This paper proposes a prototype of discrete event simulation for supply chain system analysis. Designing of supply chains requires system evaluation methodologies that enable supply chain planners to estimate various types of operational scenarios. The proposed simulation can represent information processing flows, which include demand predictions, operational orders, data monitoring operations, besides operations in traditional manufacturing system simulations. The simulation system analyzed several supply chain system operations, which are based on the actual system specifications and operational scenarios.

View full abstract

In recent years, customer's needs are diversified or complicated, and it is necessary to a flexible manufacturing system in manufacturers. In order to design the system efficiently and economically using the KAIZEN activities, a virtual manufacturing system simulation is used. However, the simulation data cannot share and simulation cannot cooperate with external systems. In this research, expression of simulation data is proposed. The expression uses the XML language. The proposed one apply to the communication between a simulator and control demands.

View full abstract

The demand to the manufacturing system is change by the diversity of customer needs and progress of IT technologies. Therefore, production system is shifted to production with flexibility of volume and period. Manufacturing scheduling is becoming more complex and more important. However, in almost manufacturing scheduler used in production system, work time is constant and the schedule to be made can not describe a real scheduler. In this paper, a manufacturing simulation is developed by object oriented method. The simulation can emulate behaviors depending on the scheduling data which is made by to external scheduling system. And by doing the feedback on the result, this research develops the system which can produces highly sophisticated schedule and evaluate the scheduling date.

View full abstract

The diversification of consumer's needs and the shortening of product's lifecycle increase uncertainty about production volume and period. Manufacturing systems are now required to realize competitive and optimized ones which can be following a fluctuation of demand in the market. Therefore, we propose a methodology for designing manufacturing systems with adaptability for market trends. This methodology consists of two methods; namely, "manufacturing system modularization" and "implementation of improvement". In this paper, the usefulness of this methodology is verified by computer simulation.

View full abstract

In Distributed Autonomous Production System (DAPS). It is difficult to realize the production progress suitable for the object of production system, because each system elements (e. g. Job, work-cell) are autonomous and selfish. Moreover, the object of the system is not stable by frequent changes of production circumstances. We develop the new production control method for the suitable production progress. The idea of this method is to harmonize system elements in order to reduce conflicts among system elements. The concrete method is that when an autonomous Job decides to dispatch an operation to the work-cell, a job considers the object function of work-cell as the constraint.

View full abstract

On designing the manual operation processes, it is important for process engineers to estimate the operation time and assume the worker's motion, in order to modify and optimize the process and workplace. 3-D Mannequin has useful functions for the engineers to make such estimation efficiently. However, the conventional 3-D Mannequin requires a lot of time to modify the process and the workplace data. To reduce the time, we have developed the support system, which enables users to modify the process and the workplace based on the already modeled data. By checking the difference of the type of motion and the position of object between the already modeled data and the modified data, the users only have to re-model the minimum required motion.

View full abstract

Kaizen activities are important to improve productivity in the manual assembly operations. However, Kaizen after the production line has been put into operation could not bring about a significant improvement because of short product life cycles. It is, therefore, necessary to achieve Kaizen before the startup using the simulation. We call such activities virtual improvements. In this study, the procedure of the virtual improvement based on a computer mannequin is proposed, and a system which automatically generates human operation models based on the optimal arrangement of a parts box is developed.

View full abstract

We have proposed a new meta-heuristic method termed Hybrid Tabu Search to solve a hierarchical logistics optimization problem. It is formulated as a p-Hub problem with capacity constraint. The method divides the solution procedure into two levels. Its upper level solves a Hub location problem based on the sophisticated Tubu Search and the lower a connecting problem by Dijkstra method in terms of Lagrangean relaxation. We also call attention on the discount effect regarding mass transport. Through numerical experiments, we confirmed that the proposed method outperformed the commercial software known popularly, and can deal with the real life problems effectively.

View full abstract

Recently, shortening of lead-time for product development has been highly required under a variety of customer demands. Under such circumstances, some engineering design problems can be formulated as multi-objective optimization problems (MOP). Though the Pareto optimality gives a rational decision norm in MOP, it will never provide a unique solution that should be the best compromise of decision maker. In this paper, we have proposed a novel multi-objective optimization method that employs soft computing technique like RBF networks. Through numerical experiments, we have shown that the proposed method can improve the design process efficiently when using not only physical model but also metamodel of the system under concern.

View full abstract

Since each company has to concentrate on core competence in the supply chain, production is never completed inside of the company. This paper proposes a cooperative scheduling technique using lagrangian decomposition coordination method. The target problem is a production model including suppliers, transportation service vendors, and makers of the products. The transportation routes can be mixed due to the cost reduction. This paper shows a mathematical model of cooperative scheduling and some results of numerical experiments.

