Browsing by Author "van Heerden AH"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    A Study of the Soft Skills Possessed and Required in the Construction Sector
    (MDPI (Basel, Switzerland), 2023-02-14) van Heerden AH; Jelodar MB; Chawynski G; Ellison S; Tummalapudi M; Harper C
    Soft skills are essential to employability and retention; therefore, if obtained and observed, they can significantly reduce sector-wide turnover. This study aims to investigate and compare soft skills that industry professionals currently possess and soft skills the industry requires and needs to attain. A questionnaire was administered using the RICS database, and 741 respondents participated in this study. Initially, the soft skills possessed and the soft skills required were analysed and compared via descriptive statistics. Furthermore, principal component factor analysis was used to identify the underlying factors and classify the identified soft skills. It was found that there are alignments and evident discrepancies between the actual skills currently possessed and the skills required by these professionals. The soft skills currently possessed by the industry were classified into three groups: (a) Ethics and Professionalism Cluster; (b) Self-Effort Management Cluster; and (c) Management—Leadership and Power Cluster. This was different to the two clusters identified for the soft skills requirements, which were: (a) trait-based cluster—less controllable; (b) training-based cluster—more controllable. The study concludes that there are controllable and less-controllable skills, which need to be possessed and managed in building professionals. Controllable soft skills are easier to train, whereas trait-based soft skills are more difficult to train and possess. The findings of this research are significant as their understanding can be used to help mitigate turnover and guide construction sector professionals to plan for the appropriate skills they require.
  • Thumbnail Image
    Item
    Review of the thermal efficiency of a tube-type solar air heaters
    (Elsevier B.V., 2024-05-11) Pardeshi PS; Boulic M; van Heerden AH; Phipps R; Cunningham CW
    There is an urgent need to provide evidence that solar air heaters can be effective for heating and ventilating low-rise buildings. Solar air heaters are devices that can convert solar energy into thermal energy for moderate and low-temperature applications such as space heating, preheating, crop drying, and the food industry. However, its efficiency is low due to the low heat transfer coefficient between the absorber and the flowing air, but they are also simple to construct and operate as there is low-risk leakage of heat transfer liquids. The two main types of solar air heaters are flat plate and tube-type. To date, flat plate solar air heaters have received the most attention in the research literature, but evidence of the efficiency gains from using tube-type solar air heaters is growing. The study aims to provide up-to-date information on tube-type solar air heaters, which will help advance the development and uptake of solar air heaters. The research showed that thermal efficiency gains could be achieved by altering the design of the solar air heater including different artificial roughness geometries inside the tubes, integrating solar thermal energy systems, application of coatings or reflectors inside the solar air heaters, or using evacuated tubes and micro heat pipe array systems. This literature study showed that evacuated tubes and micro heat pipe array systems have higher thermal efficiency than other techniques. Based on the detailed discussion of various techniques for improving the thermal efficiency of solar air heaters, a new roughness geometry was proposed.
  • Thumbnail Image
    Item
    Transfer learning on transformers for building energy consumption forecasting—A comparative study
    (Elsevier B V, 2025-06-01) Spencer R; Ranathunga S; Boulic M; van Heerden AH; Susnjak T
    Energy consumption in buildings is steadily increasing, leading to higher carbon emissions. Predicting energy consumption is a key factor in addressing climate change. There has been a significant shift from traditional statistical models to advanced deep learning (DL) techniques for predicting energy use in buildings. However, data scarcity in newly constructed or poorly instrumented buildings limits the effectiveness of standard DL approaches. In this study, we investigate the application of six data-centric Transfer Learning (TL) strategies on three Transformer architectures—vanilla Transformer, Informer, and PatchTST—to enhance building energy consumption forecasting. Transformers, a relatively new DL framework, have demonstrated significant promise in various domains; yet, prior TL research has often focused on either a single data-centric strategy or older models such as Recurrent Neural Networks. Using 16 diverse datasets from the Building Data Genome Project 2, we conduct an extensive empirical analysis under varying feature spaces (e.g., recorded ambient weather) and building characteristics (e.g., dataset volume). Our experiments show that combining multiple source datasets under a zero-shot setup reduces the Mean Absolute Error (MAE) of the vanilla Transformer model by an average of 15.9 % for 24 h forecasts, compared to single-source baselines. Further fine-tuning these multi-source models with target-domain data yields an additional 3–5 % improvement. Notably, PatchTST outperforms the vanilla Transformer and Informer models. Overall, our results underscore the potential of combining Transformer architectures with TL techniques to enhance building energy consumption forecasting accuracy. However, careful selection of the TL strategy and attention to feature space compatibility are needed to maximize forecasting gains.