@masterthesis{Nair2025, type = {Bachelor Thesis}, author = {Nair, Abijay}, title = {Development of a CANBUS network for RoboBoat}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23481}, school = {Hochschule Rhein-Waal}, pages = {70}, year = {2025}, abstract = {Autonomous marine vessels have seen use since as far back as World War I, with the Imperial German Navy's Fernlenkboot. Such vessels find more modern use in the logistics sector and as survey vessels, where the continuous operation made possible by autonomy provides distinct advantages in their respective tasks and industries. This thesis presents the development of a CANBUS-based internal network for the RoboBoat autonomous surface vessel project of the Rhine-Waal university of Applied Sciences. It is designed to be the basis on which further development of on-board electronics can take place. It will also act as a platform for further development of the RoboBoat itself, as well as for future projects by students of Rhine-Waal University's FabLab Blue that require such communication architecture. The resulting network is built around the NMEA 2000 standard and off-the-shelf microcontroller boards and network cables, which have enabled quick and fairly robust communication between the different control modules employed in the prototype. NMEA 2000 was chosen owing to its compatibility with modern marine hardware like the multi-function displays, as well as for a simple path to backwards-compatibility with older, less expensive NMEA 0183 hardware. The development process incorporates the systematic process of Engineering Design by G. Pahl and W. Beitz, which guides the processes used here to make various design decisions.}, language = {en} } @masterthesis{Muesch2025, type = {Bachelor Thesis}, author = {M{\"u}sch, Yannick}, title = {Kolmogorov-Arnold-Transformer: Untersuchung hybrider KAN-Transformer-Architekturen im Hinblick auf Skalierung, Effizienz und Interpretierbarkeit}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23376}, school = {Hochschule Rhein-Waal}, pages = {85}, year = {2025}, abstract = {Transformers have become the de-facto standard architecture in Machine Learning, particularly in Natural Language Processing. However, they are computationally expensive to train, with approximately two-thirds of non-embedding parameters residing in Multi-Layer Perceptrons (MLPs). Liu et al. (2024) revisited the long-disregarded Kolmogorov-Arnold representation theorem and demonstrated promising results in specific domains such as symbolic regression and PDE solving. Whether Kolmogorov-Arnold Networks (KANs) generalize to other domains, particularly NLP, remains an open research question. This work systematically evaluates five architecture configurations: MLP baseline, MLP with B-Spline activation, KAN with B-Spline, KAN with Mean aggregation, and Group-Rational KAN (GR-KAN) - across three model sizes (15M, 41M, 124M parameters). Experiments are conducted on text classification ("AG News") and language modeling ("FineWeb") tasks. Identical components (attention, embeddings) isolate performance differences to the feed-forward topology. Statistical validity is ensured through multiple random seeds and parameter-matched comparisons. KAN-based architectures achieve performance parity with MLPs on classification tasks (±0.5 percentage points accuracy). However, they consistently underperform on language modeling, with perplexity increases of +7 to +28 points. Training efficiency overhead ranges from 1.5-2.0× for KANs, while GR-KAN approaches baseline speed. Notably, KANs exhibit significantly higher sparsity (≈ 90\% vs. ≈ 40\%), enabling compression factors of 1.4-1.6× with minimal performance degradation. The MLP+B-Spline control condition reveals that performance deficits stem from KAN topology rather than activation functions. KANs present a viable alternative for discriminative NLP tasks but do not outperform MLPs for generative language modeling. The postulated interpretability advantage through learnable activation functions could not be empirically confirmed - learned functions degenerate to quasi-linear transformations. Future research should explore alternative basis functions and selective hybridization strategies.}, language = {de} } @masterthesis{Mujaya2026, type = {Bachelor Thesis}, author = {Mujaya, Tanaka Peter}, title = {Design and Development of a Computer Vision System for detecting Assembly Errors in LEGO Models}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23367}, school = {Hochschule Rhein-Waal}, pages = {100}, year = {2026}, abstract = {This thesis presents the design and development of a computer vision system for detecting assembly errors in LEGO models within The Digital Factory Game, a production simulation exercise at Rhine-Waal University of Applied Sciences. The system automates quality assurance by replacing manual visual inspection with machine learning-based classification, deployed on edge hardware.