SUNScholar
SUNScholar is a leading digital archive for the preservation and promotion of the research output of Stellenbosch University.
Communities in DSpace
Select a community to browse its collections.
Recent Submissions
Item
An Economic Analysis of Farm Profitability Related to New Deciduous Fruit Cultivar Choice
(Stellenbosch : Stellenbosch University, 2024-12) Smit, Mareze Nelle; Hoffmann, Willem Hendrik; Labuschagne, Iwan Frederick; Stellenbosch University. Faculty of AgriSciences. Dept. of Agricultural Economics.
South African deciduous fruit production makes a significant economic contribution both to the Western Cape Province and the national agricultural sector. The profitability of this industry affects the livelihoods of many South Africans. Typical to commercial agriculture, deciduous fruit production is exposed to various forms of risk of both natural and social origin. Cultivar diversification is a prominent form of risk management employed by deciduous fruit producers. Varying phenological cycles, i.e. time of flowering and harvest, of different cultivars mitigate risk in both the production and market environment. Although market fluctuations and adverse natural conditions heavily impact the performance of a single cultivar, a diversified cultivar composition ensures the resilience of the farm system. Choosing a cultivar composition suited to managing the specific risk profile of a farming operation, is vital to profitability.
Deciduous fruits are perennial crops requiring a substantial initial capital investment. This demands consideration of all aspects of the environments influencing cultivar success over the long-term. Apples (Malus domestica Borkh.) and Japanese plums (Prunus salicina Lindl.) are the crops used in this study. A coherent decision-making process that acknowledges the complexity of all that influences cultivar profitability of apples and plums had not yet been developed in South Africa. This study set out to determine which profitability drivers for apple and plum cultivars are necessary to consider during a planting decision in specific South African production environments.
Accommodation of complexity and uncertainty was necessary to satisfy the research problem. A systems approach in the form of a whole-farm budget model met this requirement. Four different models were constructed, namely for apples in the Koue Bokkeveld, apples in Elgin, Japanese plums near the town of Robertson, and Japanese plums in the Boland near the town of Wellington, respectively. Model compilation had to consider the unique challenges of the South African deciduous fruit industry. South Africa’s climate and geographical location from main markets creates distinct production and logistical trials. Therefore, a participatory modelling method was used, wherein 27 industry stakeholders participated in interviews regarding the problem of cultivar choice. Discussions were guided by established literature.
After model construction, model validation took place in the form of a multidisciplinary focus group discussion held for apples and plums, respectively. A group of industry stakeholders with diverse backgrounds were asked to evaluate and confirm the logic supporting the incorporation of each cultivar decision-making consideration in the models. Participants of the two discussion groups had to reach consensus regarding the validity of the logical basis of the model. Consensus was achieved in both cases.
The study concluded that a cultivar’s adaptability to an area’s climate is essential for a cultivar to be profitable. Furthermore, a wide range of marketing opportunities lowers the risk associated with profit generation. Participants in this study in both the participatory model development phase and the group discussions strongly agreed that any cultivar characteristic impacting income generation, was far more important in the decision-making process than characteristics affecting production cost.
Item
Holonic software system for coordination of a fruit treatment facility
(Stellenbosch : Stellenbosch University, 2024-12) Smit, Divan; Basson, A. H.; Kruger, K.; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.
The fourth industrial revolution, along with globalisation, has led industries to become more reliant on data for decision-making to maintain market share. The agricultural industry is no exception to these trends, particularly fruit treatment facilities, which increasingly rely on data for their operations. A large portion of the industry is still reliant on tedious human tasks, such as resource coordination and paper-based recording of vital information, which has a large potential for errors.
Therefore, the objective of this thesis is to develop a proof-of-concept Facility Operation Support System (FOSS) based on a holonic systems approach, to coordinate resources in a fruit treatment facility. Coordination in this thesis refers to the resource allocation for facility tasks and the information management that goes with it. Holonic systems have the potential to be a solution to the challenges faced by fruit treatment facilities. Holonic systems separate software entities into holons to represent the physical world in the digital space, allowing them to have autonomy and cooperability.
