“The digital transformation can be understood as the changes that the digital technology causes or influences in all aspects of human life” (definition by Prof. Erik Stolterman). In order to manufacture high value-added products and create a sustainable society, the digital transformation, based on advanced precision engineering, will be an urgent task within manufacturing systems.

This special issue consists of eight excellent research papers that focus on advanced precision engineering in manufacturing systems. All research papers were presented at the 18th International Conference on Precision Engineering (ICPE2020). The ICPE2020 covered various topics, including digital design, CAD/CAM technology, traditional cutting/grinding, non-traditional additive manufacturing, machine tools, measurement, robotics, control, and others. Held during the COVID-19 pandemic, it was a virtual conference that saw 189 papers presented in the oral session and 38 papers presented in poster session. The COVID-19 pandemic demands the innovation of current manufacturing systems with new concepts and methodologies, and the editors hope that the research papers in this special issue give us valuable information for the digital transformation of manufacturing systems.

All papers were refereed through careful peer reviews. The editors deeply appreciate the efforts and excellent work of all the authors and anonymous reviewers in realizing this special issue. Finally, we hope that future research on precision engineering in manufacturing will further contribute to the achievement of the Sustainable Development Goals (SDGs) of our global society.

In recent years, manufacturing companies have faced difficulties in securing sufficient production capabilities at factories because of many regional risks, such as natural calamities and epidemics. A production line should be designed to be reconfigured to adapt to various risks for satisfying its demands. This paper proposes a flexible and reconfigurable production line composed of a combination of line workers and multipurpose equipment called robotic cells. A robotic cell performs work (similar to a worker) using a programmable arm robot. The required tasks are allocated to workers or robots. However, it is difficult to design the line configuration and task allocation, because the number of combinations is large. Additionally, the production efficiency fluctuates depending on the correlations between the worker’s attitude, skill level, and allocated tasks. This paper describes a production-line design method using a genetic algorithm. The proposed method maximizes the availability ratio and minimizes the cost of the production line by considering the worker’s attitude toward the work.

Recently, measures to address environmental problems and resource problems have been strongly desired. Consequently, the energy saving of machine tools has been promoted at production sites. Thus, the efficiencies of motors and coolants have been improved for the energy saving of machine tools. However, replacing machines and peripherals is time-consuming and expensive. Therefore, we focused on the power consumption by the feed drive system and spindle of a machine tool during machining. Considering of the energy conservation in the feed drive system and spindle is practical and effective to existing machine tools. In this study, we propose the simulator that calculates the energy consumption of the feed drive system and spindle using the NC program. It is possible to calculate the energy consumption without actual machining and measuring. Using the simulation result, we can change the tool path and machining conditions, leading to a reduction in machining power consumption. This study devised a method for calculating the power consumption of the feed drive system using the NC program using only the reference power consumption of the machine tool. Then, the measured and simulated power consumptions for machining were compared to verify the validity of the proposed method.

Dynamically manipulating flexible objects using robots is difficult. Some studies have been conducted that only considered one type of object with known properties or that needed an identification test for string properties in advance. We propose a method to realize the dynamic manipulation of a string with unknown characteristics. We use a mass-spring-damper model for the string and repeat three steps: motion generation, real manipulation, and parameter estimation. The proposed method estimates the string properties to realize the motion objective via the real manipulation of the string. An identification test in advance was not necessary. In this study, we focus on swing manipulation. This can increase the motion energy of a string without a high-power actuator. After making a large swing, the robot can throw strings to a more distant target, such as a hammer throw. This motion is useful for explanation robots, rescue robots, and so on. We modified the proposed method to generate a swing manipulation. Then, we investigate whether swing manipulation can be performed by the proposed method and demonstrate its effectiveness via experiments with various strings with unknown characteristics.

