Alfred Mwanza
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Alfred Mwanza has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Ibidun Christiana Obagbuwa
Department of Mathematical Science and Computing, Walter Sisulu University, Mthatha 5117, South Africa
2 shared publications - Albert Antwi
Arid Region Water Research Centre (ARWRC), Sol Plaatjie University, Kimberley 8300, South Africa
1 shared publication - Edward A. N. Dakora
Department of Management Sciences, Sol Plaatje University, Kimberley 8301, South Africa
1 shared publication - Lebelo Serutla1 shared publication
Mfundo Monchwe
Sol Plaatje University
1 shared publication- Muthoni Masinde
Central University of Technology
1 shared publication
Latest Publications
- Open Access
Logistics Information Technology and Its Impact on SME Network and Distribution Performance: A Structural Equation Modelling Analysis
Introduction: This study explores the impact of logistics information technology (LIT) on supply chain relationships and distribution performance in small and medium-sized enterprises (SMEs) using South Africa as a case study. Although digital supply chain solutions are increasingly important, there is limited evidence of SME efficiency in emerging markets using LIT. Methods: This study utilises a survey of 313 SMEs from four South African provinces. Bayesian structural equation modelling (Bayesian SEM) was then used to examine LIT’s effects on distribution performances in terms of timeliness, product availability, and condition. Results: The results show that the adoption of LIT strengthens buyer–seller networks (β = 0.524, CI = [0.434, 0.613]) and improves distribution by enhancing both timeliness performance (β = 0.237, CI = [0.098, 0.372]) and product condition performance (β = 0.175, CI = [0.042, 0.259], β = 0.222, p < 0.001). However, it does not directly enhance product availability performance (β = 0.085, CI = [−0.030, 0.199]), signifying that LIT adoption by itself fails to improve product availability. The results also demonstrate that SME network relationships mediate the connection between LIT adoption and distribution performance metrics. Discussion: This study’s findings contribute to the literature and offer valuable information and guidance to policymakers as they underscore the importance for SMEs to invest in LIT integration and compatibility, as well as inventory optimisation and improved supplier communication to minimise transit time variation. Policymakers should support SMEs’ digital transformation through interventions including funding and training for LIT adoption. This study confirms the essential role of LIT in SME supply chains and illustrates that technology-facilitated relationships enhance distribution performance, which enhances SME competitiveness.
Open AccessLogistics Information Technology and Its Impact on SME Network and Distribution Performance: A Structural Equation Modelling Analysis
Introduction: This study explores the impact of logistics information technology (LIT) on supply chain relationships and distribution performance in small and medium-sized enterprises (SMEs) using South Africa as a case study. Although digital supply chain solutions are increasingly important, there is limited evidence of SME efficiency in emerging markets using LIT. Methods: This study utilises a survey of 313 SMEs from four South African provinces. Bayesian structural equation modelling (Bayesian SEM) was then used to examine LIT’s effects on distribution performances in terms of timeliness, product availability, and condition. Results: The results show that the adoption of LIT strengthens buyer–seller networks (β = 0.524, CI = [0.434, 0.613]) and improves distribution by enhancing both timeliness performance (β = 0.237, CI = [0.098, 0.372]) and product condition performance (β = 0.175, CI = [0.042, 0.259], β = 0.222, p < 0.001). However, it does not directly enhance product availability performance (β = 0.085, CI = [−0.030, 0.199]), signifying that LIT adoption by itself fails to improve product availability. The results also demonstrate that SME network relationships mediate the connection between LIT adoption and distribution performance metrics. Discussion: This study’s findings contribute to the literature and offer valuable information and guidance to policymakers as they underscore the importance for SMEs to invest in LIT integration and compatibility, as well as inventory optimisation and improved supplier communication to minimise transit time variation. Policymakers should support SMEs’ digital transformation through interventions including funding and training for LIT adoption. This study confirms the essential role of LIT in SME supply chains and illustrates that technology-facilitated relationships enhance distribution performance, which enhances SME competitiveness.
- Open Access
- 27 December 2024
- Published by MDPI AG in World Electric Vehicle Journal
- Vol. 16 (1) , 9
Using a YOLO Deep Learning Algorithm to Improve the Accuracy of 3D Object Detection by Autonomous Vehicles
This study presents an adaptation of the YOLOv4 deep learning algorithm for 3D object detection, addressing a critical challenge in autonomous vehicle (AV) systems: accurate real-time perception of the surrounding environment in three dimensions. Traditional 2D detection methods, while efficient, fall short in providing the depth and spatial information necessary for safe navigation. This research modifies the YOLOv4 architecture to predict 3D bounding boxes, object depth, and orientation. Key contributions include introducing a multi-task loss function that optimizes 2D and 3D predictions and integrating sensor fusion techniques that combine RGB camera data with LIDAR point clouds for improved depth estimation. The adapted model, tested on real-world datasets, demonstrates a significant increase in 3D detection accuracy, achieving a mean average precision (mAP) of 85%, intersection over union (IoU) of 78%, and near real-time performance at 93–97% for detecting vehicles and 75–91% for detecting people. This approach balances high detection accuracy and real-time processing, making it highly suitable for AV applications. This study advances the field by showing how an efficient 2D detector can be extended to meet the complex demands of 3D object detection in real-world driving scenarios without sacrificing computational efficiency.
