This study proposes a tool management method based on a mathematical model to optimize tool usage costs in a flexible manufacturing system (FMS), considering machine load balance. Typically, this involves multiple multi-axis machine tools equipped with large-capacity tool magazines. The proposed approach aims to minimize the total cost associated with tool usage by developing a mathematical model that assigns cutting tools to machine tools while accounting for load balancing. Tool usage cost includes expenses for purchasing and storing cutting tools. Numerical experiments demonstrate the effectiveness of this method and highlight its potential as an efficient tool management solution to reduce tool usage costs and enhance production efficiency.
We treat an issue of routing and scheduling for a parcel delivery system, which is composed of a single carrier truck and multiple identical drones. The carrier truck basically plays the role of a mobile depot of drones, and the drones perform the last-mile delivery of a parcel from the carrier truck at a stop to each of all the drone-served customers and of some flexible customers. The carrier truck also brings a parcel directly to such a customer that either can be or must be served by the carrier truck, i.e., it is either a flexible customer or a truck-served customer. The carrier truck delivers a parcel to a flexible customer if the direct delivery is more advantageous than the last-mile delivery by a drone. We propose a mixed integer program of this issue, which is an extended version of an existing formulation for a fixed truck route model of such a truck-drone system. The objective is to minimize the sum of the total duration over available truck stops and a weighted travel time of the carrier truck, i.e., it is to minimize a generalized makespan. We also demonstrate the solution quality of the proposed mixed integer program empirically by utilizing a mixed integer programming solver, and we observe an effect of the weight in the objective function and an effect of the existence of flexible customers in the numerical results.
This paper examines a multi-echelon supply chain optimization problem under simultaneous demand and supply uncertainty. A multi-budget robust optimization (MBRO) model is proposed to enhance the operational resilience of the supply chain. Compared to standard deterministic and single-budget robust models, the proposed MBRO model introduces two separate uncertainty budgets to control demand-side and supply-side deviations, allowing different levels of protection on each side. The model is formulated as a mixed-integer linear program and compared against three alternative approaches: a deterministic model, robust possibilistic programming (RPP), and cardinality-constrained robust optimization (CCRO). Uncertainty is represented through bounded percentage deviations combined with budget constraints that limit the number of parameters exposed to worst-case conditions simultaneously. A series of computational experiments are conducted using a real-world supply chain dataset from the fashion and beauty industry. Four different uncertainty scenarios are considered: nominal, high demand, low supply, and high uncertainty conditions. The results show that the MBRO model performs better than other comparison models in terms of demand fulfillment, uncertainty mitigation, and total profit. Specifically, MBRO achieves up to 33–68% higher profit compared to other models, measured as the relative gap with respect to the MBRO solution. Also, MBRO delivers up to 73% greater uncertainty mitigation compared to RPP and approximately 6% greater relative to CCRO, while maintaining prescribed service levels. Sensitivity analysis also uncovers nonlinear trade-offs between robustness budgets and profit, showing that fine-tuned protection works better than simply maximizing it. These findings highlight the benefit of independently controlling demand and supply robustness. The proposed framework also offers decision-makers a practical and adjustable approach for designing resilient supply chains under heterogeneous uncertainty.
Goods-to-Person (GTP) order picking, in which the non-value-added activity of traveling to storage shelves can be replaced by robots, has been widely studied. In GTP order picking systems, relieving pickers from walking is expected to improve work efficiency. However, because of the division of labor between pickers and robots, waiting times may occur for both parties, potentially leading to reduced utilization if appropriate operational strategies are not implemented. Therefore, this study proposes a method for determining the picking sequence and conducts numerical experiments to analyze the trade-off relationship between picker and robot utilization for different numbers of robots. The results demonstrate the importance of evaluating the number of robots from the perspectives of both picker and robot utilization. In addition, an operational strategy incorporating buffer racks is examined, and the conditions under which buffer racks become effective are clarified. Specifically, buffer racks do not provide benefits under operating conditions identical to those without buffer racks. However, they become effective when multiple robots operate simultaneously in front of the buffer racks or when case returns are conducted at night or during picker breaks. This study aims to achieve an operational strategy that effectively utilizes both the picker and robots in warehouse operations.
Fast charging congestion and the environmental value of electricity vary substantially by hour, creating a need for operational control that is congestion aware and aligned with cleaner electricity. This study proposes a scheduling based pricing design for en route fast charging on a trunk road network. The day is discretized into equal time windows, and charger availability is represented by capacity constraints defined over time windows, which endogenously generate charging delays. Drivers jointly choose routes and charging stations by minimizing a generalized cost that includes travel distance, expected waiting time, an exogenous electricity tariff, and a station and time window specific surcharge. The surcharge is optimized to minimize a system level social cost composed of travel cost, waiting cost, and the social cost of electricity consumption, while treating the resulting payments as transfers. The hierarchical problem is formulated as a mathematical program with equilibrium constraints enforcing Wardrop equilibrium. Numerical experiments on an illustrative middle mile network compare (i) no surcharge, (ii) congestion only pricing, and (iii) a dual objective pricing design that also encourages charging in hours with lower electricity social cost, using a Kansai area spot price profile as a transparent proxy. Relative to the no surcharge case, congestion only pricing disperses charging across stations and reduces total waiting with negligible additional travel. When low emissions alignment is additionally targeted, charging shifts toward lower price hours and waiting is further reduced, with a modest increase in travel distance. Overall, the framework demonstrates how implementable station and time window surcharges can jointly mitigate fast charging congestion and promote renewable-aligned charging in corridor networks.
