[in Japanese]

View full abstract

In this paper, we propose a method of capacity requirements planning (CRP) for generating robust loading plan. The proposed method determines the starting times and processing modes of production orders with the aim of minimizing the probability of overload to be occurred in the target manufacturing system. Through computational experiments, the effectiveness of the proposed method is demonstrated by comparing our method with two conventional methods.

View full abstract

Dynamic changes in manufacturing sites influence whole production planning. Therefore, production planning should be treated as comprehensive. However, the production planning has been treated as separate, such as process planning, production scheduling aad so on. In this study, we use Zero-Suppressed Binary Decision Diagram (ZDD) to represent only combinations satisfying the constraints in the comprehensive solution space. By searching in the ZDD, feasible solution can be obtainedコn this paper, we describe how to solve the dynamic production planning problem using the ZDD. Experimental results demonstrate that the ZDD can create and represent only feasible combinations adapting to dynamic changes, quickly.

View full abstract

In globalized manufacturing industry, we need suitable safety stock placement. But it is difficult to use exhaustive methods because we have to choose the amount of safety stock to each warehouse from many choices. Existing method has a problem of computational cost for simulation to get evaluation value. In this study, we focus on Kriging to reduce the number of times for simulation. It is the method to predict solutions' value by approximating the relation between design variables and responses with some kind of functions. In this paper, we show the new method is more efficient than the existing method.

View full abstract

5-axis controlled machining has a high degree of freedom, so we can get any tool attitude at each cutting point. On the other hand, the changes in tool posture may be redundancy, and cause the decrease in accuracy of machining surface. To solve this problem, in this study, 2-dimensional configuration space technique is applied to generate the tool path. If the interference surface of work piece is curved surface, There is no interference occur and the tool attitude change is optimal for the whole of machining surface. Through this technique, we can realize the optimization of the tool attitude change on the whole of machining surface. Finally, we verify the usefulness of the devised technique by machining simulation.

View full abstract

In these days, the robots safety rules have been revised. It's say that If there is an appropriate safety skills for robots, you don have to set the guard fence around the robots. So the appropriate safety skills are required. In this research, we develop the real time planning method for 6-axis robot posture with referring point cloud information with using 3D distance sensor. This system needs to solve the interference avoidance problem and course correction problem. We develop how to choose the avoidance posture for interference avoidance problem. In the experiment we mount this system to the 6-axis manipulator. We check how this manipulator moves if there is an obstacle on the way of the start to goat. As a result, we observe the robots course correction against the obstacle. In this research, we realized the real time planning method for 6-axis robot posture with referring point loud information with using 3D distance sensor.

View full abstract

This study proposes a new efficient method to reduce maximum cutting torque loaded on ball end mill with continuous multi axis controlled machining. In the proposed method, tool posture in each cutter location defined by NC program is changed gradually byreferring estimated results of instantaneous cutting force and shape information of engagement domain removed cutting edges. In the planning process, the tool posture on each cutter location is replaced by a new candidate of tool posture, which can satisfy following conditions, 1. Ball end mill can remove work piece surface with smaller cutting force, 2. The removed part of work piece with the new tool posture candidate is equal to removed part with the original tool posture, 3. Side part of cutting tool does not interference machined workpiece surface. And, this gradual replace is repeated several times. These process are applied for each cutter location in continuous tool path in NC program. To shorten calculation time, the ultra-parallel processing technology is introduced In the developed prototype system, the engagement domain and side part of cutting tool is divided into a large number of estimation point.Then, inclusion estimation process between the estimation point and the machined work piece volume is decomposed into simple estimation processes about tool swept volume.

View full abstract

In order to respond rapidly to global demand fluctuation at low cost, we challenge to develop the global supply chain management. In this paper, we present the multi-echelon parts allocation technology to support inventory control and meet demand fluctuation. This technology consists of two methods: in the planning phase, the safety stock level is simulated under the several demand fluctuation scenarios by using the inventory allocation method; in the execution phase, the supply allocation is calculated by using the mathematical optimization method・We applied this technology to supply chain evaluation and parts allocation in case of tight supply. In the result, our system contributes to reduce the global inventory and improve the order fill rate.

