Scheduling is an important problem of manufacturing system operation, and numerous studies have been conducted. Most of those studies deal with determining problem of the optimal schedule for manufacturing under fixed conditions. In real manufacturing, however, conditions are variable, and therefore, it is desirable to generate a robust schedule for change of manufacturing conditions. Researches have been conducted on robust scheduling considering change of processing time. There are other kinds of change of conditions, and faults in processing are taken into consideration in this research. We consider the situation in which some defect products are processed again in the original process after being modified by another process. The scheduling problem for this kind of manufacturing system is formulated as a mixed integer programming problem with nonlinear constraints, for which some solvers have been proposed.

View full abstract

Recent manufacturing systems need to be flexible to correspond to manufacturing situation rapidly. To meet this requirement, methodology for the distributed simulation system has been considered. Recent development of computer network technologies makes it possible to equip each facility with an independent simulation model and run the models/facilities on a network so that they can communicate with each other. This type of distributed system was termed as highly-distributed system, and manufacturing simulation for this system has been discussed. A framework and a time management manner for this manufacturing simulation with the set of models based on a master manufacturing schedule was proposed. However, there has not been any discussion about modification of the master schedule based on the simulation result. This paper proposes a scheduling method based on the framework of the highly-distributed manufacturing system.

View full abstract

[in Japanese]

View full abstract

A new tool allocation method is proposed, in this paper, to consider both load balancing and optimization of transportation based on mathematical programming. The proposed tool allocation method consists of two steps. In the first step, the individual tools are divided into tool sets based on the integer programming, in order to balance the total machining time by using the individual tool sets and reduce the number of transportation. In the second step, the individual tool sets are assigned to the MCs based on the quadratic assignment problem, in order to minimize the transportation distance of jobs. Some case studies have been carried out to verify the effectiveness of the proposed method.

View full abstract

Air arc plasma cutting, which does not requires special orifice gas, is applicable to a wide range of small-to-medium-sized manufacturing industries because it is simple and economical. However, one of its limitations is that angular precision of the bevel produced by this method is inferior. Especially, when the cutting torch is combined to the simple automated machine, this kind of fault has occurred. The authors researched plasma cutting conditions for reducing sensitivity to noise factors in this case.

View full abstract

With a progress of CAD/CAM system and CG software, design and processing of more complicated shape is enabled. Machine tools having more than five axes are used for processing of this complicated shape, and processing of high-flexibility shape it is impossible for three-axis machine tool is enabled. In this research, turning center is used for an example of multi-axis machine tool. For a way of high-efficiency processing, a new rough cutting process by using simplified polygon model is suggested, tool path calculation method of that new processing is developed, and result of efficiency improvement by this developed process is considered and rough machining with the tool path.

View full abstract

In the fields of machining process, 3D models are used for the purpose of representing motions of tool and workpiece. However, 3D models can't describe changes of state to time direction continuously, so representing machining process efficiently is difficult. The objective of this study is to represent 5-axis cutting processes explicitly by using 4D mesh models, which can describe changes of state to time direction continuously. In order to show the availability of 4D representation, tool and workpiece are defined as "4D models of accumulated volume" which have information of motion records. In this report, 4D mesh models are obtained from time-series data of 3D voxel models which are made from 5-axis cutting simulation. Generated 4D mesh models are observed as 3D models by cross-section extraction. Also, cutting areas are obtained by collision detection between tool and workpiece.

View full abstract

Annotations such as GD&T (geometric dimensioning and tolerancing) and surface roughness included in 3D annotated model are needed for machining or inspection process. However, when annotations in 3D annotated model are used at downward process, it is necessary to associate machining area with annotations. The purpose of this study is to generate machining information based on machining feature with annotation from 3D annotated model and NC data. In this study, at first, feature candidates are searched in 3D annotated model. Then, these feature candidates are matched with the machining feature from NC data. The generated model can be used for process planning or on-machine path planning.

View full abstract

Power characteristics generated by renewable energy sources significantly depend on natural environment. Hence, this kind of power sources with output fluctuation may require adequate battery systems. On the other hand, it has been reported that zinc, leaden and the other necessary metals for battery electrodes may be exhausted in the near future. We have studied about novel energy storage systems based on gravitational potential energy. In this report, concept and modeling of the mass lifting mechanisms called LMM has been explained. Moreover, optimum design of mechanical subsystems has been examined through extensive CAD approach.

View full abstract

From the viewpoint of long-term usage, modular product has been attracting attention. On this study, a virtual market simulation system for evaluating the sustainability was proposed. First, production system model that consists of platform and add-on modules is introduced. Then to realize long-term usage of modules, we focused on user's different sense in value, and we proposed trading in the market using the difference between the values of the individual user feels for the same module. Finally, we construct multi-agent simulator which consists of plural user agents with the different value concepts and modules producer agent. From the results of case study, feasibility of evaluating sustainability was showed.