This thesis follows a design process for holonic systems; therefore, a requirements analysis is conducted to evaluate stakeholders' needs, to formulate a set of functional and non-functional requirements for the FOSS. An architecture is then developed, using holonic system design principles, to satisfy the identified requirements. The architecture uses the Activity-Resource-Type-Instance holonic reference architecture for the identification and classification of holons, separating them into activity and resource holons.
The proposed architecture is implemented in a proof-of-concept FOSS for a case study, to evaluate the extent to which the architecture achieved the requirements. The case study is based on the Hortgro Phytosanitary Laboratory located in Stellenbosch. The Biography-Attribute-Schedule-Execution (BASE) architecture is used for the implementation of the architecture, using the BASE platform.
The functionality and performance of the case study implementation is verified through several experiments. The results of the experiments highlight the ability of the implemented FOSS to satisfy the identified functional requirements, as well as the extent to which the FOSS satisfies the non-functional requirements. The results specifically showcase the ability of the FOSS to coordinate both human and non-human resources for facility tasks. Furthermore, the results show that the FOSS can reallocate tasks and rapidly schedule new tasks should a disturbance (e.g. equipment failure) occur. Following the verification, an overall evaluation of the FOSS concludes that the objective of the thesis is achieved.
Item
CubeSat flight software development with limited access to satellite hardware
(Stellenbosch : Stellenbosch University, 2024-12) Slabber, Dirk Andre; Jordaan, H. W.; Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering.
This thesis demonstrates a method of developing CubeSat flight software earlier in the satellite development timeline. Through this method, more time is permitted for testing, optimization and development of a more robust software product. CubeSats consist of embedded subsystems bought as commercial components. Developers are required to integrate these subsystems into a reliable software system. However, procurement is usually done late in the development timeline. In the modern satellite industry, where missions are developed at a rapid pace, little time is left for full system software verification and testing. The proposed solution is the utilization of digital twins to replace physical components during the early stages of development. Through simulation of components in orbit and interfacing them with flight software on an On-Board-Computer (OBC), the software is continuously tested as if the full satellite is available. Flight software is developed specifically for the DockSat mission. While the software is built upon a commercial software framework, it is tailored with additional software architecture to be inherently robust and support mission-specific subsystems. The first three stages of the mission: commissioning, beaconing and detumbling, are chosen for the scope of the thesis. Additional focus is placed on the development of those software structures which are difficult to test without access to hardware. Digital twins are developed for the D2S2 simulation program running on a desktop computer. Various interfacing methods are explored, with the final design utilizing STM development boards to act as data relays between the OBC and the desktop. Digital twins of the antenna system, radio, Electrical Power System (EPS) and Attitude Determination and Control System (ADCS) are developed using the documentation provided by their
manufacturers. Testing the flight software through this method proved to be convenient and time-efficient. The results show the effects of robustness techniques applied to the flight software. It also demonstrates the feasibility and usefulness of simulated components by performing rapid and repeated testing of various orbital scenarios. Due to these results demonstrating complex and reliable satellite operation, this thesis shows that the use of this method is successful in the facilitation of robust CubeSat flight software development.
Item
Decision Support Tool For The Selection Of The Most Suitable Indoor Localization Technology Using The Analytic Hierarchy Process
(Stellenbosch : Stellenbosch University, 2024-12) Sirin, Havvanur; Von Leipzig, Konrad; Zincume, Philani; Bitsch, Gunter; Stellenbosch University. Faculty of Engineering. Dept. of Industrial Engineering.