View full abstract

In the production under diversified customer demand and global competition arising from complicated and unstable decision environment, it is hard to deal with the situation flexibly and extensively by the centralized scheduling system. Moreover, the scheduling itself becomes extremely difficult according to the increase in scale of the system. For resolution of these problems, this paper focuses on the decentralized scheduling system with two different protocols in information exchange. The we compared the characteristic of the system between the cases of using a synchronous protocol and an asynchronous one for adjusting the sub-schedule generated by each process individually. Numerical experiments are provided to reveal the feature of each protocol.

View full abstract

A variety of optimization problems determining a sequence in a mixed-model assembly line have been studied to improve the production efficiency like parts usage rate and workload. When a discrepancy among the assembly times is large in such production process, inefficiency such as line stop, increase in idle time, etc. will occur likely by using the conventional sequencing method. To cope with this problem, in this paper, we describe the model associated with the relief-men who are not in charge of regular operations. After formulating it as a combinatorial problem under mild conditions, we have developed a meta-heuristic algorithm to derive a near optimal solution. Finally, we evaluate the method through comparison based on numerical experiments.

View full abstract

This report proposes a scheduling method for a bi-criteria job shop scheduling problem. The criteria considered in this report are tardiness cost and sequence-dependent setup time. Two priority rules each of which is effective to each scheduling criterion are used to make a composite dispatching rule. One of the two rules is SIMSET rule that selects the next job with the shortest setup time SIMSET rule can make effective schedules with regard to average setup time. The other rule is constructed using a neural network. This neural network is trained by a simulated annealing method to make effective schedules with regard to tardiness cost. The proposed composite rule has the parameter α and can make effective bi-criteria scheduling by tuning it. Numerical experiments show that the effectiveness of the proposed rule.

View full abstract

The sheet production process treated in this paper has one machine and several equipments, by exchanging according to the variety of sheets. Equipment exchange causes several losses, such as setting up loss of resource as well as the total processing time. It is necessary to determine the production schedule of jobs and corresponding equipment satisfying the due date and the several requirements on sheet production. The scheduling problem should take into account the multiple objectives and becomes much more complicated. In this paper, we propose the hybrid solution method by combining genetic algorithm and dynamic programming for two-objective scheduling problem, which finds Pareto optimal solutions more efficiently rather than solo genetic algorithm.

View full abstract

Research and development of an auto-programming and 3D-simulation system for NC (Numerical Control) machining of free-form surfaces are carried out. First, the architecture and interface of the system, theoretical basis of NURBS, format of NURBS-NC machining program, compensation of cutting path and 3D-NC simulation method are described. Then, NC machining of free-form surfaces on gas turbine blade, portable telephone and vacuum-cleaner is performed by using the developed system and the results discussed.

View full abstract

Sheet metal parts are widely used as main constructional elements of industrial products such as any household electric appliances and electronic equipments. Since a process planning of sheet metal part is so complicated, that computer aided engineering for sheet metal part has been investigated. In this paper, we focus on the shape of bending punch which is determined according to the conditions of interference. The proposing method can be used for the determination of a suitable punch and check of punch by using pattern recognition. We applied this method to actual sheet metal bending process of sheet metal parts in a shop floor of a sheet metal manufacturing industry as a case study, and validity of this method was confirmed.

View full abstract

This study deals with the automation of polishing process by using of an industrial robot. In general, the polishing process has been performed by handiwork of the skilled worker. However, the environment is extremely inferior, and the automation of polishing is strongly required. As one example of the polishing automation by the robot, this study deals with the impeller consisting of complicated free-form surfaces. And we examine items that should be taken into consideration during the automation of polishing. This paper describes the selection of the polishing tool, the experiment for understanding the characteristic of the tool, and the examination of the polishing path for a complicated free-form surfaces.

View full abstract

In this paper, the author discusses the basic idea of the Intelligent Artifactual System, which can adapt to environmental change. The author points out that the system should be analyzed as a whole system, which includes the design system, the manufacturing system and recycling system. This view leads to the concepts of circulation, relativity, self-referring and latent function. In this work, as an approach to realize these concepts, the idea of Meta Design is proposed.

View full abstract

This paper presents methods using Genetic Algorithm (GA) and Genetic Programming (GP) to presume mimetic object's intention. The latest research results obtained from simulation have proved a theory that if a robot presumed an intention of a mimetic object correctly, it will learn the object's behavior efficiently. Therefore, it can be justified whether a robot got a mimetic object's intention by evaluating the robot's learning efficiency. The simulation results based on these methods also will be presented.

