual manufacturing equipment are invested under assumptions that candidate production plans and their probabilities are obtained.

View full abstract

In the assembly process of high-mix low-volume production line, the control system must be flexible and robust to adapt to changes in state of manufacturing line such as switching product types, stopping for a short time, and out of stock of parts. We proposed a multi-agent based control system of the assembly process. In the proposed control system, a built-in controller in equipment autonomously performs functions of scheduling and load adjustment with the use of a multi-agent system. In this paper, we show mechanism to connect agents automatically by using IEC61499 function blocks. In addition, function block definition tool is developed to define system configuration and equipment behavior.

View full abstract

[in Japanese]

View full abstract

Recently, when performing maintenance plant, consider the advance path and interference during loading using the three-dimensional shape data of the plant structure during device loading. However, it is necessary to perform to know the multiple measurements of the position measuring accurately the measurements of plant structure. Therefore, it is required a lot of effort. In this study, self-propelled simultaneously to obtain the measurement position and the measurement point cloud using a robot, a structure in shape to develop a system for creating fully automatically. As a result, it was possible to create the retrieved structures the shape point cloud coordinates around while moving.

View full abstract

A workpiece produced by die-casting or other casting process has a relatively low level of accuracy. Thus, finish machining is often required for such workpieces. However, it is difficult to set up a workpiece in the designated position with the work holding table. To easily make cutter paths for the workpiece in the designated position, this paper proposes machining operation method. This relies on coordinating the workpiece and cutting paths to make it easy to find the correct position. In this study, reverse engineering is used to obtain the 3D model of a workiece. The model is scanned by a 3D laser scanner. The 3D lattices point data were then constructed by interpolating and thinning the scan data. The RANSAC method is applied to primitive shape recognition for unorganized point clouds to detect planes, cylinders and spheres. The system also recognizes primitive shapes: namely, planes, cylinders, and spheres, from the STL data and match the scan data with STL data. After fitting, the system can know the position of workpiece and the machine tool table. Thus, the position can be used to set the origin of coordinated workpiece and cutting tool path for the machining operation can be generated.

View full abstract