Uncertainties of operation time of manufacturing operations are not considered in the conventional scheduling problems, however the operation time may have some uncertainty and distributions for the case of the manual operations and the long distance transportations. An approximation method is proposed in the paper to estimate the distributions of the finish time of all the operations in the scheduling results, based on the mean values and the standard deviations of the operation time.

View full abstract

This paper deals with a real-time scheduling system of the Holonic Manufacturing Systems (HMSs). In the previous papers, real-time scheduling processes based on the utility values have been proposed and applied to the HMSs. A simulation based procedure has also been implemented to estimate the future status of HMSs and to determine the utility values aiming at improving scheduling method. However, it was assumed that only one holon in the HMSs carries out the estimation, in the previous research, for the ease of the estimation process. A simulation based real-time scheduling method is newly proposed, in the paper, to improve the estimation process which enable all the holons to estimate the future status of the HMS. Some case studies of the real-time scheduling are carried out to verify the effectiveness of the proposed method.

View full abstract

We have proposed the hybrid timing policy in reactive scheduling based on cumulative delay, which can be capable of the rescheduling decision timing under dynamic environment. This paper extends our previous works to parallel machine scheduling with urgent jobs arrived randomly. We first explain about the proposed policy. Then we examine the effectiveness of the proposed policy through some scheduling simulations and demonstrate the proposed policy can out perform a typical event-driven rescheduling policy.

View full abstract

In this paper, we propose a job shop scheduling method to meet due-dates and to minimize overtime as far as possible. The sequence of operations of jobs is determined using the mixture of genetic algorithm and priority rule. The minimization of overtime is carried out by tuning a parameter based on machine loads. Numerical experiments show the effectiveness of the proposed method.

View full abstract

We have developed the production scheduling system for the hybrid model production system where the batch process existed together with the discrete system. This system can make the manufacturing schedule in the seamless way by suiting it to the business feature of the production system, constrained condition, and the ability of various resources. Moreover, supporting automization and efficiency improvement to schedule the cleaning operation, which is peculiar to manufacturing processes such as foods and beverages, and the optimization plans of the manufacturing operation plan became possible in this system.

View full abstract

For quick and effective response to dynamically varying market demands, it is strongly required to make flexible manufacturing systems more efficient by introducing effective and autonomous control and management system. For efficient operation, it is important to decide which work in which area, such as warehouse, loading station, buffer and machining centers in the FMS to be transported to which area. This paper proposes an auction-based scheduling method for FMS which is applied to select and assign a work to AGV whenever it is required. For the unmanned operation at night, the applying order of the auctions concerning from- to areas is separately determined corresponding to time zones, i.e., day, evening and night. A concept of the lowest bid price, which allows accepting participants bidding over the lowest price, is introduced enabling to relax the dominating effect of the order of bid types asked in the auction. Numerical simulations are performed to investigate the properties of the proposed methods with the lowest bid price demonstrating their effectiveness.

View full abstract

A new heuristic scheduling procedure for re-entrant flow shop scheduling problem is proposed to plan an appropriate production schedule in actual manufacturing systems. In this study, flow time, which is a index of the evaluation of the planned schedule, is optimized to decrease amount of work-in-process. Cycle time, which is another important index, is determined by the total load time on a bottleneck machine to manufacture the products in high productivity. The result indicates that the proposed dispatching procedure gives a good schedule in comparison with the optimum solution.

View full abstract

We propose a novel production system, that is able to on-demand and sustainable development. In the on-demand manufacturing, shortening at the development period is important. However, the micro electro-mechanical systems (MEMS) manufacturing use very large system, and development period is very long. Therefore, a new prototype manufacturing system for MEMS was proposed. The basic process of this system is press forming that is possible to process a metal, a polymer, and other sheet type materials. The vacuum deposition process was also prepared. Furthermore, post thermal process, ink jet type wiring, and device test. The system was constructed on 5 separated units. Process equipment set on the unit, and transfer system was also set on. The separated system was able to reorder the process when products were changed, and they also work stand-alone. This flexibility realizes "on-demand factory" system. In this case, "on-demand factory" means the products plan, develop, trial, manufacture and supply on markets demand. The proposed flexible process system affect all product terms.