View full abstract

Artifact's reusability should be enhanced, because a social system of mass-production/consumption causes many environmental problems. However, there is no methodology that improves artifact's reusability and adaptability, covering the whole artifact. The main purpose of this research is to construct the design methodology for sustainable society by using "Latent Function". Latent Function is defined as a new function appeared when artifact's surroundings changes. If an artifact has many Latent Functions, it becomes usable succeedingly. In this paper, we propose the method of estimating abundance of Latent Function in order to analyze its characteristic.

View full abstract

In this paper, the authors discuss on the semantic structure of artificial environment, which often gives mental stresses to the inhabitants. The authors point out that the artificial environment has a different semantic space from the natural environment. From the viewpoint of humanness, the artificial environment should allow unpredictable and/or unexpected elements in it. Some current examples of environmental design are introduced for discussing on this problem.

View full abstract

The research project of Biological Manufacturing System (BMS) has been carried out at Kobe University for more than ten years. This paper introduces some major results developed recently and several remaining topics for the realization.

View full abstract

This paper proposes a design method for a multi-robot system that acquires cooperative behaviors autonomously. From embodied cognitive point of view, a robot should have the capability of acquiring appropriate cooperative behaviors through its experience. One of the most important issues for this function is how to design an on-line autonomous behavior acquisition mechanism capable of developing the robot's role in an embedded environment. In our approach, reinforcement learning that uses Bayesian discrimination method for segmenting the continuous state and action spaces simultaneously is applied to a robot for behavior acquisition. In addition to this, in order to support the learning in a dynamic environment that originates from the other learning robots, neural networks are provided for predicting the other robots' moves at the next time step. The output signals are utilized as the sensory information of the reinforcement learning to increase the stability of the learning problem. A homogeneous real robot system will be built for evaluating our approach.

View full abstract

The aim of this research is to understand artificial environment, which is the most familiar environment for a personal life. The authors focused on the meaning structure in the relationships between the environment and human's awareness, who live there, Artificial environment is the accumulation of artificial objects. Each artificial object has functions, which are reflected creators intentions. However, the people who live there and use them usually get different meanings from them.

View full abstract

To pursue sustainable technology as a key responsibility for manufacturing all over the world, we have proposed a prototype system for evaluating and supporting a decision on life cycle engineering (LCE). For this purpose, giving a whole framework on the G2 software that is known as a development environment for intelligent applications, we have built a product life cycle model based on object-oriented approach as one of its element technologies. Then to deal with the multidisciplinary nature of the problem-solving, we provide a development environment for integrating a variety of element technologies by the aid of distributed information technologies. Through a case study associated with the multi-objective analysis regarding the plastic wastes, we have examined the plausibility of the approach.

View full abstract

One of the major research motivations in social science is the maximisation of the total public welfare in the society. Interactions amongst decision makers with rational behaviour in the society result in the social equilibrium with robustness. In this paper we try to take a metaphor of the social science aiming at the methodology to develop intelligent artiacial systems with adaptive behaviour.

View full abstract

Recently the competition among enterprises has become severe. Formation of an appropriate virtual enterprise is considered to be one of the effective and promising methods for survival of industries under severe competition. Partner search is one of the most essential issues to be solved in virtual enterprise. In this study, we focus on the partnering in construction of a virtual enterprise as a basic research to clarify its effective management. We developed a computer simulation model to construct a virtual enterprise through negotiations among enterprises.

View full abstract

In our economic society, the market serves as the mechanism for decentralized resource allocation. In the disciplinary of neoclassical economics, the fundamental theorem of welfare economics vouchs the Pareto optimality of the equilibrium allocation in Walrasian market. In this paper, we develop Walrasian virtual market with Market-Oriented Programming based on multi-agent paradigm, and validate on Pareto optimality of the allocation in the market by comparing with analytical methods.

View full abstract

By development of an information technology (IT) in recent years and shift of the production order in the production spot, the dealings form of a company is diversified and decision-making in dealings is becoming still more complicated. In this research, we build an "artificial market" on a computer, and try to solve macroscopic phenomena such as formation of a market price and determination of a circulation route, from micro viewpoints such as each dealings act. Our artificial market is developed based on economics of complex systems.

View full abstract

This paper describes the welding risk computing system in consideration of different production culture for each area. Recently, production-system globalization progresses with globalization of production. That is why many unexpected problems on the globalizing system are reported. Such problems caused due to the traditional production-system not to take production culture containing human elements influence into consideration. Therefore in order to take the production culture into consideration within welding area, we develop the welding risk assessment system regarding the human elements with the pattern variable of Parsons. As a result, it succeeded in building a framework of the welding risk assessment system to judge the welded work piece security ahead of welding.

