MUT Institutional Repository (IR)

MUT IR serves as the home for the intellectual output of MUT academic communities, and will ultimately include digital dissertations, faculty publications, digital special collections, open access publications, open educational resources and much more.

Browse Communities

Quick access to the main repository content types

9 communities

1. Journal Articles 1337 2. Books and Book Chapters 94 3. Theses & Dissertations 138 4. Conference/ Workshops /Seminar/ Proceedings 87 5. Institutional Memory 42 6. Speeches and Lectures 39 7. Policies / Reports / Statutes 3 8. Projects & Business plan 0 9. Government publications 0

Recent Submissions

Access status: Open Access ,
GRADIENT BOOSTING-BASED RESPONSE SURFACE METHODOLOGY WITH APPLICATION IN MODELLING POTATO POST-HARVEST LOSSES
(Murang’a University of Technology, 2026-08) Kirui, Erick Cheruiyot
Modelling the complex nonlinear systems is one of the most challenging problems in statistics and the conventional response surface models are generally not sufficient in capturing the intricate nonlinear relationship and higher order interaction and advanced machine learning models are highly predictive but have generally not been developed to have statistical interpretability and optimization properties. Therefore, there is no single modelling tool that allows for accurate predictions, interpretable inferences and response optimization of nonlinear data. To overcome this methodological limitation, this study introduced a novel Gradient Boosting Based Response Surface Methodology (GB-RSM) framework, combining the optimization and interpretability properties of Response Surface Methodology (RSM) with the nonlinear learning ability of Gradient Boosting. Specifically, the study aimed at identifying important storage factors that affect potato post-harvest losses using Response Surface Methodology (RSM); Modeling the potato post-harvest losses by a Gradient Boosting model to model nonlinear relationship; Developing a hybrid Gradient Boosting-Response Surface Methodology (GB-RSM) model for better nonlinear modelling; and comparing the performance of the developed model with the standalone RSM and Gradient Boosting models. A mixed method research design was used that combined experimental simulation and a cross-sectional survey. A four factor rotatable Central Composite Design (CCD) experiment was simulated to develop and test the proposed modelling framework and fieldwork was conducted among 628 potato value-chain actors who were sampled using stratified sampling in Nakuru and Nyandarua Counties, which were purposively selected as they are among the major potato growing counties in Kenya and intensive production activities as well as a high number of post-harvest handling activities exists, providing an appropriate setting for the evaluation of the proposed methodology. Data preprocessing was done using K-Nearest Neighbor imputation and normalization. A second order response surface model was fitted to the residuals and a Gradient Boosting model trained on the residuals to create a hybrid GB-RSM model to capture nonlinear structures not accounted for by the polynomial approximation. Furthermore, a multilevel beta regression technique was used to find factors influencing proportional post-harvest losses and Principal Component Analysis (PCA) was used to build a Composite Loss Index (CLI) for integrated modelling. The GB-RSM model was the best among the three models in terms of predictive performance (RMSE = 0.0896 and 𝑅 2 = 0.993) and more model interpretable and optimized with response capability as compared to the other two models. Storage methods, storage duration, market distance, chemical use, transport method and extension training were identified as factors that significantly affect post-harvest losses, but formal education was not significant. The integrated modelling also confirmed that storage duration, potato variety, relative humidity and temperature were the most important factors influencing the Composite Loss Index, and that the curing time did not have a significant effect in the range of the experiment. The study demonstrates that the proposed GB-RSM framework can be a powerful and precise and interpretable method for modelling complex nonlinear systems, while utilizing the complementary strengths of classical statistical modelling and machine learning. The framework is suggested to be implemented for the modelling and optimization of problems with strong nonlinearity in a variety of application areas and to be extended to other machine learning algorithms, highdimensional data and adaptive experimental designs in the future.
Access status: Open Access ,
BRANDING STRATEGIES BY COUNTY GOVERNMENTS ON SERVICE DELIVERY AMONG SMALL AND MEDIUM ENTERPRISES IN BUNGOMA COUNTY, KENYA
(Murang’a University of Technology, 2026-05) Barasa, Didmus Wekesa