View full abstract

[in Japanese]

View full abstract

In this study, auction-based scheduling method is proposed for machining job shop in order to automate scheduling and make daily management easier. Real-time scheduling using auction-based algorithms are performed to make decision of each job order. Simulation by using same algorithms is also executed to forecast tardiness of jobs and take measures such as extension of working time. In this report, methods of simulation such as execution timing, evaluation of result and countermeasures against tardiness are proposed, and its effectiveness is investigated by computational experiments.

View full abstract

A square arrayed manufacturing system (SAMS) is one of the agile manufacturing systems proposed to cope with the dynamically varying market demands in recent years. AGV system with mesh type paths considered in this study represents the AGV system of SAMS of which operation plays a key role for its overall performance. The selection of an AGV to transport a work and selection of paths to the destination node for an AGV are two essential decisions in AGV operation. This study proposes auction-based methods for these decisions. In addition to the basic auctions, time window concept is introduced to the node selection and auctions for evacuation of an AGV from its occupying node and for pre-assignment of an empty AGV to other nodes are installed to improve the performance. The effectiveness of proposed methods are investigated by simulation studies.

View full abstract

In recent years, even in a job shop type production, much higher productivity is required. Multitasking machines are applied widely to reduce preparation time and to improve productivity. Generally, the multitasking machine has the functions for both of turning and milling operations. However, it means that collisions among tools, workpieces and machine tool itself often occur during machining operations. Therefore, the systematic collision avoidance method is needed. In this study, a scheduling method is applied to change the machining sequence to generate a collision free NC program effectively.

View full abstract

This paper deals with a simulation-based scheduling method of manual work in consideration of learning by On-the-Job Training (OJT). As of skillful work such as maintenance work of trains and airplanes, workers have to leam many kinds of work with OJT While processing time of novice workers is longer than that of experts, the time will be reduced with repeating OJT. Therefore, OJT is important to keep skill level and long-term work efficiency of the organization. In order to schedule in consideration of OJT, the scheduler must incorporate a management function of workers to trace work experience changing dynamically. In this paper, a simulation method that uses a worker model and an agent-based mechanism is proposed to derive the optimal OJT strategy toward long term performance. And, the effectiveness of this model is discussed.

View full abstract

For achieving effective maintenance, it is necessary to make appropriate maintenance plans. For this purpose, we have proposed a procedure to select maintenance strategy based on technological and managerial evaluation. In the case of condition based maintenance, however, it is necessary to determine further detail, such as monitored functional state, monitoring interval and criteria for executing treatment. In this study, we propose a method for such detail planning based on deterioration progress pattern. In the method, the functional states are evaluated according to the objective of the maintenance. The effects of monitoring interval and a criterion for treatment is evaluated using Monte Carlo simulation.

View full abstract

This paper proposes a new simultaneous process planning and scheduling method to solve dilemmas posed by conflict between optimality of a process plan and a production schedule using multi agent learning based on evolutionary artificial neural networks. The effectiveness of the proposed method is confirmed by solving a problem extended a benchmark problem by including larger volumes of products, thereby demonstrating high productivity resulting from role-sharing among machines. Results also show that the proposed method can achieve better results than other proposed methods.

View full abstract

Due to rapid changes of markets and pressures of competitions, industries are gradually adopting their production way to support more diversity of customer's needs. This paper deals with an agent-based architecture of the integrated systems for the process planning and the production control in the manufacturing systems. In consideration of alternative manufacturing processes and machine tools, the process plans and the schedules of the manufacturing resources are generated dynamically and incrementally. Such an adaptable process planning system is becoming increasingly important for the manufacturing enterprise. We initiatively combine the process plan networks and the heuristic search algorithms with the negotiation protocol to generate suitable process plans in the dynamic manufacturing environment.

View full abstract

In this paper, a two-layer methodology is proposed, consisting of a process design layer and a facility design layer. The objective is the decision maker in finding technologies that can be combined to meet a set of requirements. Specifically, the approach consists of a process selection agent (PSA) that submits design requests to technology agents. If a technology agent determines that the design is possible then proceeds to design the technology and submits the design to the PSA. Although the PSA is responsible to bundle the designed technologies into a solution, technology agents help to find the best solution by assisting the PSA in pruning away inefficient partial solutions.

