This paper proposes an improved evolutionary algorithm with parallel evaluation strategy (EAPES) for solving constrained multi-objective optimization problems (CMOPs) efficiently. EAPES stores feasible solutions and infeasible solution separately in different populations, and evaluates infeasible solutions in an unusual manner, such that not only feasible solutions but also useful infeasible solutions will be used as parents to reproduce the populations for the next generation. The EAPES proposed in this paper ranks infeasible solutions based on the scalarizing function named constrained penalty-based boundary intersection (C-PBI), which is determined by objective function values and a total constraint violation value. Then, this paper investigates the performance of the C-PBI-based EAPES to search for Pareto-optimal solutions compared to the non-dominated sorting genetic algorithm II (NSGA-II) and the previous EAPES without using C-PBI. The C-PBI-based EAPES with a well-tuned parameter is most capable to explore Pareto-optimal solutions with good diversity, spread, and convergence to the true Pareto front. The C-PBI-based EAPES assigns bad rank to the infeasible solutions that are expected away from an unknown Pareto front, and does not store such solutions. Thus the C-PBI-based EAPES exhibits a higher searching capability than the previous EAPES by evaluating infeasible solutions in an appropriate balance between objective functions and total constraint violation.

Nonlinear friction is the main stabilization error source of missile’s seeker. In this paper, the frictional torque of a gimbaled-mirror seeker under changing environment temperature is analyzed. The torque is consisted of rolling friction due to elastic hysteresis, sliding friction due to pivoting and friction torque due to lubricant viscosity. Based on the theory of thermo-elasticity, axial and radial coupling thermal deformation between precision bearing and its installation structure are both included in the analysis model. Coupling deformation dramatically changes the actual contact angle and axial force of bearing, which consequently change the value of frictional torque. Temperature effect on viscosity of lubricant is also studied and improved Walther Equation is used to fit the viscosity- temperature relationship of lubricant. A measure system based on electrical measure method was established and temperature test chamber was used to simulate the temperature changing. Experiment results have shown that this model is accurate for friction torque calculation of gimbaled seeker under changing environment temperature.

An attempt of experiment / simulation integrated design optimization is performed for the blade airfoil shape of a vertical axis wind turbine. A variable fidelity Kriging surrogate model approach is utilized in this integrated design optimization, in which a surrogate model is constructed from both high-fidelity and low-fidelity sample points information. In this research, the high and low fidelity performance functions are respectively defined from experimental and numerical evaluation methods. In the numerical evaluation method, 2D steady computational fluid dynamics (CFD) simulations are performed for the evaluation of an airfoil shape. An optimal airfoil shape is explored on the accurate variable fidelity surrogate model which is constructed in 9 design variables space representing various blade airfoil shapes. The validity and effectiveness of the present integrated design optimization are discussed by comparing with another approximate optimal airfoil shape which is obtained only by CFD computations. The optimal design obtained by the developed approach showed 8% larger performance value in the experimental evaluation compared with the approximate optimal design. More efficient shape optimization can be realized by including the low-fidelity information, whose functional trends are utilized to construct accurate surrogate models.

Computer-Aided Process Planning (CAPP) systems have become essential in manufacturing environments to integrate the information between CAD and CAM systems, and to automatically generate the NC code from the CAD model. Though the future of these systems seems to belong to the use of Artificial Intelligence to create knowledge-based algorithms which emulate human decisions, the CAPP systems based on feature recognition and model matching, which use databases of previously known mechanical components to generate new process plans, are also a very interesting option due to their accuracy and smaller development costs. Many researchers have proposed different kind of feature recognition algorithms before. However, these algorithms are usually application-dependent and require external codes to identify the features of wireframe models. This paper proposes a new methodology for shape recognition and model matching stages which improves the accuracy of the recognition tasks, uses solid models instead of wireframe models and can be successfully applied to any kind of part. The methodology is based on an original coding system that links the geometric information extracted from the CAD model with the features of the part by means of an identification sequence which is detailed in the text. Also, a score system has been created for the model matching stage. The obtained results show that the system presents high accuracy in shape recognition, feature identification and model matching tasks, even when the analyzed part is similar to the ones in the database. In addition, quantitative geometric data is also extracted from the CAD model on behalf of future steps of the CAPP system, such as the NC code generation stage. In contrast to other systems, this methodology can be easily applied to the industry since it makes use of the CAD model only.