View full abstract

Implementing lean manufacturing (LM) enables companies to produce any product range in any order or quantity. To succeed in LM implementation, personnel need to have the expertise in deciding which LM tool to implement at the right time and on the right place・However, this expertise is not always available. In previous study, a simulation-based decision support system (SDSS) is proposed to address the gap between lean practitioners and simulation-based approaches in terms of expertise in utilizing simulation software tools and supporting real-time decision making by providing four functions through interactive use of process simulation. The functions are layout, zoom-in/zoom-out, task status and Key Performance Indicators (KPI) status. This study reveals the feasibility of the proposed decision support system. SDSS is able to address the complex interdependent input parameters in manufacturing line and provides lean practitioners with time to react to emerging problems, evaluate potential solutions and decide on LM implementation.

View full abstract

In this study, a methodology named Real-Virtual Fusion Manufacturing System (RVF-MS) based on CPS (Cyber Physical Systems) concept is proposed. It is designed as multi-agent systems and aims to adaptively and effectively deal with both external and internal fluctuations in real production shop floor. In this paper, an auction method to determine equipment agents to participate in decision-making in the virtual system under equipment failure is proposed. In the proposed method, the coordinator is failure equipment agents, bidders are equipment agents on the same process of the failure equipment, goods are unprocessed jobs assigned to failure equipment, respectively. In order to determine alternative equipment agents, coordinator implements auction. The effectiveness of the proposed method is verified by computational experiments.

View full abstract

This study aims at examining agile manufacturing to develop a high-efficiency lot progress scheduling method in semiconductor device manufacturing. Throughout the study, we propose a lot progress scheduling model that applies grouping and multi-agents with a focus on external and internal fluctuations. As one example of fluctuation while carrying out a certain plan, daily interruption of an express lot is considered. The concern in this situation is determining the problem-solving range between groups. We propose a one method for Re-lot assignment corresponding to the situation.

View full abstract

Conventional collision prevention systems for machine tools such as collision prevention functions on NC controllers are based on simulations using 3D models of work pieces that are made in advance. If the setting up of work pieces differs from the models, then collision accidents are unavoidable. Therefore, a collision prevention system for detecting work-piece setting defects of machine tools by simulation based on actual conditions from NC controllers and measurement data of work-pieces using 3D laser scanning is proposed. A collision prevention system for 3 Axis machine tools based on this concept had been developed in success. Moreover, a collision prevention system for 5 Axis machine tools has been developed. This paper presents a system concept, overview of the collision prevention system for 5 Axis machine tools.

View full abstract

Master production scheduling is the key information to synchronize all business units including production department and sales and marketing department. To make more effective and proactive, materials resource planning(MRP) and capacity resource planning (CRP) must be collaboratively performed in accordance with advanced planning and scheduling (APS) algorithms. In addition, it is necessary to have time buffer to protect from unexpected urgent orders, which disturb the resource usage and cause late deliveries. This paper proposes a calculation method for time buffer that is added to the normal production lead time. The proposed algorithm allows schedulers to make robust schedule for the manufacturing uncertainty.

View full abstract

In this paper, authors introduce an AR application developed for support drawing pipes in shop floor. Some pieces of pipes are not designed beforehand in design process, as they contain uncertainty earned by pipe production and assembling in a factory. These pipes are usually designed individually in shop floor by the method that is far from be digitalized. Our developed AR application measures the location of both ends of the pipe using image processing, automatically draw suitable shape of the pipe, and effectively illustrates graphics of the pipe as AR once the worker holds a tablet PC to the direction he wants to fix the pipe. The paper explains details of above Reapplications, including those system diagrams, and shows some results of demonstration for easy understanding how thy work.

View full abstract

This paper considers how to realize the efficient manufacturing process under energy-saving and limited resources. In this study, we focus on a typical machining process and attempt to model and derive an effect cutting velocity under electric power-saving from the point of view of capacity planning and machine scheduling. Some results of computational results demonstrate the applicability of planning/scheduling as decision-makings for greening manufacturing systems.

