Search results

This paper aims to present a hybrid machine learning algorithm for additive manufacturing (AM) design feature recommendation during the conceptual design phase.

In the proposed hybrid machine learning algorithm, hierarchical clustering is performed on coded AM design features and target components, resulting in a dendrogram. Existing industrial application examples are used to train a supervised classifier that determines the final sub-cluster within the dendrogram containing the recommended AM design features.

Through a case study of designing additive manufactured R/C car components, the proposed hybrid machine learning method was proven useful in providing feasible conceptual design solutions for inexperienced designers by recommending appropriate AM design features.

The proposed method helps inexperienced designers who are newly exposed to AM capabilities explore and utilize AM design knowledge computationally.

This paper aims to propose methods to evaluate the characteristic length of the melt pool for accurate fabrication and to identify the optimal process parameters in the selective laser melting process.

Specimens with the types of the scans by controlling the degree of the overlap with hatch spacing are fabricated. The scan modes are classified by statistically analyzing the results of hardness tests. According to the classification of the scans, the evaluation methods are proposed based on the observation of the shape of the solidified melt pool.

The control of the hatch spacing can reproduce all modes of the scan conditions, and hardness can be used to classify the scan modes.

The proposed evaluation methods are based on the analysis of the experimental observation so that they can be easily used for the real evaluation.

In recent years, additive manufacturing techniques have attracted much research attention because of their ability to fabricate customised parts with complex geometry. The range of composites suitable for laser-based powder bed fusion technique is limited, and has not been investigated yet. This paper aims to study the fabrication of AlSi₁₀Mg reinforced with nAl₂O₃ using the laser-based powder bed fusion technique.

An experimental approach was used to investigate the densification of AlSi₁₀Mg–nAl₂O₃ composites using laser-based powder bed fusion technique. Optimisation of the porosity was performed, and microstructure evolution was evaluated.

In this study, laser volumetric energy density (approximately 109 J/mm³) was found to be required for the fabrication of AlSi₁₀Mg–nAl₂O₃ composites with a relative volumetric density approximating 99%. The use of laser volumetric energy density resulted in larger grains. Columnar grain structure was observed via the use of electron backscatter diffraction mapping.

This paper examines the processing of new aluminium composite material suitable for the fabrication via the laser-based powder bed fusion technique.

Hoping to uncover the physical principles of the vibration of the functionally graded material (FGM) microplate, by which the authors can make contributions to the design and manufacturing process in factories like micro-electro-mechanical system (MEMS) and other industries.

The authors design a method by establishing a reasonable mathematical model of the physical microplate composed of a porous FGM.

The authors discover that the porosity, the distributions of porosity, the power law of the FGM and the length-to-thickness ratio all affect the natural frequency of the vibration of the microplate, but in different ways.

Originally proposed a model of the micro FGM plate considering the different distributions of the porosity and scale effect and analyzed the vibration frequency of it.

The purpose of this study investigates the asymmetric effects of brand origin confusion (BOC) on consumer preference and the purchase of local versus foreign brands in China. Drawing on the general country‐of‐origin (COO) literature and recent developments in brand‐origin studies and the emerging market phenomenon globally, it proposes and test a model of the asymmetric effects of BOC on consumer preference and the purchase of local versus foreign brands in China. This study intends to help to explain from a new angle the decreasing competitiveness of foreign brands in emerging markets, such as China.

The study pretest on nationally distributed brands across seven product categories resulted in a final set of 67 brands: 35 foreign and 32 local. Four hundred respondents evaluated measures related to brand origin, brand awareness, brand value, brand preference, and brand purchases in the previous six months. Hierarchical regression analysis was used in data analysis.

The hypotheses on the asymmetric effects of BOC between local and foreign brands in China were mostly supported. Specifically, the results showed that local brands are likely to be in an advantageous position when there is a high level of BOC. However, as the brand knowledge of consumers increases, the effects of BOC decrease.

This study provides evidence of the asymmetric effects of BOC between local and foreign brands and the moderating role of brand knowledge for local brands in China. It fills a gap in the international branding and marketing communication literature, and offers meaningful managerial insights for both local and international companies to formulate effective branding and marketing communication strategies in China and possibly in other emerging markets.

The purpose of this paper is to perform a systematic review on the application of different multi-criteria decision-making (MCDM) methods in solving the site selection problem across multiple problem domains. The domains are energy generation, logistics, public services and retail facilities. This study aims to answer the following research questions: Which evaluating criteria were used for each site selection problem domain? Which MCDM methods were frequently applied in a particular site selection problem domain?

The goals of the systematic review were to identify the evaluating criteria as well as the MCDM method used for each problem domain. A total of 81 recent papers (2014–2018) including 32 papers published in conference proceedings and 49 journal articles from various databases including IEEE Xplore, PubMed, Springer, Taylor and Francis as well as ScienceDirect were evaluated.

This study has shown that site selection for energy generation facilities is the most active site selection problem domain, and that the analytic hierarchy process (AHP) method is the most commonly used MCDM method for site selection. For energy generation, the criteria which were most used were geographical elements, land use, cost and environmental impact. For logistics, frequently used criteria were geographical elements and distance, while for public services population density, supply and demand, geographical layout and cost were the criteria most used. Criteria useful for retail facilities were the size (space) of the store, demographics of the site, the site characteristics and rental of the site (cost).

This study is limited to reviewing papers which were published in the years 2014–2018 only, and only covers the domains of energy generation, logistics, public services and retail facilities.

MCDM is a viable tool to be used for solving the site selection problem across the domains of energy generation, logistics, public services and retail facilities. The usage of MCDM continues to be relevant as a complement to machine learning, even as data originating from embedded IoT devices in built environments becomes increasingly Big Data like.

Previous systematic review studies for MDCM and built environments have either focused on studying the MCDM techniques itself, or have focused on the application of MCDM for site selection in a single problem domain. In this study, a critical review of MCDM techniques used for site selection as well as the critical criteria used during the MCDM process of site selection was performed on four different built environment domains.