}, language = {en} } @masterthesis{Joseph2025, type = {Bachelor Thesis}, author = {Joseph, Balogun}, title = {The Role of Online Marketplaces in Reshaping Agribusiness Sales and Distribution: Evidence from Ibadan and Lagos, Nigeria.}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23358}, school = {Hochschule Rhein-Waal}, pages = {69}, year = {2025}, abstract = {1.1 Context and Background Online marketplaces are applications or web sites on which products are advertised and sold. They are used in the agricultural sector to bring sellers and buyers together. Farmers advertise about their products and customers give their orders. The payment may be done prior or after delivery. The tools enable farmers to sell without traveling. They also decrease the number of intermediaries. There are sites that allow the customer to rate sellers. There are also people sharing updates on delivery or assisting with tracking goods (Eweoya et al., 2021). Trade in farm goods is evolving in Nigeria, owing to online trade. Most of the farmers would sell in open markets only. The prices were frequently low and middlemen absorbed the better part of the profit (Morepje et al., 2024). It is now easier to connect farmers and buyers by way of mobile phones and enhanced internet. Through online, farmers are able to access directly to homes, stores and even export firms. They save time and minimize wastes because customers do not need to travel in order to make an order. This idea was tested with some of the pioneer platforms in Nigeria (Abubakar et al., 2019). Farmcrowdy began as an initiative to finance small growers and match them with consumers. Agricorp constructed a system to exchange crops such as ginger and grains between states. OjaNow also connects small farms with homes and restaurants who desire fresh food. The services are used to assist farmers in selling more as they remain on their farms. Customers are able to shop and deliver farm products to their front door by making an online order. The platforms also assist the farmers to know what is being demanded as well as the price that the buyers are willing to pay. They operate most of them on basic mobile phones and hence can be used even in places with poor internet connections (Ogbeide-Osaretin and Ebhote, 2020). Online market places have changed the way farmers market their produce because they reach a greater distance than the local area. Most farmers used to sell within their towns only, but now digital platforms enable them to sell to the buyers in such cities 2 like Lagos, Abuja and Ibadan. By using such platforms, they are no longer restricted to set days of the market since sales may be carried out at any given time. This flexibility is time-saving, generates more turnover, and allows farmers to deliver food to a larger group of consumers, such as hotels, stores, and schools (Mani and Singh, 2022). Online tools have also decreased the price of selling farm produce in addition to increasing their markets. Farmers will have the opportunity to self-service their prices and deal directly with buyers by eliminating transport costs and bypassing several intermediaries. This enhanced control on prices does not only enhance their earnings, but also promotes fairness and transparency in trade. Buyers, in their turn, have an advantage of price comparison between various sellers prior to acquiring a particular good, which causes more competitive pricing and low fraud potential (Xiao, et al., 2021). The use of online platforms in integrating with local logistics and transport networks has made delivery much faster. Today many platforms collaborate with professional drivers and delivery crews that provide the delivery of the orders and their transportation at the same time. This has been especially beneficial to perishable products like tomatoes, vegetables, and fruits, whose delivery must be done in a timely manner to preserve their freshness and quality. Consequently, fewer produce is wasted as it goes to waste prior to reaching the consumers hence less waste and the general efficiency of the supply chain. Buyers get healthier and fresher food, and sellers save some money and losses of unsold or spoilt products. This credibility also builds trust between the two parties, as customers get to trust merchants with quality goods, and sellers go out of their way to keep their standards (Agrawal et al., 2021). Farmcrowdy, Thrive Agric, OjaNow, and Agricorp are the ones that have introduced similar systems in Nigeria, where both large-scale and smallholder farmers can engage in an effective and efficient digital trade (Akinwale et al., 