Offset computation for expanding a polyhedral object by an offset radius is a fundamental geometric process frequently used in manufacturing applications. This process combined with the triple-dexel representation solid model has become popular because of its robustness and compatibility with parallel processing using a graphics processing unit (GPU). In parallel geometric processing, conflicts between threads must be avoided. Thus, we propose a novel parallel offsetting algorithm free from conflicts between threads. The triple-dexel model is a combination of x-, y-, and z-axis-aligned dexel models. Each dexel model is defined based on an orthogonal grid given on a coordinate plane. We subdivide the grid into several sub-grids of a fixed size in advance. For each sub-grid, a block of GPU threads is assigned. As each GPU thread always processes different dexel elements from the other threads in this method, no conflict occurs. Our research group has previously presented a parallel offset computation algorithm for a polyhedral solid model that also uses a triple-dexel representation model and a GPU. In the previous algorithm, the surface polygons of the model are classified into several groups in advance. The parallel offset computation of multiple polygon groups is realized by selecting groups of polygon in which the offset processing of the polygons does not affect one another. This selection process is time-consuming. Computational experiments were performed to analyze the performance difference between the current algorithm and our previous algorithm. In our experiments, the current algorithm achieved speedups of 1.4 times to 3.2 times compared to our previous offsetting algorithm.

Image measurement technology – widely used in present society – has made substantial progress. It involves processes such as image input, target extraction, and measurement of the extracted region to obtain information from an image. These processes are computationally intensive because they require a large amount of information such as complex features, which is often an obstacle to improving and speeding up image processing tasks. In contrasts, living organisms easily recognize their own surroundings in real time. In cognitive science research studies, for example, visual affordance studies have shown that organisms perceive and recognize their surrounding environment and objects from ambient light, which is formed by reflected and scattered light in the environment. By applying this natural mechanism to image measurement technology, it is possible to obtain the information necessary to recognize the surrounding environment by observing ambient light without necessarily detecting or recognizing the object. In this study, we propose a direct method of assessing the surrounding environment by capturing ambient light as luminance.

This study aims to build a machine scheduling method that involves the cutting tool management in parallel-type flexible manufacturing systems. These systems consist of multi-axis CNC machine tools and are equipped with an automated tool changer and a large-capacity tool magazine. The target scheduling problem could be described as a multi-objective parallel scheduling problem. We consider the availability of cutting tools stored in the magazine as so-called “machine eligibility,” and propose a two-phase scheduling method for tool allocation and job sequencing on machines to minimize the workload balance between machines and the total completion time. Two mathematical models for tool allocation are provided: a machine-eligibility-based model and an enhanced version of the model that considers each cutting tool. A series of computational experiments demonstrates the effectiveness of the proposed method. We also clarify the relationship between schedule performance measures and job routing flexibility in the system.

Planetary gear trains (PGTs) are widely used in many machines and are one of the most important mechanisms in hybrid and electric vehicles. Previous research, based on empirical knowledge gained from the automobile industry, indicates that high carrier-support stiffness and low ring-gear support stiffness are required to reduce ring-gear errors. Therefore, here, we evaluate the vibration characteristics of a PGT as a function of the support stiffness, which is varied by inserting urethane rubber into the driving shaft. We conducted experiments using a 2K-HV-type tester, which contains a coaxially rotating and revolving planet gear shaft based on a universal joint. This mechanism allows the observation of the inner workings of the mechanism with the use of a transparent acrylic carrier. We were able to detect the so-called “bounce” phenomenon consisting of a swaying motion when the rotation of the ring gear is reversed, and this result was confirmed by our internal observations of the mechanism. It is evident that the index of vibration increases due to the bounce phenomenon because the reversal of the ring gear causes a larger vibration than that of the carrier because the ring gear can vibrate without restraint, unlike the planet gear that is sandwiched between the sun and ring gears. Furthermore, the influence of the radial support stiffness of the driving shaft, load torque to the output shaft, and acceleration time of the reversing gear on the “bounce” phenomenon were evaluated. We found that a larger load torque corresponds to a greater difference depending on the acceleration conditions of the sun gear. During reversal, at the moment when the rotation speed is zero and rotation recommences, the ring gear exerts the maximum force, and the larger is the load torque, the greater is the effect of the difference in acceleration.