For the nurse scheduling problem, we propose a hybrid method that combines mathematical programming and machine learning. In the method, a mixed integer programming (MIP) problem firstly provides a shift schedule, which, however, is not always acceptable to a chief nurse because there often exists a hidden condition; this is one of the major practical difficulties in nurse scheduling. The method secondly trains a deep neural network (DNN) model with practically acceptable schedules. As a result, the DNN model automatically fixes an input schedule, which is initially the one obtained from the MIP problem, so that it is acceptable. To evaluate the capability of this method, we perform a simulation involving another MIP problem, called the virtual chief nurse problem (VCNP). The VCNP implements virtually hidden conditions, such as the incompatibility of nurses, and its solutions are used as target data for training a DNN model. Through this simulation, the DNN model learns the hidden conditions, thereby modifying schedules appropriately. This demonstrates the promise of our hybrid approach. We also conducted an experiment in collaboration with a hospital’s nursing department. In the experiment, a chief nurse repeatedly fixed the initial schedule generated by the MIP solver, and a sufficiently good schedule was obtained after a few iterations. These results suggest that the proposed method can effectively capture hidden practical constraints and improve the overall acceptability of nurse schedules.
Improper sit-to-stand (STS) motion imposes excessive loading on the knee joint and surrounding muscles, yet access to professional physical therapist (PT) supervision remains limited due to a global shortage. This situation highlights the necessity of developing autonomous training systems capable of delivering motion guidance equivalent to that provided by PT. To address this issue, the present study proposes an optimized STS training protocol along with a portable feedback device for home-based rehabilitation. The optimized protocol was established by measuring STS motions experimentally at different speeds and performing zero moment point (ZMP) balance analysis. These analyses informed the development of a predictive model capable of estimating the STS trajectory based on fast and slow-motion characteristics. A portable device integrating a nine-axis inertial measurement unit (IMU) and an asymmetric vibrotactile module was developed to provide real-time, directional haptic feedback emulating guidance from a PT. The proposed system generates individualized motion recommendations using the predictive model and guides users toward the optimized protocol through iterative training. Experimental evaluations with human participants demonstrated three main findings: (1) the proposed asymmetric vibration device provides effective and intuitive interaction during STS exercises; (2) the STS motion model exhibits high reliability in predicting appropriate movement patterns; and (3) training performance progressively improves with repeated practice sessions under various modes using the wearable device. These results confirm the feasibility and effectiveness of the proposed system for home-based STS rehabilitation.
Double forging operation mechanism refers to the use of two manipulators to jointly hold a forging piece within a forging workshop to complete the forging process. However, during the lifting of the forging piece by the dual-machine clamping mechanism, the symmetrical arrangement of the forging manipulators results in different trajectories of the dual clamps. The coupling between the two machines and the clamped forging piece significantly affects the stability of the system. This paper starts from the working characteristics of the dual-forging operation mechanism, analyzing the working requirements of the suspension mechanism and the forging piece under the coupling conditions of the dual-forging operation mechanism. Considering the constraint, the constraint relationships between the buffering mechanism and the dual-forging operation mechanism, a configuration method for the buffering mechanism based on redundant constraints is proposed. Finally, the main motion mechanism of the dual-forging operation mechanism is constructed under the condition of satisfying the functionality of the single-forging operation mechanism, resulting in multiple combinations of dual-forging operation mechanism configurations. This configuration method alleviates the synchronization and coupling issues in the lifting motion of dual-forging operation mechanisms from a mechanical principle perspective, and improves the synchronization characteristics of the system, providing guidance for the design of dual-forging operation mechanisms.
In practical working environments, quadruped robots encounter significant challenges due to terrain variations, with sloped terrain being among the most common and difficult to navigate. Designing robotic structures specifically for sloped terrains offers an effective solution to the challenges posed to conventional machinery and human labor. Inspired by the goat's superior balance and climbing ability, largely attributed to its body-limb structure and efficient locomotion, the body structure, transmission system, and center of mass adjustment mechanism of a goat-inspired quadruped robot were designed based on caprine anatomy. RecurDyn V9R2 was used to simulate the multi-body dynamics of the quadruped robot during slope locomotion. The simulation results show that the three-segment body generates periodic inter-segment rotation and torsional-spring torque during slope transition, indicating its potential to attenuate low-frequency body pitching and terrain-transition disturbances. In addition, under a prescribed horizontal disturbance on a 15° straight slope, the robot with the active COM slider recovered its forward motion, whereas the fixed-slider case lost stability.