View full abstract

The need of computer-aided work support systems is increasing for reliable work and efficient work training in small lot production of diversified products. We have proposed that machine recognition of human work is an important function for autonomous computer support system. In this paper, RFID (Radio Frequency Identification system) is considered as a device for identifying work objects, estimating work position and monitoring other information during human work. A prototype system, which has a RFID reader/writer mounted on a glove, is developed to detect the RFID tag without special operation. The result of detection test in assembling work shows that the system can be used for monitoring work sequences.

View full abstract

This paper describes the innovation spread-able production system that continues taking in technical innovation regularly. These days, evaluation of competitive power changes in especially the production system, evaluation of which apparently shifts from production efficiency to marketability of the product developed by the innovation. The importance of generated and shared innovation-knowledge is increasing for the production system. In this research, the experiment in generating of an innovation and the simulation of it are curried out for the purpose of clarifying the human resource constitution ratio in the production system. As a result, generating of an innovation is specified and the suitable constitution ratio of easily transmitting the knowledge is found and in human resource viewpoint one idea to organize the innovation spread-able production system is proposed.

View full abstract

This paper presents Toyota's production structure in detail. Since the 1990s, the Japanese economy has undergone transformation such as an information society and globalization, and this paper analyzes how Toyota and its parts suppliers have been coping with these changes, so as to maintain vitality. Toyota still maintains its hierarchical structure almost intact. One big change is found in its "wide extension, " which means Toyota recommends its parts suppliers to supply parts to other automobile assemblers. This helps its parts suppliers to secure sales and to decrease costs due to the larger amount of production, and this provides basis to international competitiveness in order to survive fierce global competition.

View full abstract

This paper describes the relationship between industrial policy and development of machine tool industry in India. Although industrial and import licensing was liberalized to promote competition among machine tool manufacturers during 1980s, they did not react. Competition became tough after automobile policy was changed as a part of economic reforms. Demand of machine tools expanded and buyer demanded more sophisticated machine. Big machine tool manufacturers tried to introduce TPM and TQC to improve quality and labour productivity. Indian manufacturers are still protected by high import duty. While it is being reduced, they can adjust because they have ability to reduce costs. However, they take time to establish them selves in export market. Protection, competition and expansion of demand was necessary for development of machine tool industry.

View full abstract

This paper deals with evaluation of ease of assembly for mechanical system. In order to realize the evaluation, RRV and wavelet analysis of RRI have been investigated. Some assembly experiments using a table system with unidirectional assembly concept have been carried out. Through the experiments, the level 3 information of wavelet analysis has been found to be useful for evaluation of assembly. Furthermore, (i) assembly from left and right side and (ii) concurrent assembly operation have been observed to be difficult.

View full abstract

In this paper, three main objectives are hung out. The first is to propose a method to visualize and estimate the thought processes of designers in a quantitative fashion. A newly proposed concept, "Importance Factors Connectivity Map" is to display connectivity between importance factors in checking a detail drawing in a three-dimensional space graphically and quantitatively. The second is to reveal similarities and differences of thought morphology developed based on the proposed method in terms of experience and nationality.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

A production planning support system based on a Genetic Algorithm (GA) has been developed for production planning linked with a parts supply planning at PWB manufacturing section two or more, and achieving a system for optimizing the capability of the factory. Technical features of the developed algorithm include (1) Reduction of solution space by placing adaptability functions at different hierarchical levels, (2) Increased efficiency of searching the proximity area of the optimum solution, and (3) Parallel computation using the island model employing multiple computers. The developed algorithm enables the rapid creation of an optimized production schedule.

View full abstract

Oil filter for automobile has relatively short lifetime, average one year, and the weight of those disposals has reached approximately 2.5 million tons per year. Those were treated as industrial wastes and the environmental impact has not been negligible. To improve the environmental performance, we analyzed the conventional oil filters by using Life Cycle Assessment (LCA). As a result, we found the fact that the level of environmental load was higher not only in disposal process of life-end products but also in manufacturing process including material refinement. Based on the analysis, we developed the new oil filter and its production system from the zero base so that the total environmental impact could be minimized. The structure of the new product is drastically changed. Filter element is directly formed from the fiber materials to the desired shape. At the same time, part counts are reduced down to possible minimum numbers and thermal recycle has become possible because of the no metallic part contained. In the production system, the direct 3D-forming technology has been developed in-house to realize the new 3D formed oil filter. At the same time, we aimed at the ultimate revel of environment-free plant by adopting all possible technologies in energy saving and zero emission.

View full abstract