2023). Most farmers access them via mobile phones as majority of the applications are phone-based. These sites are very popular in cities where there are greater internet connectivity. They are used by farmers to add the name of crops and buyers to make orders. During harvest, when there is plenty in supply, use is greatest. These tools are more acceptable by young farmers compared to the old. They are also common among traders and restaurants as they require constant supply. Information is also 3 useful on online platforms. Farmers are aware of the demand of crops and their price rates. This helps them plan better. They grow what people want. They also make the best decision regarding the time to sell. Buyers utilize the same information to determine whether the prices can increase or decrease. Other websites provide notifications to make fast decisions. This enhances better planning of farmers and reduces waste (Alkahtani et al., 2021). Online sales have simplified the process of farming, but a number of issues also persist. Previous research has concentrated largely on the rural farmers with little attention paid to the use of these digital tools by the smallholder and urban farmers. Visiting the city like Ibadan and Lagos, farmers have special marketing and distribution issues that are not the same as found in the countryside (Sadiq et al., 2025). A large number of these urban farmers work on small pieces of land and depend on fast sales in order to avoid losses after harvest. Nevertheless, they have problems with the lack of storage facilities and the inadequacy of transport systems (Agri et al., 2020). Few studies have examined these particular constraints in the smallholder farmers (Kabo-Bah \& Bannor, 2025). The available research is rather inclined to investigate farmers access to the online platform instead of studying the impact of these tools on logistics, the speed of delivery, or storage (Trang et al., 2024; Olamide et al., 2024). Consequently, some critical questions will go unanswered, including the ways in which digital platforms can mitigate food wastage or make supply more efficient. Smallholder and urban farmers should be given more attention in future studies to offer a complete picture of the effects of online marketplaces (Akanni and Gabriel, 2020). There are numerous impediments in agribusiness in online marketplaces. Trust is a major concern. Online payments are a worry to the farmers since they are not sure that they are safe. Customers are concerned with counterfeits that are sold via these sites. A lot of farmers also cannot access because of poor power supply and weak network. Other farmers are not phone savvy and like their traditional mode of selling (Eweoya et al., 2021). One of the concerns is capacity building. The farmers should be well trained on how to utilize online sales tools. In their absence, a number of them might not enjoy the use of digital platforms. Hence, further research is necessary that will help to investigate the ways these tools are applied in the day-to-day life of farmers and how they can be trained to achieve better results. Online sales can empower farm 4 markets in Nigeria, albeit, its implementation requires farmers to leverage it effectively. It will be possible to understand which areas these platforms are most effective and which ones provide difficulties and use this information to develop more effective training and support programmes by the government and agricultural support groups. It will contribute to reinforcing what is already effective and bridging the gaps that may still exist (Ogbeide-Osaretin and Ebhote, 2020). Therefore, online stores are influencing the movement of food between farms and Nigerian households. They create new sales opportunities and enhance crop planning but there are still numerous problems. The studies are required to observe the impact of these platforms on storage, transportation as well as urban agriculture. An explicit analysis of the areas will inform policies that can empower farmers and enhance food supply.}, language = {en} } @masterthesis{Sueyruege2026, type = {Bachelor Thesis}, author = {S{\"u}yr{\"u}ge, Ey{\"u}p}, title = {Vergleich von FPGA und Mikrocontroller f{\"u}r die Echtzeitverarbeitung von Audiosignalen zur Richtungserkennung mittels FFT und TDOA einschließlich LED-basierter Visualisierung}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23229}, school = {Hochschule Rhein-Waal}, pages = {57}, year = {2026}, abstract = {Das Ziel dieser Arbeit ist es, ein praxisnahes System zur Echtzeitverarbeitung von Audiosignalen mit dem Fokus auf der Richtungsbestimmung von Schallquellen zu entwickeln. Als Hardwareplattformen werden daf{\"u}r ein Field Programmable Gate Array (FPGA) und ein Mikrocontroller (MCU) eingesetzt. Dabei konzentriert sich die Arbeit auf die Untersuchung der beiden Plattformen, wobei die Leistungsf{\"a}higkeit, Genauigkeit und Ressourceneffizienz verglichen werden. Zu diesem Zweck werden Algorithmen der Signalverarbeitung, besonders die Fast Fourier Transform (FFT) und die Time Difference of Arrival (TDOA), implementiert und ausgewertet.