The portfolio manager is responsible for the tactical and operational management of concurrent initiatives that share the same resources. Sharing resources considerably increases complexity by creating resource dependencies between tasks, especially in highly uncertain environments. Therefore, resource allocation in environments with multiple projects is challenging for portfolio managers. According to previous research, classical mathematical approaches effectively address resource and project scheduling issues with low complexity or in a comparatively static environment. However, they have limitations when characterizing interdependencies in multi-project environments, working with uncertainty, and adapting the analysis to dynamic change. This study proposed a simulation method for human resource allocation for multi-project management in a project portfolio system with resource buffers. We have outlined how resource buffers can be allocated in multi-project management and how this impacts portfolio performance. Different quantities of resource buffers have been evaluated in various situations, considering varying degrees of uncertainty and complexity, to determine their impact on performance. In addition, the effect of varying resource buffer sizes on completion time reduction has been studied, which may benefit portfolio/project managers in practice.

Smart additive manufacturing (AM) is a complex process that integrates several digital technologies in design, planning, manufacturing, control, and services to make them more responsive and adaptable. Smart AM has demonstrated the ability to provide excellent digital technology for implant, tissue, or organ printing in the medical field. An anterior cruciate ligament (ACL) tear is an injury that occurs during strenuous physical activity. One of the implants utilized in ACL reconstruction is an interference screw. The polymers used to make an interference screw for ACL surgery are typically bioresorbable polylactic acid (PLA). Investigations on the effects of process parameters such as nozzle diameter and printing speed parameters on interference screw printing properties at 90° build orientation are still limited. Smart AM technology-related materials and processes are the primary areas of this study. This study evaluated the effects of printing speed and nozzle diameter settings, as well as the mechanical, biological, and physical characteristics of PLA interference screws printed at a 90° build orientation. The interference screws were 3D printed using material extrusion-based AM. The dimension deviation was 0.04 to 0.33%, and the density was 1.1 to 1.25 g/cm³. The A, B, and C interference screws were torque tested for all in a good clamping area, while the D-interference screw was in a clamping area. The weight loss of the interference screws in the biodegradation profile decreased after five weeks. The optimal process parameters were 50 mm/s printing speed and 0.3 mm nozzle diameter. The findings contribute to the production strategy of 3D-printed interference screw-based polymeric materials. It can be helpful to understand how different manufacturing parameters function and how those parameters affect the quality of the final product.

Hammer forging is a traditional manufacturing technology to produce high stiffness product and is widely used in heavy industries. In the hammer forging, the product is produced through several blows. Unlike forging using mechanical press, the product is produced by the energy in the hammer forging, and consequently the energy distribution in blows plays a crucial role for producing highly accurate product. Conventionally, the energy distribution is determined by the trial-and-error method. This paper proposes a methodology to determine the optimal energy distribution using numerical simulation and design optimization technique. It is important to minimize the total energy for energy saving and the risk of crack for product quality. To produce the highly accurate product, underfill should be avoided, which is handled as the design constraint. Therefore, multi-objective design optimization is performed to minimize the total energy and the risk of crack without underfill. The numerical simulation is so intensive that sequential approximate optimization that response surface is repeatedly constructed and optimized is adopted to identify the pareto-frontier between the total energy and the risk of crack. It is clarified through the numerical result that the proposed approach can determine the energy at each blow effectively. The experiment is also conducted to examine the validity of the proposed approach.

This study proposes a method to automatically generate numerical control (NC) programs from 3D computer-aided design (CAD) models in STL (Standard Triangulated Language) format. In STL format CAD of models, the three-dimensional shape is represented as a collection of triangular meshes. This format has the advantage of eliminating the need for concern about compatibility issues with data formats of 3D CAD models. On the other hand, there is a challenge in obtaining machining features that represent the characteristics of the three-dimensional shape for extracting the machining removal region. The previous study proposed a method to extract the machining region and obtain geometrical features such as boss or pocket shapes from only a 3D CAD model in STL format. However, that method was limited to product shapes represented only by independent boss shapes or pocket shapes. Those forms could be machined only by contour line machining, also called 2.5-axis machining. For general product shapes, it is necessary to recognize chamfers or filet shapes and free-form surfaces. Furthermore, the ideal machining process should be based on not only the desired shape, but also the extracted machining features to shorten the machining time as much as possible. Therefore, this study proposes a method to extract machining features such as chamfers, filets or free-form surfaces and to calculate the tool path necessary for machining. This study also proposed a method to extract pocket or slot machining features from multiple boss shapes for efficient machining. The machining experiments were conducted to validate the effectiveness of the proposed method. The results demonstrate that the tool path automatically generated using the proposed method from a 3D CAD model in STL format can be machined without any trouble and with the accuracy needed for practical use.