View full abstract

[in Japanese]

View full abstract

The effective education from an expert to a newcomer is an important issue to consider the meaning 0f accumulation of technical skill succession and know-how, 0r efficient training to a worker in the cell manufacturing system. We proposed the operator allocation method in OJT using a skill index for the cell manufacturing system of a task divide and a task exclusive system. However, in order to improve transfer and productivity of skill, the situation where two or more operators were allocated the same task occurs in an actual shop floor. In this study, we consider the operator allocation method which allows collaborative operation, and scheduling, although the procedure of task divide is maintained.

View full abstract

We have already proposed a job shop scheduling method to meet due-dates of jobs and to minimize overtime as far as possible. The proposed method can efficiently search schedules using the mixture of genetic algorithm and priority rule. In the method, however, a priority rule was used only for the decision of job sequence. The decision for overtime was carried out using the genetic algorithm alone. In this paper, we propose the more efficient scheduling method by applying the effective knowledge of priority rule not only for the decision of job sequence but also 11)「that of overtime. Numerical experiments show the effectiveness of the proposed method.

View full abstract

The estimation of accurate cutting time before cutting is important for improving productivity. Estimation of cutting time calculated by "length of tool path / command feed rate" can easily be obtained in commercial CAM systems because NC programs contain machining information, such as tool positions, feed rate, etc. However, the actual cutting time is longer than cutting time calculated by" length of tool path /command feed rate "due to acceleration and deceleration which occurs along the tool path. In this study, an estimation system of ・accurate cutting time based on the control system of the machine tool is proposed. Experiments have been conducted for cutting complex shapes using the NC machine tool to compare the actual cutting time, estimated cutting time of commercial CAM systems, and the estimation error of the proposed system was only 4%, while the estimation error of a commercial CAM system was 51%.

View full abstract

Our group has been focusing on Modeling and analysis of NC tool paths via Physics-Based Methods. This modeling based on Physics-Based Methods is necessary for the understanding the cutting process and the load balancing approach which enable the tangible reduction of cycle time. But it is difficult to decide the optimized cutting condition because many factors must be considered in cutting process even though Physics-Based simulations were able to be done on CAD/CAM system. S6 1n this study, the new system based on synthesis of this NC analysis and response surface methodology was created for the optimization of NC tool paths. The optimization system showed the flexible NC optimization designed using tool life equation and machine specification. Additionally, the consideration about depth of cut like the tool path geometry was done.

View full abstract

In this paper, we propose a column generation method for solution of rectangular cutting stock problem considering three-stage guillotine cutting constraint・ Considering the manufacturing constraints, we set the problem setting as following. Stock materials are roll-shaped. The required numbers of cut pieces are 1000 for each type of pieces. We used 30 types of cut pieces.

View full abstract

The objective of this research is to propose an extension of the Overall Equipment Effectiveness (OEE) by including information on OEE prediction through simulation modeling and experiments procedure (SMEP) in order to extent the benefit of OEE as key performance indicator (KPI). To succeed in total productive maintenance (TPM) implementation, a company needs to have the expertise in deciding which workstation or production lines needs priority improvements. This process simulation approach enables the decision-makers t0 forecast the output of manufacturing processes and the effectiveness of TPM, based on the input values. This study reveals the feasibility of the proposed decision support system. SMEP is able to address the complex interdependent input parameters in manufacturing line and provides the company with time to react to emerging problems evaluate potential solutions and decide on TPM implementation.

View full abstract

This paper describes about the development of mobile robot and localization system. The mobile robot mechanical system is build based on 3 wheels type and it includes of 5 units of ultrasonic sensors and dead reckoning sensor at each drive wheel. The mobile robot used skid steered method for its mobility. The ultrasonic sensors reading are filtered using Extended Hessian Normal Form (EHNF) to represent a model plane. Based on UMBMark method, the encoder is tune to reduce the systematic error and non-systematic error. Finally, the localization of the mobile robot is based on Extended Kalman Filter (EKF) method to estimate the mobile robot position in a global map.