- 27 December 2024
- Published by MDPI AG in World Electric Vehicle Journal
- Vol. 16 (1) , 9
Open AccessUsing a YOLO Deep Learning Algorithm to Improve the Accuracy of 3D Object Detection by Autonomous Vehicles
This study presents an adaptation of the YOLOv4 deep learning algorithm for 3D object detection, addressing a critical challenge in autonomous vehicle (AV) systems: accurate real-time perception of the surrounding environment in three dimensions. Traditional 2D detection methods, while efficient, fall short in providing the depth and spatial information necessary for safe navigation. This research modifies the YOLOv4 architecture to predict 3D bounding boxes, object depth, and orientation. Key contributions include introducing a multi-task loss function that optimizes 2D and 3D predictions and integrating sensor fusion techniques that combine RGB camera data with LIDAR point clouds for improved depth estimation. The adapted model, tested on real-world datasets, demonstrates a significant increase in 3D detection accuracy, achieving a mean average precision (mAP) of 85%, intersection over union (IoU) of 78%, and near real-time performance at 93–97% for detecting vehicles and 75–91% for detecting people. This approach balances high detection accuracy and real-time processing, making it highly suitable for AV applications. This study advances the field by showing how an efficient 2D detector can be extended to meet the complex demands of 3D object detection in real-world driving scenarios without sacrificing computational efficiency.
- Open Access
- 24 January 2024
- Published by IntechOpen
Online Assessments in a Changing Education Landscape
COVID-19 threatened to bring education system to its knees across the globe, forcing educators to adopt e-learning as an alternative. While other teaching and learning activities could still go ahead online, the challenge was how would authentic assessments be conducted? E-learning can facilitate administration of online assessments, allowing educators to evaluate students, provide feedback while at the same time providing students with access to assessments from anywhere and at any time. In the post-pandemic era, creativity is likely to become a critical component of online assessments, allowing the use of technology-enabled multimedia tools to enhance assessment experience to enable students various ways to demonstrate their understanding of concepts. This chapter examines role of e-learning, creativity, and technologies in online assessments and their potential to enhance educators’ experience. It discusses challenges and proctoring tools used in, and also online assessments as a future permanent feature of education to support Massive Open Online Courses (MOOCs) and hybrid systems.
- 24 January 2024
- Published by IntechOpen
Open AccessOnline Assessments in a Changing Education Landscape
COVID-19 threatened to bring education system to its knees across the globe, forcing educators to adopt e-learning as an alternative. While other teaching and learning activities could still go ahead online, the challenge was how would authentic assessments be conducted? E-learning can facilitate administration of online assessments, allowing educators to evaluate students, provide feedback while at the same time providing students with access to assessments from anywhere and at any time. In the post-pandemic era, creativity is likely to become a critical component of online assessments, allowing the use of technology-enabled multimedia tools to enhance assessment experience to enable students various ways to demonstrate their understanding of concepts. This chapter examines role of e-learning, creativity, and technologies in online assessments and their potential to enhance educators’ experience. It discusses challenges and proctoring tools used in, and also online assessments as a future permanent feature of education to support Massive Open Online Courses (MOOCs) and hybrid systems.
- 31 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
Assessment of Business Intelligence Fitness in Performing Tasks: Case of a University of Technology, Free State, South Africa
Business intelligence (BI) systems has been adopted in academic institutions and the public sector for decision-making. The literature review shows a lack of studies on BI fitness for task performance. In this study, BI fitness is evaluated in tasks carried out by academics and support staff in University of Technology, Free State, South Africa. A model which indicates task characteristics and technology characteristic fitness integrated to Task-Technology-fit (TTF) model is proposed. A survey research design was considered suitable for this study is the best method for the findings to the entire population. The quantitative research approach was adopted. Online survey questionnaire was developed through SurveyMonkey. An email questionnaire link was sent to 60 academic teaching and support staff. The response rate was 42 out of 60. The data collected were analysed using the IBM Statistical Software for Social Sciences version 27. The analysis was performed using the partial least squares structural equation modelling. (PLS-SEM) The findings confirmed that technology construct had a very high predictive power towards the task technology fit construct. Task construct has a strong relationship with task technology fit construct. These findings reveal that the integrated tertiary system exhibits an excellent task technology fit by supporting specific staff tasks.
- 31 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
Assessment of Business Intelligence Fitness in Performing Tasks: Case of a University of Technology, Free State, South Africa
Business intelligence (BI) systems has been adopted in academic institutions and the public sector for decision-making. The literature review shows a lack of studies on BI fitness for task performance. In this study, BI fitness is evaluated in tasks carried out by academics and support staff in University of Technology, Free State, South Africa. A model which indicates task characteristics and technology characteristic fitness integrated to Task-Technology-fit (TTF) model is proposed. A survey research design was considered suitable for this study is the best method for the findings to the entire population. The quantitative research approach was adopted. Online survey questionnaire was developed through SurveyMonkey. An email questionnaire link was sent to 60 academic teaching and support staff. The response rate was 42 out of 60. The data collected were analysed using the IBM Statistical Software for Social Sciences version 27. The analysis was performed using the partial least squares structural equation modelling. (PLS-SEM) The findings confirmed that technology construct had a very high predictive power towards the task technology fit construct. Task construct has a strong relationship with task technology fit construct. These findings reveal that the integrated tertiary system exhibits an excellent task technology fit by supporting specific staff tasks.
- 12 May 2023
- Published by Springer Nature
- p. 129-153
Coronavirus Lung Image Classification with Uncertainty Estimation Using Bayesian Convolutional Neural Networks
- 12 May 2023
- Published by Springer Nature
- p. 129-153
Coronavirus Lung Image Classification with Uncertainty Estimation Using Bayesian Convolutional Neural Networks