}, language = {de} } @masterthesis{Ustritski, type = {Bachelor Thesis}, author = {Ustritski, Nikita}, title = {Competitiveness of Green vs. Traditional Companies: A Comparative Study in the Automotive, Hospitality, and Fashion Industry}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23123}, school = {Hochschule Rhein-Waal}, pages = {80}, abstract = {This thesis compares short-term and long-term competitiveness between green and traditional companies in the automotive, hospitality, and fashion industries. Existing research rarely examines multiple industries together or evaluates competitiveness using both time perspectives, creating a clear research gap. The research question asks how green and traditional companies differ in financial performance, innovation, consumer trust, and regulatory responsiveness. The hypothesis proposes that green companies show lower short-term competitiveness, but higher long-term competitiveness than traditional firms. A qualitative comparative case study approach is applied using secondary data, including annual reports, sustainability disclosures, ESG ratings, and international regulatory frameworks. Six companies are analysed: Tesla and Toyota in the automotive sector, Hilton and Wyndham in hospitality, and Kering and H\&M in fashion. Their competitiveness is assessed through a structured framework that distinguishes between immediate business outcomes and long-term strategic development. The findings show that traditional companies perform better in the short term due to efficient cost structures, established operating models, and stable supply chains. Green companies, however, achieve stronger long-term competitiveness by investing in innovation, transparency, and alignment with tightening regulatory and consumer expectations. Tesla, Hilton, and Kering demonstrate superior long-term performance, while Toyota, Wyndham, and H\&M maintain advantages mainly in the short term. The results support the hypothesis and highlight sustainability as an important driver of long-term competitiveness. The study offers cross-industry insights for academics, managers, and policymakers and identifies opportunities for future research on sustainable business models.}, language = {en} } @masterthesis{AchaleArrahNyama2025, type = {Bachelor Thesis}, author = {Achale Arrah Nyama, Bateatey}, title = {Development of Sustainable Nanofibre Wound Dressing Using Solution Blow Spinning.}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-23107}, school = {Hochschule Rhein-Waal}, pages = {69}, year = {2025}, abstract = {This thesis investigates the production of polylactic acid (PLA) fibrous mats using a low-energy, airbrush-based solution blow spinning (SBS) method with dimethyl carbonate (DMC) as a greener solvent. The study addresses how changes in PLA concentration and air pressure influence fibre formation, fibre diameter distribution, and non-woven mat structure, and examines the implications for incorporating tea tree oil(TTO) and pectin into the resulting materials. PLA solutions in DMC were prepared at multiple concentrations, and the SBS process was evaluated using realistic deposition observations and calculated estimates of material consumption. Fibre morphology was analysed by scanning electron microscopy (SEM), revealing fibrous structures and parameter-dependent differences in diameter distributions and mat heterogeneity. Chemical structure was assessed via FTIR spectroscopy, which indicated that the produced fibre mats were dominated by PLA-characteristic spectral features (and possible DMC contributions), while providing limited evidence for substantial pectin or TTO contributions in the measured spectra under the tested conditions. Thermal analysis (STA) was used to interpret important PLA thermal transitions and to identify possible traces of composites within the fibres. A qualitative hydrophilicity approach based on water droplet wetting behaviour was applied as an initial indicator of surface wettability and potential pectin effects, while recognising its limitations compared to full contact-angle measurements. Overall, the results support SBS with DMC as a viable route to produce PLA fibrous mats and show clear process-structure relationships. However, they also show that better mixing methods and stronger confirmation evidence are needed at every step of the characterisation chain in order for additive incorporation to work.