Grinding is one of the practical machining processes for manufacturing CFRP components without causing damages such as delamination and burr-formation. Although the design flexibility of grinding wheels can contribute high productivity and precision in CFRP machining, chip loading in grinding process still remains a significant problem. The chip loading reduces number of effective cutting grains on the wheel surface, resulting in higher grinding forces and a higher risk of the damages on machined surfaces. Therefore, this study proposes in-process dressing by pulsed laser irradiation to suppress the chip loading and enable continuous long-term grinding CFRP without suspending operation of grinding. Thermosetting CFRP (CFRTS) and thermoplastic CFRP (CFRTP) are used as work material in experiments for investigation on transitions of chip loading occurred in grinding, and influences of operation parameters are examined. Then, the pulsed laser dressing is employed for the grinding CFRP under various conditions to investigate the feasibility and effects of the proposed dressing method. Our findings indicate that in-process laser dressing effectively reduces grinding force and chip loading for both matrix materials. Additionally, CFRTP exhibits higher grinding forces than CFRTS and is more prone to cause chip loading. Therefore, PLD is more effective for the grinding of CFRTP.
In ball-end milling of curved surfaces, the visual quality of machined surfaces is influenced not only by cutting conditions but also by the arrangement of cutter location (CL) points generated by CAD/CAM systems. However, the relationship between CL point arrangement and appearance-based surface quality has not yet been fully clarified or quantified. This study investigates the effect of the shift in CL point arrays between adjacent cross-feed tool paths on machined surface quality using the luminance difference Ld, a quantitative index representing visual surface perception. A unidirectional scanning-line tool path was considered, and the shift amount is defined as the displacement of CL points in the feed direction, normalized by the segment length determined by the tolerance. A geometric model was developed to derive the relationship between shift amount and the tilt angles of cusps formed between adjacent tool paths. Based on the cosine law of illumination, the resulting luminance variation was predicted, indicating that larger shifts increase luminance contrast on the machined surface. Cutting experiments were conducted with different shift amounts, and luminance difference images were obtained. The results revealed a strong positive correlation between shift amount and Ld, demonstrating that larger shifts lead to greater luminance variation and deterioration of appearance-based surface quality. These findings show that the shift amount of component points serves as a practical parameter for predicting and controlling visual surface quality. Moreover, the most favorable appearance is obtained when component points of adjacent tool paths are aligned perpendicular to the feed direction.
An effective approach to tooth-profile expression using normal polar coordinates has been proposed by Watanabe in 1949, but not applied to practical design. This method represents the tooth profile using the variables obtained by drawing a normal line from any point on the tooth profile and finding the intersection with the operating pitch circle—that is, the length of the normal line from the point on the tooth profile to the intersection, the inclination angle of the normal line, and the coordinate value of the intersection. Previous research has shown that it is relatively easy to obtain and analyze the generated tooth profile by applying normal polar coordinates to external gears. In this research, this method was applied to internal gears including non-involute gears, and a method to analyze the trimming problem was proposed. First, for each rotation angle of the pinion cutter, the distance from the line passing through both centers of the internal gear and the pinion cutter to the farthest point on the pinion cutter and the closest point on the internal gear was calculated. These values could be calculated from the variables of the normal polar coordinates. Next, these two values were compared for each rotation angle of the pinion cutter, and if there was a point at which the distance to the farthest point on the pinion cutter was greater, it could be determined that trimming would occur. This method was applied to a specific arbitrary tooth profile to obtain the limit number of pinion cutter teeth at which trimming would not occur when machining an internal gear with a specific number of teeth.
Surface texturing has been widely used to improve lubrication performance under oil-lubricated conditions; however, its effectiveness is generally limited to low contact pressures, and extreme-pressure (EP) additives are typically required under severe conditions. In this context, nanostripe-inducing structures (NIS) formed by depositing multilayer films of a soft metal (Ag) and hard metal (Cr) on a microscale ridge-array substrate offer a potential approach for reducing friction through material and structural design without relying on EP additives. In this study, the friction and wear properties of Ag/Cr NIS substrates were investigated under oil-lubricated conditions using a Si3N4 ball in reciprocating sliding tests at normal loads of 0.3 and 1.0 N. The effect of sliding direction on ridge orientation was examined, and the sliding speed was varied on formed wear scars. Surface characteristics were analyzed using confocal microscopy, atomic force microscopy, current-distribution measurements, and laser-induced breakdown spectroscopy (LIBS). The results show that the friction coefficient during parallel sliding increased with the sliding cycle, whereas it remained lower during orthogonal sliding. When the sliding speed was varied, the friction coefficient increased with speed, particularly under low friction. The friction coefficient in the orthogonal direction was 30%-50% lower than that in the parallel direction. At a 1.0 N load, the wear depth in parallel sliding was approximately twice that in orthogonal sliding. Current-distribution measurements and LIBS indicated partial transfer of Ag onto the ridge tops, thus suggesting that solid lubrication reduced friction and wear.
