Search results

Simulation-based optimisation (SO) is a popular optimisation approach for building and civil engineering construction planning. However, in the framework of SO, the simulation is continuously invoked during the optimisation trajectory, which increases the computational loads to levels unrealistic for timely construction decisions. Modification on the optimisation settings such as reducing searching ability is a popular method to address this challenge, but the quality measurement of the obtained optimal decisions, also termed as optimisation quality, is also reduced by this setting. Therefore, this study aims to develop an optimisation approach for construction planning that reduces the high computational loads of SO and provides reliable optimisation quality simultaneously.

This study proposes the optimisation approach by modifying the SO framework through establishing an embedded connection between simulation and optimisation technologies. This approach reduces the computational loads and ensures the optimisation quality associated with the conventional SO approach by accurately learning the knowledge from construction simulations using embedded ensemble learning algorithms, which automatically provides efficient and reliable fitness evaluations for optimisation iterations.

A large-scale project application shows that the proposed approach was able to reduce computational loads of SO by approximately 90%. Meanwhile, the proposed approach outperformed SO in terms of optimisation quality when the optimisation has limited searching ability.

The core contribution of this research is to provide an innovative method that improves efficiency and ensures effectiveness, simultaneously, of the well-known SO approach in construction applications. The proposed method is an alternative approach to SO that can run on standard computing platforms and support nearly real-time construction on-site decision-making.

The purpose of this paper is to propose a system dynamic simulated process model for maintenance work management incorporating the Fourth Industrial Revolution (4IR) technologies.

The extant literature in physical assets maintenance depicts that poor maintenance management is predominantly because of a lack of a clearly defined maintenance work management process model, resulting in poor management of maintenance work. This paper solves this complex phenomenon using a combination of conceptual process modeling and system dynamics simulation incorporating 4IR technologies. A process for maintenance work management and its control actions on scheduled maintenance tasks versus unscheduled maintenance tasks is modeled, replicating real-world scenarios with a digital lens (4IR technologies) for predictive maintenance strategy.

A process for maintenance work management is thus modeled and simulated as a dynamic system. Post-model validation, this study reveals that the real-world maintenance work management process can be replicated using system dynamics modeling. The impact analysis of 4IR technologies on maintenance work management systems reveals that the implementation of 4IR technologies intensifies asset performance with an overall gain of 27.46%, yielding the best maintenance index. This study further reveals that the benefits of 4IR technologies positively impact equipment defect predictability before failure, thereby yielding a predictive maintenance strategy.

The study focused on maintenance work management system without the consideration of other subsystems such as cost of maintenance, production dynamics, and supply chain management.

The maintenance real-world quantitative data is retrieved from two maintenance departments from company A, for a period of 24 months, representing years 2017 and 2018. The maintenance quantitative data retrieved represent six various types of equipment used at underground Mines. The maintenance management qualitative data (Organizational documents) in maintenance management are retrieved from company A and company B. Company A is a global mining industry, and company B is a global manufacturing industry. The reliability of the data used in the model validation have practical implications on how maintenance work management system behaves with the benefit of 4IR technologies' implementation.

This research study yields an overall benefit in asset management, thereby intensifying asset performance. The expected learnings are intended to benefit future research in the physical asset management field of study and most important to the industry practitioners in physical asset management.

This paper provides for a model in which maintenance work and its dynamics is systematically managed. Uncontrollable corrective maintenance work increases the complexity of the overall maintenance work management. The use of a system dynamic model and simulation incorporating 4IR technologies adds value on the maintenance work management effectiveness.

Stock price prediction is a hot topic and traditional prediction methods are usually based on statistical and econometric models. However, these models are difficult to deal with nonstationary time series data. With the rapid development of the internet and the increasing popularity of social media, online news and comments often reflect investors’ emotions and attitudes toward stocks, which contains a lot of important information for predicting stock price. This paper aims to develop a stock price prediction method by taking full advantage of social media data.

This study proposes a new prediction method based on deep learning technology, which integrates traditional stock financial index variables and social media text features as inputs of the prediction model. This study uses Doc2Vec to build long text feature vectors from social media and then reduce the dimensions of the text feature vectors by stacked auto-encoder to balance the dimensions between text feature variables and stock financial index variables. Meanwhile, based on wavelet transform, the time series data of stock price is decomposed to eliminate the random noise caused by stock market fluctuation. Finally, this study uses long short-term memory model to predict the stock price.

The experiment results show that the method performs better than all three benchmark models in all kinds of evaluation indicators and can effectively predict stock price.

In this paper, this study proposes a new stock price prediction model that incorporates traditional financial features and social media text features which are derived from social media based on deep learning technology.