Considering that humans perform handwriting task with reduced powers by contacting elbow or wrist on a table, it is reasonable to deem that manipulators can save energy and simultaneously accomplish writing tasks precisely like humans by bracing intermediate links such as elbow or wrist. First this paper discusses equation of motion of robot under bracing constraint condition, based on the robot's dynamics with constraint condition including motor dynamics. Then a control method to utilize the constraint dynamics is proposed to control simultaneously bracing force and hand's trajectory in work space. Even though the model used for the simulation analyses is simple four-links manipulator, the simple structure can help understand even more clearly the effects got by bracing part of the manipulator. This paper demonstrates the merits of the strategy to utilize bracing by comparing the contacting motions with non-contacting motions, including ; (1) the energy consumption can be reduced; (2) the hand trajectory tracking becomes accurate; (3) there is an optimum contacting point that minimize the energy consumption on condition that trajectory-tracking task be given to the hand.

This paper will describe and simulate the dynamic response of an innovative suspension system for vehicle seat. In which, the isolated object is supported by a wedge mechanism which comprises a horizontal spring, a roller and a wedge. Besides, the dynamic stiffness of system is corrected by an auxiliary mechanism (AM) including a roller (follower) which always contacts the circular surface of the cam via another horizontal spring. Hence, this seat suspension prototype not only broad the excitation frequency region but also remains the load bearing capacity and is named by “model with AM”. The dynamic stiffness-configurative parameter relationship is presented. Then, a virtual mechanical prototyping model of the system is created by using SOLIDWORKS software. Next, by applying dynamic analysis in COSMOSMotion module of the Solidworks product family, the real behaviour of system is tested and evaluated without using the traditional build-and-test method. With this way, it can increase the quality of product by reducing manufacturing cost and error. The simulation results show that the model with AM outperforms the linear counterpart. Besides, these results furnish a useful insight for the design and manufacture of the seat suspension system for vehicle.

In this study, the cognitive load that occurs in assembly works requiring human-robot collaboration is investigated. It is aimed to detect the effects of the workplaces designed with the approach of extended cognition on cognitive load and gender differences in cognitive load. Two workplace designs were used to investigate the task load that occurs in human-robot collaboration works and the effect of extended cognition on cognitive load. In both designs there was a sedentary workplace and same assembly work which requires human-robot collaboration. Differently, for the experimental group, the manual drawing was integrated to work table to detect the effect of extended cognition. For both groups task load is measured by using NASA-TLX (Task Load Index) method. Experimental results showed that the design with the extended cognition concept helps to reduce mental task load for male participants which is an important issue for yield, work safety, and quality in human-robot collaboration works. Experimental results also showed that there are no significant differences in cognitive load between genders. Unlike other studies, in this paper, human-robot cooperative works are examined according to gender difference with a point of extended cognition. The importance of the workplace design for human-robot interaction is presented.

The goal of this study is to investigate the effect of actuator position on the theoretical energy consumption of an electrically powered bipedal robot. Specifically, the study focuses on the legs of the afore-mentioned robot because their function is limited, as they are primarily used for either locomotion or standing, and thus, this is suitable for optimization. Furthermore, the hip and knee joints constitute the most powerful joints, and thus, these are studied first. A mathematical formulation is adopted to represent the relationship between the actuators and the robots joints. A genetic optimization is used to minimize the energy loss due to motor winding resistance and no-load torque by altering the position of the actuator with respect to the joints. The results indicate that the energy lost during typical motions can be reduced by up to 30% by using bi-articular actuation. Additionally, most benefits can be realized by adding only a bi-articular actuator between the hip roll and knee pitch axes.

