Jiqian Dong, Sikai Chen, Mohammad Miralinaghi, Tiantian Chen and Samuel Labi
Perception has been identified as the main cause underlying most autonomous vehicle related accidents. As the key technology in perception, deep learning (DL) based computer…
Abstract
Purpose
Perception has been identified as the main cause underlying most autonomous vehicle related accidents. As the key technology in perception, deep learning (DL) based computer vision models are generally considered to be black boxes due to poor interpretability. These have exacerbated user distrust and further forestalled their widespread deployment in practical usage. This paper aims to develop explainable DL models for autonomous driving by jointly predicting potential driving actions with corresponding explanations. The explainable DL models can not only boost user trust in autonomy but also serve as a diagnostic approach to identify any model deficiencies or limitations during the system development phase.
Design/methodology/approach
This paper proposes an explainable end-to-end autonomous driving system based on “Transformer,” a state-of-the-art self-attention (SA) based model. The model maps visual features from images collected by onboard cameras to guide potential driving actions with corresponding explanations, and aims to achieve soft attention over the image’s global features.
Findings
The results demonstrate the efficacy of the proposed model as it exhibits superior performance (in terms of correct prediction of actions and explanations) compared to the benchmark model by a significant margin with much lower computational cost on a public data set (BDD-OIA). From the ablation studies, the proposed SA module also outperforms other attention mechanisms in feature fusion and can generate meaningful representations for downstream prediction.
Originality/value
In the contexts of situational awareness and driver assistance, the proposed model can perform as a driving alarm system for both human-driven vehicles and autonomous vehicles because it is capable of quickly understanding/characterizing the environment and identifying any infeasible driving actions. In addition, the extra explanation head of the proposed model provides an extra channel for sanity checks to guarantee that the model learns the ideal causal relationships. This provision is critical in the development of autonomous systems.
Details
Keywords
Oluseyi Julius Adebowale and Justus Ngala Agumba
Labour productivity in construction has fallen behind other industries in most of the world and has declined continuously for decades. Although several scholarly research projects…
Abstract
Purpose
Labour productivity in construction has fallen behind other industries in most of the world and has declined continuously for decades. Although several scholarly research projects have been conducted to salvage the prevalent low labour productivity in construction, contractors in the construction industry have continued to grapple with the devastating impact of low productivity. The purpose of this study is to determine key areas of focus necessary to promote productivity growth in construction.
Design/methodology/approach
Bibliometric and scientometric assessments were conducted to map the existing construction labour productivity (CLP) studies and establish key focus areas in the research domain. The keywords “Construction Productivity” OR “Construction Labour Productivity” OR “Construction Labor Productivity” OR “Construction Worker Productivity”.
Findings
Emerging trends in the CLP research field are reported. The study also determined the most productive authors and collaboration among authors, most productive journals, most active regions and publications with the highest impact in CLP research.
Research limitations/implications
Documents published in the Scopus database were considered for analysis because of the wider coverage of the database. Journal and conference articles written in English language represent the inclusion criteria, while articles in press, review, book chapters, editorial, erratum, note, short survey and data paper were excluded from analysis. The study is also limited to documents published from 2012 to 2021.
Practical implications
The study brought to the awareness of the industry practitioners and other construction stakeholders, the key knowledge areas that are critical to promoting productivity growth in construction.
Originality/value
Except bibliometric analysis, previous research studies have used different approaches to investigate productivity in construction. The study presented future research directions through the emerging knowledge areas identified in the study.
Details
Keywords
Sherin Kunhibava, Zakariya Mustapha, Aishath Muneeza, Auwal Adam Sa'ad and Mohammad Ershadul Karim
This paper aims to explore issues arising from ṣukūk (Islamic bonds) on blockchain, including Sharīʾah (Islamic law) and legal matters.
Abstract
Purpose
This paper aims to explore issues arising from ṣukūk (Islamic bonds) on blockchain, including Sharīʾah (Islamic law) and legal matters.
Design/methodology/approach
A qualitative methodology is used in conducting this research where relevant literature on ṣukūk was reviewed. Through a doctrinal approach, the paper presents analyses on the practice of ṣukūk and ṣukūk on blockchain by discussing its legal, Sharīʾah and regulatory issues. This culminates in a conceptual analysis of blockchain ṣukūk and its peculiar challenges.