Metropolitan areas suffer from frequent road traffic congestion not only during peak hours but also during off-peak periods. Different machine learning methods have been used in travel time prediction, however, such machine learning methods practically face the problem of overfitting. Tree-based ensembles have been applied in various prediction fields, and such approaches usually produce high prediction accuracy by aggregating and averaging individual decision trees. The inherent advantages of these approaches not only get better prediction results but also have a good bias-variance trade-off which can help to avoid overfitting. However, the reality is that the application of tree-based integration algorithms in traffic prediction is still limited. This study aims to improve the accuracy and interpretability of the models by using random forest (RF) to analyze and model the travel time on freeways.

As the traffic conditions often greatly change, the prediction results are often unsatisfactory. To improve the accuracy of short-term travel time prediction in the freeway network, a practically feasible and computationally efficient RF prediction method for real-world freeways by using probe traffic data was generated. In addition, the variables’ relative importance was ranked, which provides an investigation platform to gain a better understanding of how different contributing factors might affect travel time on freeways.

The parameters of the RF model were estimated by using the training sample set. After the parameter tuning process was completed, the proposed RF model was developed. The features’ relative importance showed that the variables (travel time 15 min before) and time of day (TOD) contribute the most to the predicted travel time result. The model performance was also evaluated and compared against the extreme gradient boosting method and the results indicated that the RF always produces more accurate travel time predictions.

This research developed an RF method to predict the freeway travel time by using the probe vehicle-based traffic data and weather data. Detailed information about the input variables and data pre-processing were presented. To measure the effectiveness of proposed travel time prediction algorithms, the mean absolute percentage errors were computed for different observation segments combined with different prediction horizons ranging from 15 to 60 min.

Present for more than 20 years, 3D food printing (3DFP) technology has not experienced the same widespread adoption as its non-food counterparts. It is believed that relevant business models are crucial for its expansion. The purpose of this study is to identify the dominant prototypical business models and patterns in the 3DFP industry. The knowledge gained could be used to provide directions for business model innovation in this industry.

The authors established a business model framework and used it to analyse the identified 3DFP manufacturers. The authors qualitatively identified the market’s prototypical business models and used agglomerative hierarchical clustering to extract further patterns.

All identified 3DFP businesses use the prototypical business model of selling ownership of physical assets, with some variations. Low-cost 3D food printers for private usage and dedicated 3D food printers for small-scale food producers are the two primary patterns identified. Furthermore, several benefits of 3DFP technology are not being used, and the identified manufacturers are barely present in high-revenue markets, which prevents them from driving technological innovation forward.

The extracted patterns can be used by the companies within the 3DFP industry and even in other additive manufacturing segments to reflect upon, refine or renew their business model. Some directions for business model innovation in this industry are provided.

To the best of the authors’ knowledge, this is the first quantitative study to give an account of the current 3DFP business models and their possible evolution. This study also contributes to the business model patterns methodological development.

In current business management, knowledge is considered to be a strategic resource that can strengthen an organization’s competitiveness. Today, under the process of continuous globalization, almost all companies are rapidly exposed to global competition regardless of their scale or type of business. However, multinational management is very complicated and uncertain and it is hard for multinationals to effectively coordinate and manage their global value chains. In light of this, the utility of multinational management based on knowledge is increased. The present study examines multinational firms’ knowledge management systems, knowledge creation processes and global supply chain performance and attempts to reveal any significant linkages between these latent variables. For this research interest, we proposed 18 items to measure four types of knowledge creation processes (SECI) designed by Nonaka (1994) and revised by authors considering the global business environment, in particular involving the global supply chain management concept. Utilizing the confirmed SECI model, 128 sample companies were classified into four groups according to the levels of their knowledge creation processes. The empirical results of this study reveal important linkages between a multinational firm’s knowledge management system and knowledge creation process, as well as between its knowledge creation process and global supply chain management performance. In particular, the current work suggests that the creation and conversion of tacit knowledge as well as explicit knowledge can be effectively supported by information and communication technology.

Having defined liquidity, the aim is to assess the predictive capacity of its representative variables, so that economic fluctuations may be better understood.

Conceptual variables that are representative of liquidity will be used to formulate the predictions. The results of various machine learning models will be compared, leading to some reflections on the predictive value of the liquidity variables, with a view to defining their selection.

The predictive capacity of the model was also found to vary depending on the source of the liquidity, in so far as the data on liquidity within the private sector contributed more than the data on public sector liquidity to the prediction of economic fluctuations. International liquidity was seen as a more diffuse concept, and the standardization of its definition could be the focus of future studies. A benchmarking process was also performed when applying the state-of-the-art machine learning models.

Better understanding of these variables might help us toward a deeper understanding of the operation of financial markets. Liquidity, one of the key financial market variables, is neither well-defined nor standardized in the existing literature, which calls for further study. Hence, the novelty of an applied study employing modern data science techniques can provide a fresh perspective on financial markets.