View full abstract

Under growing concerns with sustainable society, green logistic optimization is becoming a keen interest to provide a plausible solution aiming at qualified service in global and competitive distribution system. As a key technology for such deployment, we have engaged in The studies on vehicle routing problems (VRP) noticing that transportation cost and/or CO2 emission actually depend not only travelling distance but loading weights. As a special variant of such VRP, many researches are recently turned to the VRP with simultaneous pickup and delivery. Hence, we propose a novel saving-based hybrid method that is amenable for various real world applications under the present concern. Finally, numerical experiments are provided lo validate effectiveness of the proposed method through comparison.

View full abstract

In the assembly process of high-mix low-volume production line, the control system must be inedible and robust to adapt to changes instate of manufacturing line such as switching product types, stopping for a short time, and out of stock of parts. In this paper, we propose 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. This system decides on actions in the manufacturing line depending on the BOM. In response to changes in production quantity and product type switching, the system automatically performs Load adjustment when equipment are added or changed. And we show mechanism of the multi-agent based control system to work with limited resources.

View full abstract

Manufacturing industries face to diversity of the market where customer orders are more and more complex, meeting their specific needs within short delivery time. In order lo shorten the delivery time, design processes should perform concurrently to the manufacturing processes according to the master production schedule (MPS). Nevertheless, one-of-a-kind production usually needs a design process before production orders are released because production items cannot define in advance. This paper proposes concurrent execution management of design and manufacturing by using design bills. Design bill represents family production items which can be produced by deciding the design sub-problems. The concurrent MPS generation algorithms are illustrated with a simple example. Finally, the paper shows capability of the new design process management and suggests some future challenges.

View full abstract

At chemical plant and such as factory, all plant maintenance work must be done accurately based on process document. Accordingly we propose a method that automatically recognizes progress of work from pictures of machines which are on maintenance or dismantling to indicate accurate tasks depends on progress to workers. We implement our system to tablet PC and we embedded in the client server system to manage data centrally. Finally we prove the effectiveness of our system by operational test that pictures which are taken at workspace are correctly sent to server to judge which process the work does and show tasks to client side PC automatically and correctly.

View full abstract

In recent years, efficient and agile manufacturing is aimed at in many manufacturing industries. In sheet metal processing, reducing the waste of raw materials and managing the suitable production schedule are the important factors for such manufacturing. In order to reduce the waste of sheet metal, the optimal cutting layout is determined by nesting. 0n the other hand, the manufacturing sequence for all manufacturing processes is decided through scheduling after nesting. Typically, nesting and scheduling have different objectives essentially, so they have been considered separately in each management sector. However, they are affected with each other and there occurs a certain trade-off between them. Therefore, those problems should be considered simultaneously in order to increase the efficiency of entire manufacturing activity. Under such understanding, we have proposed an integrated method of scheduling and nesting. In this method, we apply Bottom Left algorithm and Dispatching rule in order to solve the nesting problem and the scheduling problem respectively. In this study, to enhance the ability of nesting method, we propose the re-assignment method. Finally, we verify the effectiveness of the proposed method through some computational experiments.

View full abstract

In recent years, the situations surrounding manufacturing systems arc becoming increasingly large-scale and uncertain, and structures of them are becoming complex. Scheduling for such large-scale and complex manufacturing systems is more difficult and time-consuming task. For deciding the suitable schedule efficiently in such manufacturing systems, concurrent scheduling method for hierarchical distributed manufacturing systems has been proposed. In this study, we extend this method by considering the overlapping term of multiple jobs in order to increase the efficiency of entire manufacturing activity. Finally, we verify the effectiveness of the proposed method through some computational experiments.

View full abstract

A new method is proposed to judge the feasibility of a schedule. Here, the no-buffer scheduling problem is targeted because of its complexity due to the consideration of the blocking of machines. The proposed method focuses on the problem where the operation orders of machines axe given in advance, which is a simple linear programming. By adopting the two-phase method, the feasibility judgement of this problem can be performed.

