Search results

The compressive strength of concrete depends on many interdependent parameters; its exact prediction is not that simple because of complex processes involved in strength development. This study aims to predict the compressive strength of normal concrete and high-performance concrete using four datasets.

In this paper, five established individual Machine Learning (ML) regression models have been compared: Decision Regression Tree, Random Forest Regression, Lasso Regression, Ridge Regression and Multiple-Linear regression. Four datasets were studied, two of which are previous research datasets, and two datasets are from the sophisticated lab using five established individual ML regression models.

The five statistical indicators like coefficient of determination (R²), mean absolute error, root mean squared error, Nash–Sutcliffe efficiency and mean absolute percentage error have been used to compare the performance of the models. The models are further compared using statistical indicators with previous studies. Lastly, to understand the variable effect of the predictor, the sensitivity and parametric analysis were carried out to find the performance of the variable.

The findings of this paper will allow readers to understand the factors involved in identifying the machine learning models and concrete datasets. In so doing, we hope that this research advances the toolset needed to predict compressive strength.

The metallic alloys and their components fabricated via laser powder bed fusion (LPBF) suffer from the microvoids formed inevitably due to the extreme solidification rate during fabrication, which are impossible to be removed by heat treatment. This paper aims to remove those microvoids in as-built AlSi10Mg alloys by hot forging and enhance their mechanical properties.

AlSi10Mg samples were built using prealloyed powder with a set of optimized LPBF parameters, viz. 350 W of laser power, 1,170 mm/s of scan speed, 50 µm of layer thickness and 0.24 mm of hatch spacing. As-built samples were preheated to 430°C followed by immediate pressing with two different thickness reductions of 10% and 35%. The effect of hot forging on the microstructure was analyzed by means of X-ray diffraction, scanning electron microscopy, electron backscattered diffraction and transmission electron microscopy. Tensile tests were performed to reveal the effect of hot forging on the mechanical properties.

By using hot forging, the large number of microvoids in both as-built and post heat-treated samples were mostly healed. Moreover, the Si particles were finer in forged condition (∼150 nm) compared with those in heat-treated condition (∼300 nm). Tensile tests showed that compared with heat treatment, the hot forging process could noticeably increase tensile strength at no expense of ductility. Consequently, the toughness (integration of tensile stress and strain) of forged alloy increased by ∼86% and ∼24% compared with as-built and heat-treated alloys, respectively.

Hot forging can effectively remove the inevitable microvoids in metals fabricated via LPBF, which is beneficial to the mechanical properties. These findings are inspiring for the evolution of the LPBF technique to eliminate the microvoids and boost the mechanical properties of metals fabricated via LPBF.

Machine learning (ML), and deep learning in particular, is gaining traction across a myriad of real-life applications. Portfolio management is no exception. This paper provides a systematic literature review of deep learning applications for portfolio management. The findings are likely to be valuable for industry practitioners and researchers alike, experimenting with novel portfolio management approaches and furthering investment management practice.

This review follows the guidance and methodology of Linnenluecke et al. (2020), Massaro et al. (2016) and Fisch and Block (2018) to first identify relevant literature based on an appropriately developed search phrase, filter the resultant set of publications and present descriptive and analytical findings of the research itself and its metadata.

The authors find a strong dominance of reinforcement learning algorithms applied to the field, given their through-time portfolio management capabilities. Other well-known deep learning models, such as convolutional neural network (CNN) and recurrent neural network (RNN) and its derivatives, have shown to be well-suited for time-series forecasting. Most recently, the number of papers published in the field has been increasing, potentially driven by computational advances, hardware accessibility and data availability. The review shows several promising applications and identifies future research opportunities, including better balance on the risk-reward spectrum, novel ways to reduce data dimensionality and pre-process the inputs, stronger focus on direct weights generation, novel deep learning architectures and consistent data choices.

Several systematic reviews have been conducted with a broader focus of ML applications in finance. However, to the best of the authors’ knowledge, this is the first review to focus on deep learning architectures and their applications in the investment portfolio management problem. The review also presents a novel universal taxonomy of models used.