A bolter miner is a new type of mining machinery with cutting and anchoring functions. Owing to its structural diversity, the cutting head performance of a bolter miner varies. However, there is still no effective method to evaluate the cutting head performance. Based on multi-aspect factors influencing the cutting head performance, the main factors such as the load fluctuation coefficient, mean torque, and specific energy were taken into account. Combined with the characteristics of multi-criteria decision-making approaches and the superiority of game theory in calculating the weight of the cutting performance indexes, an evaluation method based on the game-theory-extenics model is proposed. In addition, by combining the geological conditions of a particular coalmine, three pick layout schemes for the cutting head are compared, the final one of which was chosen through a performance evaluation. The results were successfully applied to the first shield bolter miner in China, and good results were obtained for engineering application.

In hardware/software (HW/SW) co-design, HW/SW partitioning is the most important step since it determines which components are implemented in hardware and which are implemented in software. Since most of HW/SW partitioning problems are NP hard, heuristic methods have to be utilized to solve them, especially for the large size problems. GPU-based heuristic methods to accelerate HW/SW co-design are a promising way to reduce run time. However, the existing methods cannot deal with very large embedded applications because of GPU resource limitations. This paper presents a method to overcome the GPU resource limitations for very large partitioning while keeping a reasonable runtime. First, at the stage of computing the costs of the candidates, we propose a fast method of 2-flipping computing for very large HW/SW co-design. Our method is also general and can deal with both odd and even numbers of nodes. More importantly, our method avoids utilizing double-precision arithmetic units, which are scarce resources in GPU architecture. Second, since the GPU is constrained by memory limitations and the costs of candidates cannot be directly stored in the GPU's global memory, we present a time-space tradeoff strategy to break memory limitations for very large HW/SW partitioning. In this way, the following steps can be run under the constraint of GPU's memory limitations. Third, an in-place removal of infeasible solutions is proposed to reduce the overhead of global memory by half when the neighborhood is compacted. Fourth, when evaluating the tabu status of feasible candidates, we present a bitwise representation of tabu status to minimize the transfer overhead. Finally, we conduct a number of experiments. The results show that the proposed 2-flipping method of single precision data types works well. The results also demonstrate that the proposed approach expands the number of nodes of the task graph from 10,000 to 30,000 under the limitation of the GPU's global memory of 6 GB. The correlations between compression intensity and solution quality are analyzed to ensure the fairness and soundness of our method. Our work is general and can provide guidance for other applications.

During the process of kinematic structure enumeration using graph theory, isomorphism identification of graphs is an important and complicated problem. In this paper, an improved hamming number method (IHNM) was proposed to detect isomorphism for kinematic chain with multiple joints. Firstly, it is difficult to deal with multiple joints in mechanism analysis and synthesis, so the joint-joint matrix is used to describe the kinematic chain. The matrix can uniquely represent the kinematic chain. Then, the improved Hamming matrix is obtained by the improved hamming number method. The joint Hamming number, chain Hamming number and joint Hamming string were introduced, which can determine whether the kinematic chain is isomorphic or not. Eventually, the examples demonstrate that this new method can be accurately and effectively performed.

This paper considers a two-machine re-entrant flowshop scheduling problem in which there are two classes of jobs with different urgencies, i.e., urgent jobs and normal (not urgent) jobs. The objective of this problem is minimizing total tardiness of one class of urgent jobs and maximum completion times of the other class of normal jobs. To solve this problem, a lower bound and several heuristic algorithms for the problem are proposed. To evaluate the performance of developed algorithms, computational experiments are performed on randomly generated problems, and results are reported with analysis. The suggested algorithms show better performances compared with the ones used in real manufacturing systems in terms of solution quality and computation time.

This paper describes a proposal on a new method that enables the appropriate and practical decision making in bidding before submitting a proposal to participate in the bidding for a large-scale international infrastructure project. Such a method has been demanded by potential bidders in order to overcome the recent trend that Japanese companies have given up in participating in such projects in spite of opportunities to extend their business in highly potential overseas market. For this, we have established a comprehensive evaluation method for risk assessment with reference to actual experiences of one of the leading Japanese consultants in establishing the basis of the proposed method, and conducted a post-evaluation to verify and validate the proposed method based on assumptions for the Suvarnabhumi Airport Rail Link (ARL) construction project in Thailand. Based on the assessment, we have obtained a positive result to ensure the effectiveness and the validity of the theory assumed as well as the proposed method, with suggestion for further improvement.