Recent product components have been designed as a combination of complex forms for advanced functionality. Such high degree-of-freedom forms are shaped by various manufacturing processes. In the industrial supply chain, the deviation of a manufactured form from its designed form must be verified to ensure product quality. Three-dimensional (3D) measuring systems are typically used for verification. To validate the product manufacturing performance accurately, a measurement procedure that can derive the measurands of forms with small measurement uncertainty is necessitated. For simple geometries such as flats, spheres, and cylinders, precise measurement and uncertainty evaluation methods have been established; however, those for complex forms are still being developed. In this study, measurement uncertainty is assessed by measuring a calibrated gauge and comparing the measured with calibrated values. Several gauges configured with basic geometric elements have been proposed as a reference for complex form measurements, and some issues remain. Some of the gauges do not sufficiently simulate the forms of actual product components, whereas other gauges are difficult to calibrate with small uncertainties. Herein, we discuss a possible approach for solving these problems in the metrology of complex forms using 3D measuring systems. First, a new gauge concept is proposed. The gauge is designed by extracting the complex features of the actual product form and segmenting the gauge’s form into various circular arc curvatures. The geometric parameters are well calibrated, and the measurable angular range of the curvature is not limited. A representative gauge based on this concept is described. Next, a robust and small-uncertainty calibration method for complex-form objects is described. The area encompassing the target cross-section is measured, and an envelope is generated on the probe tip mechanical surface from the probe tip centers to determine the measured surface. Finally, the calibration and measurement uncertainty evaluation for the geometric parameters of the new gauge are quantitatively presented. It is verified that the proposed calibration procedure is applicable to freeform measurements, and that both simple and complex forms can be measured within a 1.5 μm uncertainty.

Total internal reflection is an optical imaging technique for nanoparticle tracking and observation employing the scattered light from an evanescent field near the interface or reference surface. Generally, the nanoparticle behavior is the three-dimensional Brownian motion in an aqueous medium. The motion can be traced by an optical microscopy, but it cannot be traced by an electron microscopy technique. In the three-dimensional nanoparticle moving position, the X and Y positions are parallel to the surface, which can be traced by the general microscopy techniques. However, the height position Z of a nanoparticle perpendicular to the surface could not be traced without the longitudinal scanning method. Here, a novel method is proposed to investigate the 3D position of nanoparticles by applying multi-wavelength evanescent fields microscopy, which has a high spatial resolution in the Z-direction without longitudinal scanning. This paper focuses on the verification of measurement in the Z-direction. A piezoelectric actuator was employed to control the nanoparticle displacement in height Z. Standard polystyrene 100 nm particles were randomly adhered on a spherical tip that connected with the piezoelectric actuator. The spherical tip was essentially made from an optical adhesive (n = 1.348) with a refractive index close to the water for decreasing the unnecessary signal from the tip-self during nanoparticle observation in the water. The proposed method could obtain the multi-wavelength scattering lights from the observed nanoparticles by an 8-bit color camera with higher than 50 frames per second recording to investigate the 3D nanoscale tracking. The X and Y positions of nanoparticles were determined by the centroid of the scattering light intensities. The height Z was determined from the logarithm ratios between the detected scattering light intensities of both wavelengths. The measurement repeatability of the absolute difference in height between nanoparticles could be measured less than ±16 nm by using the proposed method. The penetration height measurability range was approximated at 250 nm from the reference surface.

In the cutter path computation for the five-axis machining, it takes much time to determine proper cutter postures in machining. We propose a novel method for calculating all possible cutter postures that can be used in machining mold’s surface without collisions in a short time. In determining the cutter postures for each cutting point, intersection detection between a line segment and a set of polygons is frequently carried out. Latest graphics processing units (GPUs) are equipped with a hardware called ray tracing (RT) cores dedicated to image processing in the 3D computer graphics. We use this RT core technology for accelerating the intersection detection and consequently reducing the time and cost necessary in the cutter posture determination. We also present the numerical experiments conducted to verify the effectiveness of the proposed method.