View full abstract

In this study, focusing on building an environment, which realizes user's LOHAS: lifestyles of health and sustainability, an emulator for a LOHAS environment is proposed. In recent days, LOHAS people are gradually increasing day by day. Moreover, the 2011 Tohoku earthquake and tsunami has changed the lifestyle of many people to LOHAS. However, since the change is suddenly, the engineering was not following the LOHAS. Thus, the construction of the engineering for realizing LOHAS is one of urgent problems. Using the proposed emulator, the data for analyzing a relationship between changes of the environments and the energy balance of the LOHAS environment. In this paper, a data serving system and a visualization system of the emulator are proposed. Using these systems with a sensor system, the total system visualizes the energy balance of the emulator. By an experiment using an experimental system implementing these systems, a validity of the proposed system was shown.

View full abstract

The lake water pollution is a problem in recent years. To elucidate the mechanism of water pollution, water quality monitoring is important. Propose a measurement system to provide water quality information needed to elucidate mechanisms of water pollution. This device is equipped with a sensor system, measurements performed by floating the lake. This system has to be cheap to use a large number of devices, and it is necessary for operation to be simple. This device transmits water temperature and positional information using Android device. In this paper, overview of the proposed Environmental monitoring system is first explained. Then, the flow and temperature measurement experiment explains. Finally, experimental results have shown the temperature can be measured using an Android device.

View full abstract

We present a game theoretic model in which one manufacturer and its suppliers are coordinated. Quality of products from suppliers is considered on the decision process. Due to fluctuation of demand resulting from varying quality levels from suppliers, manufacturers are subject to financial risk. A game theoretic approach to resolve coordination between manufacturers and suppliers under risk and uncertainty is proposed. The objective is to help decision making for global manufacturing in order to improve quality in competitive global business. Computational experiments are conducted to demonstrate the effectiveness of the proposed algorithm.

View full abstract

[in Japanese]

View full abstract

This paper deals with a method for planning of design and manufacturing processes of some groups of products in product family deployment. In general, it is important to reduce the design and manufacturing costs to achieve the economic efficiency in product family deployment, but the planning problem is very complex to solve, because the design and manufacturing processes are planned simultaneously. To cope with the difficulty, in this paper, we try to model the design and manufacturing processes as a mathematical programming problem, and to optimize the problem by relaxing the constraints, which are corresponding to the change of the business strategy in product family deployment. Furthermore, some numerical examples are demonstrated to show the efficiency of the method in this paper.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

The purpose of our research is to establish a simulation method to evaluate the manufacturing systems considering the energy consumption and the productivity concurrently. In this paper, firstly requirements for simulation to evaluate the energy consumption and the productivity are analyzed. Secondly a simulation method considering the requirements is proposed. A UML model that is defined facility state transitions and relationships between the facility state and the energy consumption is proposed. Manufacturing system simulation that is implemented in the proposed UML model is also proposed and developed. A system that provides a function to visually evaluate dynamical changes of the energy consumption per unit of the production throughput along time by our proposed simulation is also proposed and developed. Finally a case study for semiconductor manufacturing systems is carried out to confirm efficiency of our proposed simulation method.

View full abstract

Laser forming is expected to form sheet material without dies and molds. Laser forming is a plastic deformation method using local heat stress. It enables to rapidly make a prototype and decrease production costs because of die and mold less forming. Although many studies on laser forming have been done so far, there are few studies on forming curved surfaces. The object of this paper is to develop a method to generate the scanning paths and the scanning condition for forming curved surfaces by laser. The method based on the maximum and minimum curvature lines as the scanning paths. Since the two curvature lines are orthogonal, they don't influence each other in forming. The determination method of scanning speed of the laser for obtaining a demand shape was carried out based on the basic experiment data.

View full abstract

When machining curved surfaces by an NC machine tool, there are two problems. One is that the feed rate cannot reach the specified value. The other is that the quality of machined surfaces is decreased by big acceleration and deceleration due to machine control. These problems are caused by that the method to generate cutter location data does not consider the sampling time of the NC controller of a machine tool. To solve these problems, approximation to cutter locations with series of circular arcs is proposed. Circular arc interpolation algorism is complicated and not rational. In this paper, two methods to generate NC programs are proposed. The first one is to make approximation paths of the cutter locations with circular arcs based on the locations of the center of curvature. The second one is to make the length between the start point and the end point of a circular arc equal to an integral multiple of the "specified feed rate F × sampling time T". The effectiveness of the proposed methods is confirmed by simulations and experiments.

View full abstract

The posture optimization of a moving object for collision avoidance is one of important issues to control production resources such as 5 axis machine tools, robots and so on. In generally, a rotating matrix, which changes the posture of the object in an appropriate direction, consists of trigonometric functions, or no-linear functions. In this paper, a new set of equivalent constraints in the rotating matrix is proposed to plan the optimum posture of the object by mixed integer programming.