View full abstract

Manufacturing systems has been designed to be rational optimal under static situations. Resently however, situations surrounding manufacturing systems are becoming increasingly large-scale and uncertain, and structures of them are becoming complex. It is difficult to deal with such incomplete conditions using existing top-down methods. Self-organization-based Biological Manufacturing Systems (BMS) has flexibility and adaptivity under complex and uncertain situations. On the other hand, local competition occurs among agents, or the system is trapped in local solutions, mainly because the sum of local goals of agents is not equal to the system goal. This study proposes a design approach for self-organization-based BMS by introducing agents' bounded rationality, to generate effective solutions under incomplete conditions, through avoidance local competition without top-down control or additional information. This paper shows the results of simulations, which indicate the existence of differences between two types of bounded rationality, and that designing appropriate bounded-rational agents might be an effective approach to derive adaptive solutions.

View full abstract

A cutting force model for end mills with consideration of runout effects is introduced in this paper. The cutter runout caused by cutter setup can be expressed a location angle and a radical offset. An estimation method, which can accurately identify these two parameters from a surface geometry, is also proposed. The feasibilities of the cutting force model and the runout determination method are demonstrated by comparing estimated and measured cutting forces.

View full abstract

In order to realize the articulate link between design and manufacturing and a traceable manufacturing system, it is necessary to represent the semantics of information independently of any implementations. A significant amount of research for machining process planning and machining feature extraction has been carried out to help realize an integrated CAD/CAPP/CAM system. However, all data are not used throughout product development processes. In this research, the Digital Semantic Model that fulfills the above requirements is proposed. In this paper, modeling and implementation of Digital Semantic Model for 5-axis machining is reported.

View full abstract

One of the most important performance indices of the machine tools is the machining accuracy in the multi-axis machining process. Therefore, it is required to establish systematic analysis and design methods for the multi-axis machine tools, from the viewpoint of the machining accuracy of the complicated products. The objective of the research is to establish mathematical models representing the shape generation motion deviations of machine tools, on the basis of the geometric deviations of the relative motion between a pair of connecting units with respect to the priority of the geometric features. A case study on the kinematic motion deviations of the three-axis machining centre is carried out with the proposed method.

View full abstract

The objective of this study is to develop a system to determine optimal machining conditions for minimizing energy consumed in a machine tool. To realize this objective, a module to estimate cutting energy and energy consumed in each unit of a machine tool is developed. Cutting energy in square end milling and ball end milling is estimated by the cutting force which can be derived from the cutting area and the specific cutting force. And, energy consumed in each unit of a machine tool is calculated by the electric energy model defined with basic experiments.

View full abstract

This paper presents laser processing technology that can be applied to magnetic interception of linear solenoid. This laser processing technology is used to open small holes without dross and sputtering by jet assisted air gas at a defocus position of a pulse YAG laser. In addition, high-speed technology is realized by laser processing while turning a part. This technology contributes to miniaturization and cost reduction, and linear solenoid have been manufactured since 2002.

View full abstract

This paper proposes a novel simulation framework that integrates discrete-event modeling with system-dynamics modeling. The former has strength in system performance evaluation; meanwhile, the later has an advantage of representing feedback mechanisms in complex systems. This paper summarizes this framework and represents application examples.

View full abstract

This paper presents an assessment for technical and economic feasibility of electricity generation from biomass. In particular, we focus on determining the optimal location of the biomass collection and electricity generation facilities using gasification technologies. This is accomplished by representing this decision making task as a location-routing problem (multiple allocation). Specific equipment selection and cost evaluation of the electricity generation facilities is accomplished by means of a method for the synthesis of biomass conversion processes. In order to illustrate the proposed approach. we present a case study for biomass-to-electricity generation in Nagano city.

View full abstract

In a construction production, management of various kinds of parts/units under corroboration project with various conditions such as status, position, attributes, construction order, delivery order, and so on. That makes difficult to grasps and manages construction projects by traditional production systems. In this study, details of actual construction processes management are found by listening comprehension for a general conductor and a parts/units maker. Then demands in construction process management are extracted. From these results, architecture and process flow of construction management system is proposed by using RFID.

View full abstract

This paper deal with control design using 3D model of the production facilities and the control framework which has the data structure and feature showing the effect in 3D model. Furthermore, reuse methods of design information are considered from the view point of control systems. Therefore, control system analysis and modeling are achieved and modeling methods of device controls and input/output (I/O) controls in 3D model are described. As a case study, the control object framework of processing and transfer facility control is proposed.