Elastomer end-milling has attracted attention for use in the small-lot production of elastomeric fragments because the technique is an applicable method for a large variety of materials and does not require the preparation of expensive and time-consuming moulds. In order to effectively utilize elastomer end-milling, it is necessary to ensure the machining accuracy of elastomeric parts machined through this technique. However, the control method of machining error in the elastomer end-milling has not been presented since most machining services of the elastomeric part are based on enterprise-dependent dexterities or know-how. The objective of this paper is to construct and utilize a machining error model for elastomer end-milling. A statistical model based upon physical states and machining conditions is introduced and investigated. In this paper, a framework for modelling the machining error in elastomer end-milling is also proposed. In the framework, the candidates of model variables are evaluated based on the preliminary experiments. Moreover, a statistical model is constructed by using the selected variables. Candidate variables are cutting conditions and predictable physical state variables such as workpiece deformation and cutting force. The framework is investigated by evaluating error prediction with the experimental results. An identified error model from limited machining cases can estimate the machining error of different machining cases. The results indicate that the proposed modelling method is capable of supporting to achieve model-based precision elastomer end-milling.

The objective of this research is to investigate an on-machine estimation method to achieve efficient and fast estimation of the fixturing force and workpiece deformation. The estimation enables us to visualize workholding states and improves machining accuracies of thin-walled parts. In this research, a systematic estimation method of workholding states which combines fixturing simulation and locally measured strain is proposed and evaluated. The proposed on-machine estimation method is evaluated in different workholding conditions (clamping sequences and fixturing forces). Estimated fixturing force and workpiece deformation for a clamped thin-walled workpiece were compared to the results from the engineering experiments. From the comparison, it becomes clear that the proposed method has the feasibility to detect improper workholding states such as insufficient fixturing force or excessive deformation.

In powder bed fusion (PBF), a type of metal-based additive manufacturing (AM) process, metal powders are deposited on a substrate and melted through selective laser-beam irradiation. Among AM processes, PBF yields excellent dimensional accuracy, and the built parts can be applied to molding dies and topology-optimized parts. Furthermore, PBF can be used to build porous structures. In this study, a highly functional die casting method was established using PBF, which involved a release agent supplied through the porous structure to the surface of the proposed die. The arrangement of the porous structure made it possible to apply the release agent to the deep groove, which is not possible using a spray as in conventional supply methods. The laser-irradiated area was visualized to confirm pore formation, and the building conditions of the porous structure suitable to supply the release agent were investigated. The resulting die casting characteristics were evaluated. Considering the obtained results, guidelines to build dies or molds for die casting with porous structures are recommended. The amount of release agent could be controlled at each position of the die casting die, and the melted Al alloy did not penetrate the porous structure. In addition, the obtained Al alloy casting did not exhibit any castability defects. Moreover, suitable control of the supply of the release agent enabled enhancement of the die casting characteristics.

Polishing pad conditioning is essential for achieving stable chemical mechanical polishing (CMP). While diamond disk conditioners (DDCs) are often applied, flexible fiber conditioners (FFCs) have been proposed as a new conditioning tool. FFCs are intended for roughening the pad surface appropriately by bundling fine wire fibers. Our previous study demonstrated the fine conditioning characteristics of FFCs for a hard urethane foam pad. In this study, the conditioning characteristics of an FFC are compared with those of a DDC. First, we evaluate the conditioning performance of an FFC using SUS fibers on a soft urethane foam pad. The result indicates that on a soft pad, the SUS-FFC can restore the pad surface asperities more finely, as confirmed via the stabilized number of contact points based on contact image and luminance value distribution analyses. Subsequently, for a metal-contamination-free FFC process intended for semiconductor CMP, we develop an FFC fabricated using polyether ether ketone (PEEK) and verify its performance via the CMP test of a silicon oxide film. It is shown that the hard pad can be conditioned using the developed PEEK-FFC; this implies that a stable removal rate can be realized immediately after pad break-in conditioning.

This study aims to improve the efficiency of free-form surface machining using a five-axis controlled machine tool and a barrel tool. The barrel tool has cutting edges with curvatures smaller than its radius; thus, its pick feed width is larger than that of a conventional ball end mill with the same tool radius. Therefore, the machining efficiency can be improved. Barrel tools can be effectively utilized in a five-axis controlled machine tool. When five-axis controlled machining, tool interference may occur, which should be avoided during actual machine operation. Additionally, a sudden change in tool posture adversely affects the quality of the machined surface. This paper proposes a method to obtain the cutting points that render cusp heights below the target value. A method for generating an interference-free tool path, in which the tool posture changes continuously, is also proposed. The usefulness of the developed methods was confirmed through machining simulations.