Findings
This paper reveals that digitizing ṣukūk issuance through blockchain remedies certain inefficiencies associated with ṣukūk transactions. Indeed, structuring ṣukūk on a blockchain platform can increase transparency of underlying ṣukūk assets and cash flows in addition to reducing costs and the number of intermediaries in ṣukūk transactions. The paper likewise brings to light legal, regulatory, Sharīʾah and cyber risks associated with ṣukūk on blockchain that confront investors, practitioners and regulators. This calls for deeper collaboration in research among Sharīʾah scholars, lawyers, regulators and information technology experts.
Research limitations/implications
As a pioneering subject, the paper notes the prospects of blockchain ṣukūk and the current dearth of literature on it. The paper would assist relevant Islamic capital market entities and authorities to determine the potential and impact of blockchain ṣukūk in their respective businesses and the financial system.
Practical implications
Blockchain ṣukūk will assist in addressing issues inherent in classical ṣukūk and in paving the way to innovative solutions that will facilitate and enhance the quality of ṣukūk transactions. For that, ṣukūk would require appropriate regulatory technology to address its governance and regulation peculiarities.
Originality/value
Integrating ṣukūk with blockchain technology will add value to it. The paper advances the idea that blockchain ṣukūk revolutionises ṣukūk and enhances its practice against known inadequacies.
Details
Keywords
Francisco Muñoz-Leiva, Doaa Herzallah, Ismael Ramón Sánchez-Borrego and Francisco Liébana-Cabanillas
This study examines the role of logotypes in advertising effectiveness on s-commerce platforms by analyzing the visual attention paid by the consumer to fashion branding �…
Abstract
Purpose
This study examines the role of logotypes in advertising effectiveness on s-commerce platforms by analyzing the visual attention paid by the consumer to fashion branding – wordmarks or combination marks – and their subsequent recall.
Design/methodology/approach
The study examines the main areas of visual representation of the brand (VRB) on the Instagram network and the user’s corresponding areas of interest on a mobile-device screen. Attention and recall of the VRB are assessed in light of different classification variables (users’ gender, age and level of experience in s-commerce tools) to better understand how VRB may be leveraged by fashion retailers to encourage purchasing behavior. To achieve this objective, a mixed experiment design based on the eye-tracking methodology and a self-administered questionnaire is carried out.
Findings
The results indicate that visual attention, gender, age and s-commerce experience all contribute to determining users’ recall of the brand logo to which they are exposed on-screen. By considering the different s-commerce user profiles that exhibit different visualization behaviors, fashion retailers will be better placed to improve their online advertising campaigns and, ultimately, increase brand sales. The findings also point to promising future research directions on the effectiveness of branding strategies.
Originality/value
This highly innovative study provides in-depth insights into advertising effectiveness in terms of attention and recall, according to the main types of VRB for two specific s-commerce tools used by a high-street fashion brand, namely, its profile on Instagram Shop and its profile on Instagram Stories.
Details
Keywords
Tinyiko Vivian Dube and Lorette Jacobs
This paper aimed to determine the extent to which academic libraries and information services were extended due to the emergence of COVID-19 in the Gauteng Province, South Africa.
Abstract
Purpose
This paper aimed to determine the extent to which academic libraries and information services were extended due to the emergence of COVID-19 in the Gauteng Province, South Africa.
Design/methodology/approach
Founded on a pragmatism paradigm, the sequential explanatory research design was adopted to engage with participants and respondents on their experience of library services extensions to support users during the COVID-19 pandemic. Data were collected using online questionnaires and interviews. Cluster and purposive sampling were used and data for the quantitative part were analyzed using the Statistical Package for the Social Sciences (SPSS), whilst qualitative data were analyzed manually.
Findings
Findings revealed that academic libraries operating in a higher education environment provided extensive support to remote users during the COVID-19 pandemic. This was done through the utilization of a variety of technology utilization, ranging from traditional e-mail support to the use of technology related to Artificial Intelligence such as the BOTsa, which is a Chatbot aimed to assist users in receiving speedy responses to library-related inquiries.
Originality/value
This study is unique in that it focuses on academic libraries that operate in higher education environments where support for achieving academic endeavors becomes imperative to ensure the smooth execution of teaching and learning activities within the restrictions put in place due to the COVID-19 pandemic. Adaptions and improvements to academic library services during and post-COVID-19 era were successful in ensuring that remote users could obtain similar services and access to information as was the case before the outbreak of the COVID-19 pandemic.