The lack of appropriate Decision Support Tools (DST) is a major challenge in the industrial environment where the selection of high-precision Indoor Localization Technologies (ILT) is crucial. Companies often face difficulties analyzing the advantages and disadvantages of different Indoor Localization Technologies due to the wide range of selection criteria involved. These criteria include accuracy, coverage area, power consumption, cost, scalability, response time, and robustness. The extensive range of criteria can lead to uncertainty and sub-optimal investment when an inappropriate decision is made. This research project aims to tackle this challenge by developing a tool that supports users in the industry to easily select the most suitable Indoor Localization Technology based on their specific requirements. The main theoretical implication of this project is the extension of theory concerning the combination of a multi-criteria decision-making method and the Technology Selection (TS). By integrating the multi-criteria decision-making method, the Analytic Hierarchy Process (AHP), the DST systematically evaluates diverse criteria, ensuring a comprehensive analysis of ILT options. This theoretical advancement addresses a critical gap in existing research by providing a structured approach to TS that balances multiple factors. The main practical implication for managers is the proposition of a practical artifact that companies can use to support the decision-making process. The developed tool is designed to be a systematic and well-founded solution, capable of handling the complex decision environment faced by industries. It also allows for the consideration of the subjective evaluations of the decision-makers, ensuring that personal insights and organizational priorities are incorporated into the decision-making process. The tool represents an industryspecific solution designed to meet the requirements of ILT selection, making it highly relevant and applicable in real-world scenarios. Additionally, the tool features a clear navigation system that guides users through the selection process. By providing transparent insights into the AHP decision method, the tool enables companies to make informed decisions with confidence. The project's methodology includes a Systematic Literature Review on Technology Selection and Indoor Localization Technologies, ensuring that the tool is grounded in the latest research and best practices. Overall, this research project not only advances theoretical knowledge in the field of decisionmaking and TS but also provides a valuable, practical tool for industries. The Decision Support Tool developed through this project is poised to become a critical asset for companies, facilitating optimal investment in ILT and enhancing the overall efficiency and effectiveness of industrial operations.
Item
A financial and managerial evaluation of the impacts of adopting radio-frequency identification technology on a conservation agriculture farm in the Swartland.
(Stellenbosch : Stellenbosch University, 2024-12) Sinclair, Henry William; Hoffmann, Willem; Strauss, Johann; Stellenbosch University. Faculty of AgriSciences. Dept. of Agricultural Economics.
Global population growth and urbanisation have increased the demand for food significantly. The global population figure surpassed eight billion in 2022 and two billion more people are projected for 2050. More than half of the population lives in urban areas, a figure expected to rise to 70%, expanding urban areas by 1.2 billion square kilometres. This growth intensifies food insecurity and pressures agricultural resources, necessitating increased food production often encroaching on fragile ecological regions. This increased pressure on agriculture necessitates sustainable farming practices. Fragile ecosystems, like the Amazon and Sahel, are particularly at risk from expansion. Conservation agriculture (CA) is a sustainable solution that reduces greenhouse gas emissions, conserves resources and enhances soil health. It is based on three principles: minimal soil disturbance, maintaining crop residues and crop rotation. CA improves yields, reduces vulnerability to climate shocks and provides broader environmental benefits, like carbon sequestration and reducing water system pollution. CA reduces soil disturbances, while precision livestock farming (PLF) technologies are designed to assist producers in managing livestock in a more efficient and data-driven manner. The contributions of CA towards sustainable production are well-known and proven, while the impacts of PLF and livestock towards sustainable production are less known. This study implements trial data from Langgewens Research Farm and technical PLF data, specifically radio-frequency identification, to evaluate the financial viability of this technology within a CA system in the Swartland region. Agriculture production systems are complex, consisting of various interrelated components. A whole-farm budget model is developed within a systems framework to evaluate and compare two typical production systems aligned within CA principles. As a benchmark for comparison, the model was based on a typical farm in the relatively homogeneous farming region of Middle Swartland. Research began with trial data on a specific crop rotation system — System E — from Langgewens Research Farm. This data was adapted for use in financial analysis and integrated into the typical farm model. Additionally, technical radiofrequency identification (RFID) data on sheep production was incorporated to evaluate and compare the viability of PLF technology within a CA. This study highlights the substantial potential of RFID and PLF technologies in transforming modern agriculture. The detailed analysis and sensitivity evaluations clearly demonstrate that these advanced tools not only enhance gross income per hectare but also offer substantial benefits in terms of efficiency, animal health, overall farm management and profitability and risk mitigation. By integrating technologies within a CA system, farmers can make more informed decisions, optimise resource use and improve their environmental footprint.