Modern supply chains are subject to a wide range of risks, such as demand uncertainty, natural disasters, and terrorist attacks. Considering supply chain design, this study investigates the logistics network design problem with facility disruptions that are caused by the above-mentioned risks. In light of recent events such as 2011 Tohoku earthquake and tsunami, how to design a robust supply chain network at risk of unexpected facility disruption has become a pressing concern. To find solutions to this design issue, in this paper we have formulated three types of stochastic supply chain disruption risk problem to offset potential losses from network disruptions. Thereat, the distribution centers are split into two types: reliable distribution centers (RDCs) and unreliable distribution centers (UDCs). RDCs are considered to be completely resilient to disruption and are available for both primary and backup assignments, while UDCs are affected by such disruptions and can be assigned as only a primary facility. To find the optimal solutions to these problems, commercial software is applied. Through numerical experiments, comparison among the models is taken place and prospects for risk management against facility disruption are discussed.

View full abstract

In this study, a viewpoint from greenness has been deployed into the logistics optimization associated with VRP. In terms of the foregoing method developed one of the authors, the idea has been extended so that it can cope with not only the generalized Weber model but also pick-up transportation model. Numerical experiment is provided to demonstrate the idea and to validate the proposed method.

View full abstract

Under recent global competition and short product life cycle, we need to cope with highly demanded problem-solving in manufacturing. With this point of view, this study provides a practical hybrid method for a hierarchical logistic network optimization belonging to a tactical level. It includes a new idea to solve VRP with time window and multi-objective analysis between economics and service of delivery. Numerical experiment has been carried out to validate effectiveness of the proposed approach.

View full abstract

Supplier selection is an essential task within the purchasing function of supply chain management because it provides companies with opportunities to reduce various costs and realize stable and reliable production. In this paper, we present fuzzy multi-objective linear programming (FMOLP) with three important goals: cost minimization, service level maximization and purchasing risk minimization. To solve the problem, we introduce an enhanced two-phase approach of fuzzy linear programming. First, the minimum achievement level desired by decision maker is set to the solution from max-min operator approach. Then, problem formulation is reconstructed to improve the minimum achievement level resulted from the first phase. Numerical experiment is provided to demonstrate our approach.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Hydrogen is attracted as new clean energy, and a fuel cell vehicle using hydrogen has been studied by many researchers . However a HFCV (Hydrogen Fuel Cell Vehicle) is many problems, in order to be a future popular vehicle. One of the problems is too long filling time. Currently, the fast filling time may be exceeded over ten minutes at hydrogen at pressure of up to 70 MPa. The filling time is about three times of that of a gasoline vehicle. A fast filling procedure for vehicles with compressed hydrogen storage tank was studied by Steffen Maus et al. The long filling time was caused by high temperature of the storage tank at pressure of up 70 MPa. The storage tank temperature rose up to about 60 K from the initial temperature . The storage tank temperature should not exceed the limit at 80 degrees Celsius. Therefore, hydrogen gas was cooled about –20 degrees Celsius before filling. An aim of our study is that cruise distance and filling time of the HFCV are as the same level as those of the gasoline vehicle. In this paper, hydrogen filling is controlled by PFC (Predictive Function Control) with FF (Feed Forward Control) and 1-DTM (1-Dimensional Thermodynamic Model) method. The storage tank pressure is controlled by PFC and FF. The storage tank temperature is calculated using 1-DHD which calculate the storage tank temperature from the storage tank pressure. To demonstrate performance of the proposed method, results of dynamic simulations using the proposed method are shown.

View full abstract

Recently, Just-In-Time (JIT) modeling methods, which includes locally weighted PLS, have attracted much attention because they can cope with changes in process characteristics. To construct accurate JIT models, it is very important to select an appropriate subset of samples. In past researches, samples were selected on the basis of the Euclidean distance and so on, however, the conventional methods cannot construct accurate models in some cases. In this research, a sample selection method to construct highly accurate locally weighted PLS models is developed.

View full abstract

There are many single-loop controllers in a plant. The interaction among whole control loops has never been analized. In this presentation, an approach appicable to real plants is proposed.