View full abstract

This research proposes a novel demand forecasting method which will work effectively even in such circumstances where extrapolate-able demand patterns are hardly available. The method uses the market mechanism to aggregate tacit knowledge of the firm's sales people on the future demand of a product into a continuous forecasted demand distribution. A new type of prediction security and an original market maker algorithm suitable for the security are introduced and furnished into an intra-firm prediction market system. As a result, sufficient liquidity is secured for the market even when the number of the traders is small, and the market maker can output at any time an aggregated demand forecast of the traders as a continuous distribution. An agent simulation model, where each trader has learning capability, is also developed to study how the method works, whether the output distribution is adequate, how quickly the output converges, etc.

View full abstract

While manufacturing factories in Japan strengthen high-variety low-volume manufacturing more and more, there is a great demand of higher measures for it from manufacturing sites. The real bottleneck has been an inspection process at the final assembly line which decides manufacturing model and ability. We developed a platform which solves this bottleneck and inspects high-variety of products with cross-model and cross-group. Furthermore, the platform also has a capability to shorten the inspection time and control the quality of manufacturing line.

View full abstract

In the recent years, consumers' demands are diversifying rapidly and complexly. In order to deal with the fluctuation of such consumers' demands effectively, manufacturers compose production planning and production scheduling by taking unconfirmed orders into consideration. Due to the difficulty of estimating consumers' demands, estimation errors occur when the unconfirmed orders become confirmed orders. Thus, it is important to understand how the estimation errors affect on the productivity in target manufacturing systems. In this paper, we mount an APS system on a Distributed Virtual Factory (DVF) which is modeled as an assembly type factory. The effects of the unconfirmed orders are observed from variety of viewpoint including product cost through experiments with the DVF.

View full abstract

Japan needs to maintain the international competitiveness of its industries and further develop the economy by overcoming resources, environmental and demographic constraints. Simultaneously achieving environmental conservation and economic growth is a policy goal of the Third Science and Technology Basic Plan. In particular, the plan's promotional strategy for the "MONODZUKURI Technology" field cites "human-centered manufacturing practices" as a basic policy, and declares Japan's determination to maximize the added value created through a value chain as a policy challenge. The center of a value chain is the production system, which should be aimed at benefiting both humans/environment and economy. Here is a summary of what this article proposes as particularly notable directions of future R & D on human- and environment-conscious Manufacturing System in Japan.

View full abstract

To optimize a large-scale logistic problem in a numerically effective manner, we developed a hybrid method termed hybrid tabusearch whose algorithms composed of tabu search and graph algorithms. In this study, to elaborate such a strategic decision from an operative level, we have concerned with the logistic network problem associated with vehicle routing and extended our previous method. Finally we verified the effectiveness of the proposed solution method through numerical experiments.

View full abstract

When evolutionary computaion is applied to the problem of the behabior coordination in a homogeneous multi-robot system, the adaptive specialization is important. Coevolution is useful for this problem. We sugest new criteria for coevolution and combination technique of individual. Consequently we attempt to acquire cooperative behavior using symbiosis coevolution and cooperative evolution.

View full abstract

The best quality, a price of the most small, and a short delivery date are always demanded from the market from an artificial thing (product and service). For this, it is requested that the design specification decided by the design phase be efficiently made an embodiment at stages of production. This means the conversion of the reformation and the conception is requested from the machining and the assembling method that is the center of the production activity. The function concept is introduced into the process design method as one aspect that deals with such a demand in this research. That is, it proposes a new idea and the frame to pursue the essence of an individual process from the function side, for the dependence of these functions to examine, and to give novelty to the process design.

View full abstract

Manufacturing activities in the industries are now deelply associated with advanced information technologies. A complex manufacturing space (CMS) is a formalization of a manufacturing system synthesized by a real system and a virtal system. In this paper, the interface betweeen human users and a CMS is mainly discussed, after the brief introduction of a CMS. Universal pointing is a key technology which enables a user to easily indicate various types of objects in a CMS with many input devices. Cyber annotation is a intension recording method to the indicated object by universal pointing. A prototyped system for universal pointing and cyber annotation demosntrates the image of user interaction in a CMS.

View full abstract