View full abstract

We propose a method for multiple exchanges of job operation orders to cancel a violation of prohibition by mathematical programming. The method supports the local search for scheduling problems under no-buffer constraint. This constraint results in a blocking state which prevents any new operation after a machine completes job, and this blocking complicates a partial change of a schedule once made. When the exchange of two consecutive operations causes a violation of prohibition, the violation can be cancelled by exchanges of job operation orders on the other machine. The proposed method is for minimizing the number of exchanges. The method is also adopted to simulated annealing for improvement of a schedule planned by a dispatching rule.

View full abstract

The additive manufacturing has much possibility to realize the small-batch production. When a user fabricates a large product, the product model is usually divided into some segments in a feasible size for the processing machine. We call it parallel fabrication because some segments can be processed in parallel at a time. This approach has a merit to reduce the lead time to build one product. Therefore, the segmentation and connection design is required. In the present paper, the segmentation and connection design for parallel fabrication of product is dealt with. The purpose of this study is to propose a general segmentation and connection method to satisfy geometrical requirements. As the first step of this stud}らsome joint methods for cyliandrical shapes are proposed. These models are fabricated and evaluated with some experiments.

View full abstract

Additive manufacturing (AM) is effective in custom-made fabrication. Currently, most orthoses are fabricated by prosthetists. If the fabrication processes are replaced with AM, the lead-time to completion can become brief. The authors have already proposed a finger orthosis, which has favorable structure for AM. However, 3D-CAD modeling for it is laborious and necessary work because the figure varies among patients. In this study, a 3D-CAD model template is proposed. The modeling process is divided into two processes, which are structure building and dimension adjustment, for users t0 operate it easily. In this software system, morphing algorithm is adopted not only for easy operation but also for easy addition of templates. The proposed system was implemented and a test orthosis was modeled.

View full abstract

Recently, a soft actuator was developed using an electro-conducting polymer and an ionic conducting polymer. Moreover, stereolithography that uses a femtosecond laser was researched as a method of multi photon-sensitized polymerization. Last year, we tried a more stable and more rapid stereolithography method for fabricating an electro-conducting polymer using a gelatine. Although, it has a effect for stabilizing solidification, it makes reduction of thickness of the solidification area. The refore, we tried the reduction of repetition rate of femto second laser for the stabilitation of solidification area and we found more stable condition for rapid stereo lithography for electro-conducting polymer.

View full abstract

A game theoretic model consisting of a manufacturer and multiple suppliers under demand uncertainty is addressed. The manufacturer and the suppliers are analyzed by a Stackelberg game where the manufacturer is a leader and the suppliers are followers. The paper addresses one manufacturer and its suppliers who are involved in purchasing and production with uncertain demands. Quantities and quality of components are decided by the selected suppliers. The quality of components purchased from suppliers is unknown to the manufacturer while it is known to the suppliers. Two scenarios (average case and worst case) are investigated for the manufacturer to estimate the quality of components. Computational experiments are conducted to illustrate the features of the proposed models with different parameters. The results show the validity of the proposed model.

View full abstract

Robot assembly systems contain one or more flexible robots which can carry out a large number of operations, and can conduct flexible assemble processes. This paper proposes a conceptual design environment for such robot assembly systems. The proposed conceptual design environment composed of three key features, a simplified input device and a quantitative evaluation tools, and a visualization tool. Using miniaturized robot-type human interfaces, design engineers can rapidly determine the posture of assembly robots, and then, design engineers can also rapidly evaluate the assembly system using the quantitative evaluation tool and the visualization tool. In this paper, the proposed design environment is used to design an example assembly system to illustrate the effectiveness of this design environment.

View full abstract

Today mass consumption society cause excessive environmental loads and depletion of energy resources. This paper describes a system that seeks the most appropriate path between houses in the local area, in order to reuse used products for each other house. We 90t the idea from reuse of used furniture and home appliances in neighborhood that is habitually done in foreign country. This system reads the predefined map data that has information of house data on reuse. The system seeks a pair of houses that has the shortest route of distance on the map, among the combinations of houses with appropriate reuse conditions. A simulation is made for this system which showed a good result.