The metalworking industry is currently facing a shortage of human resources. Furthermore, there is a need for a shift to mass customization, which enables low-volume production at a cost equivalent to that of mass production. To satisfy the demand, it is necessary to increase labor productivity per worker by shortening production lead times. Automated tool path generation is one of the means to reduce production lead time. However, the generation of NC programs that enable high-precision machining requires enormous amounts of time and labor, since it is necessary to modify the NC programs according to the results of test cutting and reviewing the machining conditions. One of the factors causing machining errors in cutting is considered the deformation of the workpiece due to the clamping in a vise. Hence, even if dimensional tolerance measured on the machine is satisfied, dimensional errors may occur when the workpiece is removed from the vise. The purpose of this study is to realize the automated NC program generation for high-precision pocket machining. This study developed a system that predicts the elastic deformation of the workpiece due to the clamping force using Finite Element Method (FEM) and automatically generates a tool path for machining that satisfies the dimensional tolerance when the workpiece is removed from the vise. As a result of case study, it was confirmed that the proposed system can automatically generate tool paths that improves machining accuracy.

The gear skiving process involves a machining method wherein a skiving tool continuously machines a workpiece while rotating synchronously. This method proves to be efficient for machining internal gears, addressing challenges faced by conventional machining methods. With the increasing demand for various types of tooth profiles to accommodate the high-speed rotation of motors, there arises a need for enhanced capabilities in gear skiving processes. Traditionally, it has been challenging to machine a diverse array of tooth profiles using a single skiving tool due to its design specificity for each type of tooth profile. In response, this study endeavors to develop Computer-Aided Manufacturing (CAM) algorithms capable of facilitating the rough machining of various tooth profiles with a single skiving tool in gear skiving process. This makes it possible to rough various types of tooth profiles in a short time without special tools. To realize this goal, two novel methods have been proposed. One approach involves the utilization of a diamond-shaped edge skiving tool, while the other entails moving the skiving tool laterally for each depth of cut before tool feed. These proposed methods enable the gradual machining of various types of tooth profiles in accordance with their shapes. An algorithm has been developed to calculate the amount of lateral movement required for each depth of cut based on tool and workpiece specifications. To validate these methods, an experiment was conducted to machine an asymmetrical tooth profile using a diamond-shaped edge skiving tool. As a result, the amount of error in the machined tooth profile relative to the target tooth profile was within ±25 μm except for the root. In conclusion, this study successfully introduces a new gear skiving process method capable of roughing a diverse range of tooth profiles with a single skiving tool.

Manufacturing industries have increasingly recognized the value of Lean Six Sigma (LSS) not only for improving productivity but also for enhancing sustainability. The combination of green concept with LSS has also gained popularity by reducing waste, cost, and emissions. Prompting the exploration of enablers supporting its adoption in environmental services. This study aims to investigate the relationship between Green LSS (GLSS) enablers and operational benefits (OBs) in wastewater treatment plants (WWTPs) in Malaysia. The study examines five independent variables (IVs): strategic-based enablers (S), environmental-based enablers (Env), culture-based enablers (C), resource-based enablers (R), and linkage-based enablers (L), in relation to the dependent variable (DV) of OB. Data was collected from 65 certified competent personnel working in WWTPs and analysed using validity, reliability, factor analysis, and multiple linear regression. The results indicate that the IVs significantly predict OB when the p-value is below the 5% threshold. This suggests that the factors examined have a significant impact on WWTP operational benefits. Furthermore, the R² value of 0.390 indicates that the model explains 39% of the variance in OB. Specifically, the variables S and C significantly support the hypotheses, while Env, R, and L do not significantly influence OB. These findings provide valuable insights for the wastewater service sector in improving their understanding and implementation of GLSS to enhance operational performance in a developing country, Malaysia.