Half engagement is a state that metal terminals are mated, but mechanical housings are not mated completely. When a connector is in such state, the possibility of connector separation is high even for a small external disturbance. It may cause a serious malfunctioning to the parent product. During manufacturing, however, it is difficult to detect the state because electrical functions still work properly. Lack of feeling of mating is known as the primary cause of the state. Normally the feeling of mating is unmeasurable. Therefore, it is necessary to quantify the feeling of mating to prevent the half engagement. To this end, in this paper, we have proposed a new evaluation approach for the half engagement prevention. We have adopted a probabilistic structure based on reliability theory and applied the logistic regression method as a scoring tool. Moreover, we have introduced two new evaluation factors to increase the accuracy of state detection. We have verified and compared the performance of our proposed approach to that of the conventional approach which has a deterministic structure. Throughout extensive experiments and analyses with real automotive electronic connectors, we have confirmed that the proposed approach is superior to the existing one in terms of the accuracy of state detection.

Customer demand is crucial in production planning and capacity expenditure. In fact, variation of production rate with respect to customer demand affects the performance of a production system due to the existence of relations between processes. Simulation model has been established in this study to represent a production line in aerospace part-manufacturing company. Different production rates are simulated to examine the impact from the circumstances only emphasizes on maximization of availability or equipment utilization. On the other hand, bottleneck process with critical utilization is introduced to examine interrelationship among processes. The results show that higher equipment utilization or availability ratio as promoted by Overall Equipment Effectiveness (OEE) is not always Lean especially when the blocking phenomena occurs and relations between processes has been ignored. High availability is preferable under OEE although it consumes excessive resources and results in unnecessarily higher throughput to fulfill particular demand. Besides, bottleneck process could lead to lower equipment utilization of customer processes that it connects with. The so-called interrelationship between processes is invisible under OEE which focuses on individual process. In order to justify the findings, performance ratio under OEE is computed and it shows that delay propagation of bottleneck process is not quantified. None of the increase in waiting time and lead time of each entity spends is considered in OEE. The hidden wastes exist because OEE promotes maximum resource utilization without considers demand required and on-time delivery of entity. In short, it is important to consider effectiveness of both entity and equipment by respecting the customer demand and impact of interrelationship between processes onto the entire production system.

Recent industrial globalization has increased importance of functional verification in product design considering various ways of usage. In particular, for shortening development time, it is desired to perform functional verification and necessary modification of the design plan in the conceptual design phase. From this point of view, a framework for such individual functional verification that can be applied to the conceptual design phase was proposed. In this framework, functions, behavior and physical phenomena caused by a way of usage are modeled by function decomposition tree, Petri net, and cause-and-effect graph, respectively. They are then integrated as an augmented Petri net, and analyses for verification are performed by qualitative simulation on the integrated model. Although fundamental feasibility of this framework was illustrated, there are issues to be discussed further for application to actual product design. A serious one is behavior modeling. In the Petri net behavior modeling, detailed ways have not been discussed and the modeling is fully dependent on the designer. The resultant model varies depending on the designers even if he/she has perceived the behavior itself the same, and verification result may be inaccurate/inadequate if the model is inappropriate. It is desirable to reduce the degree of this dependence as much as possible. In addition, it is difficult for unpracticed designers to represent behavior of the design plan as a Petri net, since, in general, designers are not familiar with Petri nets. It is necessary to make this behavior modeling easier for designers. This paper provides a detailed way of the Petri net behavior modeling so that any designers can generate a standard behavior model which works in the framework as successfully and easily as possible. Each behavior is described by a unitary Petri net and the behavior of the design plan is described as a modularly-structured Petri net composed of those units. The modular structure is beneficial to automatic generation of the Petri net behavior model from the behavior model which has been generated by the designer in advance as a flowchart, which most designers are familiar with. Detailed ways of generating a Petri net model from a flowchart model are also given. This is helpful for designers to generate the Petri net model. The proposed method was applied to an example and its effectiveness was proven.