}, language = {en} } @masterthesis{JagjitSingh2025, type = {Bachelor Thesis}, author = {Jagjit Singh, Humjeet Singh Gill}, title = {The effect of shearing in a twin-screw extruder on the mechanical properties of biopolymers (PLA+PBAT) using varied screw designs}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-22950}, school = {Hochschule Rhein-Waal}, pages = {75}, year = {2025}, abstract = {The primary aim of this research project was to determine the effects of shearing on the structure of a blend of biopolymers (PLA+PBAT) through tailored screw designs in a twin-screw extruder. To achieve this, 3 different twin intermeshing co-rotating screw designs of an extruder were used. The designs were mainly focused on the kneading and mixing part of the screw which has a shearing effect on the biopolymer. The experiments were carried out by keeping the temperature, screw speed and feed rate constant. To carry out the experiment, a 60:40 blend of PBAT and PLA was utilized. The compounded pellets produced from the 3 screw designs were then used in a Blown Film machine. Furthermore, tensile test was conducted on the resulting samples. The shear strength and strain of the films were calculated. Tear test of the films were additionally carried out to validate the findings. Fourier Transform Infrared (FTIR) test was conducted to study the functional groups, identifying the chemical interactions and phase compatibility of the of PLA\&PBAT blend. The results showed that the screw with 3 kneading blocks (HS) and 2 kneading blocks (MS) exhibited almost similar ultimate strength and ultimate strain. The screw design with only conveying elements had lower ultimate strength and strain. The kneading blocks played a role on the mixing and shearing of the biopolymer structure in the blend.}, language = {en} } @masterthesis{Malov2025, type = {Bachelor Thesis}, author = {Malov, Ernest}, title = {Development of selection criteria and systematic methodology for the design of novel hydrogels}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-22934}, school = {Hochschule Rhein-Waal}, pages = {82}, year = {2025}, abstract = {Hydrogels are a broad field of materials with applications in many niche areas. To advance the state of the art, research should explore new combinations of chemical and practical components to discover syntheses with novel emergent properties. However, given the vast amount of information available, it is challenging to systematically narrow the search to an effective comparison of a few select candidates. This work proposes a selection and evaluation methodology for building a comparative scope to decide on a set of hydrogel recipe components. Recipe design needs are determined using an existing bibliometric analysis of a hydrogel subfield, and available options are sourced from literature reviews, then assessed based on properties discussed in the literature. Candidates for the crosslinking initiation mechanism, the monomer system, and the crosslinking control method are evaluated separately and sequentially. Recipe permutations are possible as multiple choices are scored. The resultant recipe is a photoinitiated (LAP; irradiated at 380.5nm), thiol-ene (TMPMP-Maleimide), light-patterned (maskless PWM-DLP) hydrogel that swells and bends in response to water due to osmolyte (D-Sorbitol) diffusion. This novel formulation is not currently present in the literature; however, it does not exhibit any chemical or practical incompatibilities. A range of component concentrations, possible failure scenarios, and recommended fixes are provided for initial testing.}, language = {en} } @masterthesis{Kazimli2025, type = {Bachelor Thesis}, author = {Kazimli, Farid}, title = {Origami Applications in Engineering}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:1383-opus4-22848}, school = {Hochschule Rhein-Waal}, pages = {68}, year = {2025}, abstract = {This thesis explores how origami-inspired modular structures—specifically uniformly thick, load-carrying designs—can be prototyped using accessible 3D printing work flows. We focus on translating geometric folding principles into real materials and hinges suitable for repeated deployment by designing printable panels and connectors, and by validating performance with a bar-and-hinge simulation approach. Physical prototypes are fabricated and tested for packing ratio, deployment re peatability, and stiffness-to-mass. Results indicate that thick-panel origami can be produced with consumer 3D printers while retaining compact storage, rapid deployment, and promising structural behavior. The work outlines a practical path toward portable, reusable, and more sustainable engineering systems that minimize material use, support repair and reuse through modularity, and simplify logistics through high packing ratios.}, language = {en} }