Simulation-based optimization often requires many simulations and can be difficult to adapt due to time constraints. To solve this problem, constructing approximators for simulations, such as the finite element method using machine learning, has attracted attention. However, creating these approximators requires a huge amount of training data. Therefore, we propose an integral neural network to construct highly accurate approximators with a small amount of data. The integral neural network is a linear approximator using deductive knowledge that constrains the shape of the approximate curve between learning points by multiple regression analysis in which the basis function is determined by deductive information and an inductive learning method that suppresses overlearning of the linear approximator by compensating factors that are not expressed in the basis function by deductive information of the linear approximator. The nonlinear approximator with inductive learning is integrated with the linear approximator by compensating for the influence of factors that cannot be formulated. In this paper, to apply this method to constructing approximators for thermal analysis of power devices, we extended the method to models other than multiple regression analysis for deductive information and constructed approximators. We showed that they can be approximated with high accuracy even by non-traditional models.

In end milling operations, the tools are exposed to high temperatures and high pressures, which inevitably causes the tool wear progress. It is well known that the tool wear progress strongly depends on the temperature of cutting edge. In this study, the thermal analysis to predict cutting temperature is performed by modeling heat transfer, heat input and heat dissipation between the tool and the workpiece based on the voxel model in which the workpiece and the tool are modeled in Cartesian and cylindrical coordinate systems, respectively. The heat input is estimated by the cutting energy predicted by machining simulation with the instantaneous cutting force model. To verify the validity of the proposed cutting temperature simulation method, the predicted cutting temperature was compared with the experimental results in which tool temperatures in end-milling operation measured by a thermocouple thermometer embedded inside the end mill. The predicted maximum temperature and temperature transient responses were in qualitative agreement with the measured results.

To combat the environmental issues posed by global warming and the depletion of natural resources, the emission of greenhouse gases (GHG) and the consumption of natural resources must be economically reduced. Upgrading and remanufacturing are expected to play important roles in dealing with both issues since they can save the production of virgin materials associated with GHG emissions for assembled products by component reuse and material recycling. Although upgrading components can add additional values and avoid their value obsolescence, composing reused components with shorter physical or value lifetime in a remanufactured product leads to decreasing revenue from selling the upgrade-remanufactured product. Hence, in these cases, component reuse and material recycling can be more economical life cycle option than upgrading and remanufacturing. Furthermore, disassembly is an essential process for recovery options such as upgrading, remanufacturing, components reuse, and material recycling, and tends costly due to the labor costs of manual disassembly. Disposing without disassembly may be a better life cycle option. Therefore, life cycle options, including upgrading, remanufacturing, reusing, recycling and disposal should be suitably selected for each component based on an additional value by upgrading, physical and value lifetimes for each component. This study proposes an upgrade-remanufacturing decision method to maximize GHG saving rate and profit using 0-1 integer programming with ε constraint method. The numerical experiments are conducted using the laptop consisted of 34 components. The results in the laptop indicate that the selling price of upgrade-remanufactured product should be set to more than 2,000 Yen, and the achievement of much higher GHG saving rate such as 99% would lead to negative earnings. Additionally, the bi-objective model is expanded to multi-objective for profit, GHG saving, and recovery rates for investigation of the profit and the selected life cycle options for each component.

This paper presents a trustworthy architecture for Web3 service that establishes self-sovereign services. This architecture leverages decentralized identifiers (DID) and resource access control based on user trust scores and policies. By transitioning from centrally managed ID to DID, it will be possible to move towards ID management that does not rely on service providers. Trust scores are calculated by analyzing the social network generated from the history of past transactions between users stored in the blockchain. The degree of trust is quantified using measures of centrality and similarity. Centrality is measured by the number of user achievements and the extent of their connections. Similarity is calculated by estimating the communities they belong to using node embedding. These trust scores are clustered to generate policy based on role-based access control (RBAC). The authorization function is programmatically realized based on user credibility by comparing the results of the distribution of accessible resources by policy. In addition, a Web3 service broker acts as an intermediary between multiple service providers and users to provide optimal services through smart contracts. A Web3 service broker programmatically select services without intermediary agencies. A trustworthy architecture for Web3 service is shown as a case study of a home delivery service. In this scenario, Web3 service brokers offer optimal services for private companies and small and medium-sized businesses. This paper is an extended version of ”A Web3 Service Architecture with Self-Sovereign Identity” presented at iDECON/MS2023.