View full abstract

When a pair of control loops with negative RGA elements is adopted as the plantwide control configuration, it would become "seriously unconformable." Although it is operable and controllable for some time, it would easily suffer from getting trapped into "Trap Points" at setpoint changes from MPC/RTO or owing to disturbances, and, once trapped, it is hard to get out of it. Furthermore, there are mysterious cases called "Partly Unconformable Control Configurations" in which the conditions of "conformable / unconformable" depend on the operation zone of the process. The purpose of this paper is to propose some "Control Configuration Design Guides" to be referred by practicing process control engineers without the need of RGA evaluation.

View full abstract

Whenever product demands and/or production circumstances change, decisions about scheduling and operation of production systems are made based on accumulated know-how, taking account of various criteria such as production efficiency, product quality, safety and environmental load. The final goal of this work is the development of modeling method for supporting decision making that can deal with frequent and unexpected environmental changes. To attain the goal, we have to analyze the accumulated know-how and model the decision process made by experts. This paper first presents modeling of decision processes based on piecewise affine models by analyzing a set of operation data and then discusses the decision making process model.

View full abstract

[in Japanese]

View full abstract

Preview control is a control method that can be used when the future desired value is available to enhance the control performances to desired value changes. Though the feedforward gains for the desired value signal may be calculated using LQ optimization, it is difficult to have a specified response form, and it is not easy to suppress the overshoots in it. In this paper, we propose a method to design a preview controller to have a specified response form, and give an experimental example of a temperature control.

View full abstract

We have proposed a novel model identication methodology for ARX model based on transfer functions. The identifcation approach converts transfer functions to ARX models without any approximation. Model parameters of the transfer functions are directly estimated. In this paper, the proposed closed-loop identifcation method is successfully applied to an industrial chemical plant, which is controlled by model predictive controllers.

View full abstract

Recently, efficiency of energy conservation has been attracting much attention year by year. Hence, effective methods to evaluate energetic efficiency of existing control systems are required more than ever. To ease the evaluation of the energy efficiency, parameter identification of control systems in operation should be useful. However, the parameter identification may not succeed straightforwardly due to not only measurement noise but also disturbance. Disturbance can not be avoided and disturbance attenuation is often the major aim of feedback control where the existence of disturbance is essential. This paper thus proposes a method of parameter identification that uses multiple output signals and then extracts components orthogonal to the signal caused by disturbance. The effectiveness of the proposed method is examined by using a numerical example.

View full abstract

[in Japanese]

View full abstract

We often hear the apologetic word of "Beyond-Design-Basis" through the reports on the accident at Fukushima-Daiichi nuclear plant, which have been drawing criticism from the society. In the field of plant operation, instrument and control engineers should support operators to overcome unanticipated events by the design of human-system symbiosis systems. In this paper, human-system symbiosis model for the plant operation is proposed and operator's error in the Three-mile island accident are analysed according to this model. Results shows the importance of team operation, redundant sensor and top-down information with measured process value.

View full abstract

This paper presents a tele-control system constructed by a robot hand/arm and operator. The angle of the robot hand is controlled by the angle of the operator's finger, and the operator feels the environmental force, as detected by the touch sensor of the robot hand, constituting so-called bilateral master/slave control. The position of the robot arm is controlled by the potision of the operator's arm. In the experiments, the operator grasped and manipulated the object using a multi-fingered humanoid robot arm/hand by master/slave control feeling fingertip force by tele-operation between USA and Japan.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In 1984 Authors developed a new tension control scheme of hot rolled material as field control engineers of a maintenance section of a steel making plant. Here, looperless rolling scheme was developed using recursive least square method for the heavy material, and optimum control using modern control theory was applied to the light material. Later on, authors moved to academic area or the management division of the company. Authors reviewed the problems of the late developed scheme and evaluated revised schemes using developed experiment devices. The difficulties and counter measures for the nonlinearity of machineries, discrete time optimum control system would be discussed. Authors want to stress the needs of the control system development by young field engineers by themselves.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

This work presents a novel approach for the identification of plant model. Since performance deterioration is usually caused by changes in process dynamics, effective re-modeling is important. The proposed method is driven by events such as changes of set point, and uses the data of routine closed-loop operation. The effectiveness of the method is demonstrated through some experiments of the identification of a DC motor control system.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

With only one VSCMG (Variable Speed Control Moment Gyro), it is impossible to realize complete attitude control of spacecraft, but pointing control can be realized. In this paper, we show that the pointing control problem of spacecraft with only one VSCMG can be described as an LPV (Linear Parameter-Varying) model, and construct an interesting GS (Gain-Scheduled) controller to solve this problem. The controller guarantees both overall stability and H2 performance for a whole operating range. Moreover, distinct Lyapunov solutions are applied to design this controller. To demonstrate the effectiveness of the proposed method, we present a numerical simulation to compare the proposed method to a conventional one.