因此,就了解經濟週期和經濟發展而言,流動資金是一個極其重要的概念。本研究擬在安全資產的價格預測方面取得進步。安全資產代表了經濟的實際情況,特別是美國的十年期國債。

流動資金的定義上面已說明了; 為進一步了解經濟波動,本研究擬對流動資金代表性變量的預測能力進行評估。

研究使用作為流動資金代表的概念變項去規劃預測。各機器學習模型的結果會作比較,這會帶來對流動資金變量的預測值的深思,而深思的目的是確定其選擇。

只要在私營部門內流動資金的數據比公營部門的流動資金數據、在預測經濟波動方面貢獻更大時,我們發現、模型的預測能力也會依賴流動資金的來源而存在差異。國際流動資金被視為一個晦澀的概念,而它的定義的標準化,或許應是未來學術研究的焦點。當應用最先進的機器學習模型時,標桿分析法的步驟也施行了。

若我們對有關的變量加深認識,我們就可更深入地理解金融市場的運作。流動資金,雖是金融市場中一個極其重要的變量,但在現存的學術文獻裏,不但沒有明確的定義,而且也沒有被標準化; 就此而言,未來的研究或許可在這方面作進一步的探討。因此,本研究為富有新穎思維的應用研究,研究使用了現代數據科學技術,這可為探討金融市場提供一個全新的視角。

Clinical notes typically contain medical jargons and specialized words and phrases that are complicated and technical to most people, which is one of the most challenging obstacles in health information dissemination to consumers by healthcare providers. The authors aim to investigate how to leverage machine learning techniques to transform clinical notes of interest into understandable expressions.

The authors propose a natural language processing pipeline that is capable of extracting relevant information from long unstructured clinical notes and simplifying lexicons by replacing medical jargons and technical terms. Particularly, the authors develop an unsupervised keywords matching method to extract relevant information from clinical notes. To automatically evaluate completeness of the extracted information, the authors perform a multi-label classification task on the relevant texts. To simplify lexicons in the relevant text, the authors identify complex words using a sequence labeler and leverage transformer models to generate candidate words for substitution. The authors validate the proposed pipeline using 58,167 discharge summaries from critical care services.

The results show that the proposed pipeline can identify relevant information with high completeness and simplify complex expressions in clinical notes so that the converted notes have a high level of readability but a low degree of meaning change.

The proposed pipeline can help healthcare consumers well understand their medical information and therefore strengthen communications between healthcare providers and consumers for better care.

An innovative pipeline approach is developed to address the health literacy problem confronted by healthcare providers and consumers in the ongoing digital transformation process in the healthcare industry.

University research efforts have not been effective in developing lasting impacts on operations management in construction because of inadequate coordination between academia and industry. This study aims to describe the development of an industry–university (IU) relationship which has enabled the conduct of practically and scientifically relevant research.

Design science research was carried out between 2016 and 2019 to build a consortium between a university and 17 design, construction, technology and logistics companies for enabling process innovations in construction. The consortium conducted industry-funded research on various topics, such as takt production, lean design management, prefabrication, measurement of waste and business models supported by digitalisation. The academic and practical impacts of the consortium’s research projects were investigated through a survey and in-depth company interviews.

The paper presents a conceptual model for creating an IU relationship to support scientifically and practically relevant research. The model includes network architects who mobilised consortium development and a joint governance body that developed a shared long-term vision and selected research topics based on this vision. The results show that using the model’s approach, the consortium selected research topics that have led to both academic publications and process innovations in construction.

Using empirical data, this study describes how to create a win-win IU innovation relationship that enables the implementation of process innovations into the construction sector and, at the same time, the conduct of scientific research in construction management.

Facing the diverse needs of large-scale customers, based on available railway service resources and service capabilities, this paper aims to research the design method of railway freight service portfolio, select optimal service solutions and provide customers with comprehensive and customized freight services.

Based on the characteristics of railway freight services throughout the entire process, the service system is decomposed into independent units of service functions, and a railway freight service combination model is constructed with the goal of minimizing response time, service cost and service time. A model solving algorithm based on adaptive genetic algorithm is proposed.

Using the computational model, an empirical analysis was conducted on the entire process freight service plan for starch sold from Xi'an to Chengdu as an example. The results showed that the proposed optimization model and algorithm can effectively guide the design of freight plans and provide technical support for real-time response to customers' diversified entire process freight service needs.

With the continuous optimization and upgrading of railway freight source structure, customer demands are becoming increasingly diverse and personalized. Studying and designing a reasonable railway freight service plan throughout the entire process is of great significance for timely response to customer needs, improving service efficiency and reducing design costs.