A decade after the introduction of devolution in Kenya, county governments continue to face challenges in delivering efficient and satisfactory services to Small and Medium Enterprises (SMEs). Despite the adoption of county government branding strategies such as transparency, accountability, communication, and e-government systems, SMEs in Bungoma County still report concerns regarding service reliability, responsiveness, and trust. This study therefore examined the influence of county government branding strategies on service delivery among SMEs in Bungoma County, Kenya, while testing the moderating effect of political influence. The study was anchored on New Public Management Theory, Public Value Theory, and Institutional Theory, and adopted a cross-sectional survey research design. Data were collected using semi-structured questionnaires and key informant interviews from 372 respondents drawn from SMEs and county government officials whose job description concerns SMEs. Quantitative data were analysed using descriptive and inferential statistics, while qualitative data were analysed thematically. The findings established that government transparency branding, accountability branding, communication branding, and e-government branding all had a positive and statistically significant effect on SME service delivery, where the overall level of service delivery was unsatisfactory. The study further found that political influence had a significant moderating effect that weakened the positive relationship between branding strategies and SME satisfaction. The study concludes that effective county government branding strategies enhance SME satisfaction by improving trust, accessibility, responsiveness, and confidence in public services. The study recommends that Bungoma County Government should strengthen transparency mechanisms, improve accountability systems, invest in responsive communication channels, expand digital service platforms, and manage political interference to enhance service delivery outcomes for SMEs
Access status: Open Access ,
A LIGHTWEIGHT PROOF-OF-STAKE COMPLIANCE ENFORCEMENT MODEL FOR DATA PRIVACY PRESERVATION IN TELEMEDICINE SYSTEMS
(Murang’a University of Technology, 2026-08) Walumbe, Denis Wapukha
Telemedicine expands access to medical care but also generates highly sensitive patient data, creating significant privacy challenges. Existing solutions face three key gaps: limited computational resources that hinder the use of conventional privacypreserving techniques on telemedicine devices; existing Proof-of-Stake (PoS) models, such as Algorand and Ouroboros, are not optimized for telemedicine environments; and HIPAA and GDPR requirements are not directly enforced at the transaction level, leaving consent violations and unauthorized access undetected. To address these challenges, this study aimed to develop a lightweight PoS compliance enforcement model to preserve data privacy in telemedicine systems. The specific objectives were to analyze the features of lightweight PoS models relevant to preserving data privacy; to develop a lightweight PoS voting mechanism based on identified features; to develop a lightweight PoS model incorporating the proposed voting mechanism to enforce data privacy compliance; and to evaluate the performance and effectiveness of the proposed model. The study employs a pragmatic research philosophy and Design Science Research (DSR) design, utilizing both simulated telemedicine device data and 4,999 data points from Kaggle (MTSample) for empirical validation. The study identified four core features of lightweight PoS: voting mechanism, cryptography, storage, and architecture. A voting mechanism based on Byzantine Agreement and cluster pre-consensus, combined with cryptography, was developed to reduce message complexity. A privacy compliance model (M_LPoS) that integrates voting with the Local Privacy Decision Function (LPDF) for regulation was also created. Performance tests showed that M_LPoS achieved high throughput, low latency, and a privacy compliance score exceeding HIPAA’s 75% threshold, while reducing energy use compared against baseline models. M-LPoS optimizes computational efficiency, ensures equitable reward distribution, and embeds privacy within the architectural mechanism. This makes it ideally suited to resourceconstrained telemedicine systems. The study recommends that healthcare providers and telemedicine systems vendors adopt the proposed Lightweight PoS compliance enforcement model (M-LPoS) as a foundational privacy layer in their telemedicine platforms and implement energy-efficient processes. At the same time, policymakers should consider updating privacy standards to incorporate lightweight blockchain technologies. Future studies should validate the model using real-time telemedicine data and conduct longitudinal evaluations of stakeholder participation, decentralized governance, economic viability, and regulatory compliance. Also, further studies should focus on developing adaptive clustering algorithms and cross-jurisdictional compliance to facilitate the worldwide implementation of telemedicine.
Access status: Open Access ,
REMEDIATION OF HOME-TREATED WATER BY REMOVAL OF RESIDUAL ALUMINIUM AND PHENOLIC COMPOUNDS USING GREEN SYNTHESIZED MAGHEMITE (γFe2O3) NANO PARTICLES
(Murang’a University of Technology, 0206-08) Murage, Daniel
Water is an essential natural resource for support of life. To make water safe for human consumption, water needs to be treated. However, small-scale consumers, use home water treatment coagulants obtained from commercial shops. Example of coagulants include 'Rock Alum’ available in Kenyan retail shops. The use of aluminium based coagulants leaves residual aluminium in water. Aluminium exposure through contaminated water results in neurological disorders. Phenolic compounds occur naturally in water mainly from plants and environment. Their presence in drinking water causes muscles weakness and respiratory disorders. This study explored the preparation of maghemite (γFe2O3) nanoparticle by a green method using pineapple (Ananas comosus) peels extract and their application in remediation of treated water by removal of residual aluminium ions and phenolic compounds. The nanoparticles were characterized using X-Ray Diffraction (XRD), Ultra-Violet visible light Spectrometer (UV-Vis) and Fourier Transform Infra-Red spectrometer (FTIR). The maghemite nanoparticles (0.5 g) were used in batch adsorption of residual aluminium ions and phenolic compounds from water samples at varying pH, contact time and temperature. The de-ionized water samples treated with ‘Rock Alum’ coagulants gave residual aluminium ions at concentration of 1.9 ppm when 0.0125 g of coagulant (Rock Alum) was used. This concentration of aluminium ions was higher than permissible limits by the World Health Organization of 0.1 ppm. On characterization of maghemite (γFe2O3) nanoparticles using XRD, FTIR and UV-Vis, it was found that at high calcinating temperature of 400-450 0C gave nanoparticles with improved crystallinity and enhanced ordering of the maghemite structure. The percentage adsorption of aluminium ions was determined using Atomic Absorption Spectrometer (AAS) and UV-Vis for phenolic compounds. It was found that adsorption decreases with increase in pH for aluminium ions. Temperature had little effect on adsorption, while rate of adsorption increased with increase in contact time. Adsorption of aluminium was multilayer and fitted to Freundlich adsorption isotherm. Maghemite was found to be efficient in adsorption of phenolic compounds (Phenol and resorcinol) from spiked de-ionized water samples. Phenol was best adsorbed at pH of 5 and 7 for resorcinol. Rate of adsorption decreased slightly in both phenol (3.6%) and resorcinol (15.2%) with increase in adsorption temperature. At optimum adsorption conditions pH, temperature and contact time, mean adsorption of aluminium, phenol and resorcinol was 75.63, 72.61 and 80.09% respectively. This study recommends use of alternative water coagulants which are not aluminium-based.
Access status: Metadata only ,
INTERACTIVE FINANCIAL RISK MANAGEMENT STRATEGIES AND FINANCIAL SUSTAINABILITY AMONG NON-ROTATING INFORMAL FINANCIAL GROUPS IN KIRINYAGA COUNTY, KENYA
(Murang’a University of Technology, 2026-05) Njeru, Caroline Muthoni
The use of non-rotating informal financial groups (IFGs) as development intervention tools continue to gain popularity among developing countries yet their long-term viability remains under-explored. Despite the widespread uptake, these groups face excessive risk exposure and high rates of premature disintegration. It is estimated that 25% disintegrate due to liquidity constraints and loan defaults. Additionally, members lose 15-25% of savings annually. These trends raise serious concerns about the adequacy of IFGs’ risk management strategies. This study examined interactive financial risk management strategies and their influence on financial sustainability of non-rotating IFGs in Kirinyaga County, Kenya, focusing specifically on internal controls, diversification, financial literacy training, fiscal management, and their interaction effects. Anchored in contingency management theory and supported by social capital, adverse selection and resource-based theories, the study employed an explanatory embedded mixed-methods design. The target population was 60 non-rotating IFGs with 806 members registered at the county department of social services. Data was collected from 424 respondents drawn using a combination of stratified, census, purposive and simple random sampling. Primary data were gathered through questionnaires and key informant interviews, with instrument reliability and validity established through a pilot study in Murang’a County using 10% population. Data was analyzed for descriptive and inferential statistics using STATA version 13. Specifically, descriptive statistics included measures of central tendency and dispersion while inferential statistics drew from correlation and multiple regressions. Pearson correlation coefficient was deployed to test the relationship between risk management strategies and financial sustainability. Interaction-based logistic regression within a generalized linear model framework was employed to predict potential outcomes. The regression yielded positive log-likelihood, an indication of good fit. Results showed that interactions between financial literacy training and internal controls and between financial literacy training and diversification positively and significantly influenced financial sustainability. Conversely, interactions between fiscal management and internal controls and fiscal management and diversification had significant negative effects attributed to overly rigid fiscal controls. Although trends suggest potential synergistic effects between internal controls and diversification, and between financial literacy training and fiscal management, these interactions were not significant. The findings emphasize the complex interplay among risk management strategies. By integrating sound fiscal management and financial literacy training practices within internal controls systems and diversification projects, non-rotating IFGs can foster resilience and long-term financial sustainability. The study therefore recommends solid risk management approach that integrates internal controls, diversification, financial literacy training and fiscal management as their alignment rather than isolated implementation is critical for strengthening resilience and achieving long-term financial sustainability. Research findings provide appropriate data and policy recommendations to help formulate appropriate support programs for development partners, national and county governments. Future studies investigating the design and effectiveness of training programs is recommended.