View full abstract

A renewable-energy-driven environment monitoring system NOMAD for lake water has been addressed. The system employs several drifting buoys to observe lake water quality and flow. Previously developed system employs commercially available network service as sensing information storage. However, it has significant communication constraints such as data pouring speed as well as data conversion. The system also uses socket communication with data server process so that the battery performance tends to be deteriorated in monitoring. In this study, data up loading system for sensory information management has been implemented using web-based application. The system verification is also experimentally conducted.

View full abstract

After the Great East Japan Earthquake March 11, 2011, result of the power outrage of the electricity and stopping supplies the lifeline of water, electricity, and gas, utilization of the renewable energy; which is one of sustainable resourcess, is reconsidered again in Japan. Especially in Fukushima prefecture, a ground source heat exchange is considered as one of the important sources. However, due to the much cost of digging die ground for the heat exchanger, it is difficult to build generally. In this study, to overcome the cost, it is proposed that the idle well utilized as the heat exchanger. Based on the proposed method, the snow-melting system by the heat sampled from the heat exchanger is build. In this paper, the concept of the proposed system is described. In addition to this, die monitoring system checking and controlling the operation of the proposed system is described. Using the monitoring system, tele-operation of the system is realized.

View full abstract

The current trends of rising energy price compel manufacturers to improve energy efficiency and saving. This paper proposes an energy-aware scheduling method using an energy-block modeling which considers combination of each machine's energy consumption pattern. The energy-block represents whole power profiles during one processing job and that is modelled as combined multiple rectangle energy pieces Each machine has specific energy consumption pattern for a processing job. This paper examines single objective to minimize total tardiness by changing the parameter of constraint of energy max peak. The proposed model is applied to an experiment with some cases of different combinations of blocks, and it is confirmed the balance of energy max peak and make span.

View full abstract

A systematic method has been proposed, in previous paper, to allocate human operators to machine tools and to determine machining schedules considering statistical distributions of machining times. It was assumed that the processing times executed by the human operators are represented by random variables according to normal distributions, in the previous paper. However, normal distribution cannot express distributions of machining time enough e.g. case that probabilities of delay are larger than that of earliness. A new method is proposed, in this paper, to assign human operators to machine tools and to determine machining schedules considering statistical distributions of processing times under the assumption that the processing times executed by the human operators are represented by random variables according to beta distributions.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Let I = I_1 ∪ I_2 ∪…∪ Im denote a union of m sets of items, where for each i = 1,2,…,m,I_i = {I_i_k|k=1,2,…,n} denotes a set of n items of the i-th type and Iik denotes the k-th item of the i-th type with an integral weight Wik and an integral priority γik.The food mixture packing problem asks to find a union I' = I'_1 ∪ I'_2 ∪…∪I'm of m subsets of items where I'_i ⊆I'_i so that the total weight of chosen items for r is no less than an integral target weight T, and the sum weight of chosen items of the i-th type for I's is no less than an integral indispensable weight bi. The total weight of chosen items for I' is minimized as the primal objective, and the total priority of chosen items for I' is maximized as the second objective. For the case in which there are two types of items (i.e., m = 2), an O(nT) time dynamic programming algorithm has been designed. A new upper bound on the sum weight of chosen items of each type in an optimal solution is presented, and it empirically improves the execution time of the O(nT) time dynamic programming algorithm.

View full abstract

A demand of organ regeneration increase in recent year. There is a method using a scaffold in organ regeneration. Here, we assume that clinical engineers operate a 3D printer to generate a scaffold of organ regeneration in a hospital, and a possibility of a scaffold creation by a general and commercial 3D printer is discussed in this research. A basic discussion regarding a minimum member diameter and a minimum space between members regarding a scaffold are carried out when a filament material is PLA (polylactic acid). The obtained their results are feasible for the scaffold creation.

View full abstract

It is easy for laser forming to control power and laser forming does not use a tool, so there is not wear of a tool. There are a lot of advantage in laser forming. But laser forming is not practical because of working difficulty. In this paper, I observed SUS304 wire after laser forming by SEM. I processed SUS304 wire and measured bending angle. Additionally l got temperature of SUS304 wire by simulation.

View full abstract