View full abstract

In our study, we propose a distributed cooperative approach for flexible flowshop scheduling. This method breaks up the system into sub schedulers, and they integrates as a single system. Since each scheduler needs only local information, it is effective to apply the proposed method into a large production system. Our target model handles lot formation problem and setup problem. In this model it is important to decide lot formation and machine allocation concurrently, when we decide scheduling. In this paper we propose a method that each sub-scheduler optimizes not only own local schedule but lotsizing to the model. We compare our method with existing methods, and show its effectiveness.

View full abstract

Semiconductor manufacturing process flow is consisted of hundreds of process steps using hundreds of machines repeatedly, which is called a re-entrant flow shop, and it is very difficult to optimize the production schedule. Furthermore, in recent years, it has been becoming more difficult to use machines effectively because the product-mix becomes much more complex and the production lot size becomes much smaller. Therefore, it is indispensable to integrate a variety of scheduling methods into a solution to achieve the high production efficiency. In our previous researches we applied the well-known cooperative scheduling method, Lagrangian Decomposition and Coordination method, into an abstracted model based on the actual production line. Then we have verified the proposed method can create schedule effectively. In this paper we show the effectiveness for also larger models and make clear the issue in applying it into the actual scale model.

View full abstract

[in Japanese]

View full abstract

In this paper, we study a new method for robot programming: view-based teaching/playback. It can achieve more robustness against changes of task conditions, such as initial positions of the object, than conventional teaching/playback. The proposed method is composed of two parts: teaching phase and playback phase. We used range images instead of conventional grayscale images, which were used in our previous studies. In application to pushing tasks by an industrial manipulator, view-based teaching/playback with grayscale images failed to carry an object to a goal in the case of changing lighting conditions or color of the object. On the other hand, view-based teaching/playback with range images succeeded even in such changes. The method with range images was also successful in pushing from initial positions that were not identical to those in the teaching phase.

View full abstract

This paper presents the three types of remote control method for intuitive robotic arm manipulation. The first method covers Wii remote control method using acceleration sensor and gyro sensor of two kinds of approaches; the servo movement and the arm movement. The second method covers data glove-based manipulation of two types of approaches; the gesture-based interface and master-slave-based one. The third method covers the voice command-based manipulation. The manipulation experiments using a 5 DOF robotic arm system were conducted with each of these methods. The results show that all of these methods performed the interactive manipulation of the robotic arm but some differences were observed in recognizing the command given by the subject due to the experiment conditions. Considering the results, this paper discusses the feasibility of robotic arm-based communication device in manufacturing environments.

View full abstract

Although a number of people are dead and missing because of the destructive East Japan earthquake, there exist many survivors because of successive evacuation drills. Therefore, evacuation drills are important requirements for schools and public places that house several individuals. Furthermore, for the effective validation of drill exercises and evacuation planning, evacuation simulations have increased significantly. Similar problems have been observed in recent aircraft accidents worldwide. However, most studies in this area have dealt with 2-D problems and are inadequate for realistic 3-D environments. In this paper, an extensive 3-D mass evacuation/rescue simulation method is addressed using a kinematic digital human model (KDH) with precise human joints. As an application to aircraft accident, mixed evacuation simulation considering extended KDH for injured people transportation has been effectively demonstrated.

View full abstract

In this research, the simulation analysis of shovel-truck coupling behavior in Open-pit mine. The mining machines are getting larger for efficient mining in Open-pit mines. Initial cost and operational cost increase with it is getting larger. Therefore, it is necessary to aim at improvement in the working efficiency of mining machines. It is effective consider to shovel-truck coupling behavior. In this research, we aimed to quantitatively analysis of the working efficiency of shovel-truck coupling behavior. This report clarified following three waiting: waiting for truck's spotting, waiting for unloading to truck, and waiting for truck staying at the loading point. This report explains an analysis model and the discrete simulation to examine various situations. In Simulation analysis of according to spotting system, In the case of the single spotting system, the number of waiting always occurred, and it resulted in hardly occurring by the double spotting system.

View full abstract

A life support system using robotic physical agents for partial disabled people has been addressed. Concept and design of a newly developed end-effector, which assist user's daily operations of home appliances such as sanitary ware, healthcare equipment and so on, is proposed. The device can control the distributed appliances in user's room environment by connecting its end-effector to the corresponding attachment on appliance. Experimental results of a prototype device called "M-GEAR" show the device motion characteristics as well as the effect of connecting tolerances on the power transfer performance.

View full abstract

Skill education in manufacturing industry has not been improved because there are a lot of difficulties in skill succession. Therefore we firstly clarify the present situation of expert skills, and then propose effective training programs. For this purpose, we discuss a visualization of expert skills using relationships between skills for promotion of efficiency of skill succession in the machine tool industry.

View full abstract