View full abstract

This study proposes a method to control flapping-wing micro aerial vehicle (Flapping-wing MAV) with less control inputs. Firstly, complex time-varying system for dynamics of Flapping-wing MAV is approximated by an averaged system. Then, utilizing an approach called virtual holonomic constraint, a control law with only two inputs (stroke amplitudes of left and right wings) is derived to stabilize six attitude parameters. A numerical simulation is finally conducted to show the effectiveness of the control law with respect to the averaged system.

View full abstract

Currently, unmanned automatic flight of helicopters is expected to the expansion of operational area. However, the small scale helicopters are very sensitive for disturbance and Ground-Effects. In this study, we derive the model of both Ground-Effects and motor dynamics through experiments of a RC helicopter. These models compensate both Ground-Effects and delay of the motor respectively. We use integral sliding mode control to attenuate the disturbance and modeling error. We improve the control dynamics inside the boundary layer by considering motor dynamics. Then, we verify the efficacy of proposed methods through experiments.

View full abstract

Unmanned aerial vehicles are very useful to inspect inside of disaster building such as fire, explosion and radiation. It is impossible to generate flight path prior the flying inspection. This paper proposes a flight path generating algorithm using information amount measured by a mounted visual sensor. The results of numerical simulations show that an aircraft keeps the center of corridor in the building.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In steel production, the continuous casting is a key process to dominate the quality of the rolled products and the efficiency of the steelmaking plants. In this talk, several application results of the control theory in the continuous casting will be shown. The Q-parameter controller design, an advanced control design technique, will be mainly described, which has been developed to suppress the periodical disturbance caused by the bulging of the solidified shell. Some experimental results on actual casters will be explained.

View full abstract

Computational fluid dynamics (CFD) simulators have recently been used for optimization in die casting and various other fields. However, solving an optimization problem with a CFD simulator (CFD optimization problem) has the issue of uncertainty in the evaluated values from the CFD simulations. Such problems are, of course, difficult to optimize compared to general problems. In this study, we propose an optimization algorithm that can search for good solutions to CFD optimization problems. We have applied the algorithm to the optimization of the sprue shape for die casting at an actual plant.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

As for the movement calculation of the interface which consists of molten metal and the air in the mold filling simulation, the VOF method is adopted by much software. The MARS (Multi-interface Advection and Reconstruction Solver) method has advantages the highly capture accuracy of the interface shape and an accurate estimate of the quantity of advection. The MARS method is applied to simulate the mold filling behavior. It becomes clear that the present simulation is carried out by enough accuracy and functions.

View full abstract

In order to fit in foreign markets, local support for the foundry equipments should be performed easily. In case of programmable controller (PLC) which is used to control foundry equipments, it is advisable to choose a PLC that is familiar by local engineers and receive support form the PLC maker without difficulty. One side, it takes time to apply existed PLC program to another maker PLC, and also it is easy to occur unexpected bugs. In this paper, a built global-support control system of automatic pouring machines is introduced.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In this study, a high thrust electro-magnetic proportional valve actuator with a few poles was designed and tested. The dimensions of this actuator were varied, thrust and electro-magnetic characteristics were examined by Finite Element Method analysis and experiment. As a result, it became clear that the thrust force of this actuator was above 2 times as large as that of conventional solenoids.

View full abstract

In this study, we evaluate characteristics of force control of stiffness-changeable finger using balloon actuator by predictive functional control (PFC) system. The predictive functional control is one of the model based predictive control (MPC) schemes, which predict the future outputs of the actual plant over the prediction horizon and compute the control effort over the control horizon at every sampling instance. In this paper, PFC control performances of the finger using balloon actuator are reported.

View full abstract

[in Japanese]

View full abstract

The steady-state and dynamic friction behaviors of pneumatic cylinders are experimentally investigated under various operating conditions. It is shown that three parameters in the modified LuGre model are affected by the pressures in the cylinder chambers and the measured dynamic friction behaviors can well be simulated by the modified LuGre model taking the dependences of the three parameters on the pressures into consideration.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract