Book of Abstracts

Retrieval-Augmented Generation (RAG) has developed into a practical approach for contextualising Large Language Models (LLMs) in external knowledge sources [1]. While earlier research has focused on improving retrieval accuracy, embedding quality, and prompt construction, most existing RAG pipelines implicitly assume that retrieval succeeds. In practice, retrieval frequently fails due to missing data, low semantic relevance, or conflicting sources [2], yet generation continues regardless, producing unreliable or misleading outputs.

This paper identifies the absence of failure awareness as a fundamental architectural gap in current RAG systems. We propose a failure-aware RAG architecture that introduces retrieval confidence scoring, explicit failure mode classification, and structured fallback strategies inspired by reliability engineering. By treating retrieval as a fallible subsystem rather than a guaranteed prerequisite, the proposed design improves robustness, observability, and trustworthiness of RAG-based software systems deployed in real-world environments.

The scalable delivery of healthcare services remains a critical challenge in many regions due to shortages of skilled clinicians and increasing diagnostic workloads. Automated clinical report generation from medical images offers a promising solution; however, existing approaches are often limited by high computational complexity and poor deployability in resource-constrained settings.

In this work, we present an engineering framework for scalable multimodal artificial intelligence (AI) systems for clinical report generation, with a focus on retinal imaging. The proposed approach integrates a SigLIP visual encoder with a MedGemma fusion module and a Gemini 2.5 large language model (LLM), while leveraging Low-Rank Adaptation (LoRA) for parameter-efficient fine-tuning. This design significantly reduces computational and memory requirements, enabling deployment on limited hardware without compromising performance.

Evaluated on the DeepEyeNet dataset, the model achieves substantial improvements over prior non-local attention architectures, including gains in BLEU, ROUGE-L, and CIDEr metrics. Clinical validity is further supported through expert evaluation by certified retinal specialists, confirming the coherence, accuracy, and acceptability of generated reports. Ablation studies demonstrate the effectiveness of attention-focused adaptation in enhancing cross-modal alignment and clinical grounding.

More importantly, the framework enables scalable deployment of AI-assisted diagnostics, making it suitable for real-world healthcare environments where resources are limited. This work highlights how engineering innovation in efficient multimodal AI systems can bridge the gap between advanced machine learning models and practical clinical deployment, providing a viable pathway toward improving diagnostic accessibility and achieving national healthcare impact.

As of 2026, the shift in Artificial Intelligence has moved from experimental “proof-of-concepts” to industrial-scale engineering aimed at national-level transformation. This abstract explores key engineering innovations - spanning infrastructure, autonomous systems, and public service architecture - that are driving significant impact on national economies, security, and social welfare.

Core innovative pillars. Agentic workflows and autonomous systems: moving beyond responsive chatbots, the rise of AI agents (software capable of reasoning and executing multi-step tasks) is revolutionising logistics and urban mobility. This includes the integration of autonomous trucks and drones into national supply chains, significantly reducing costs and energy consumption. AI infrastructure engineering: the development of automated MLOps (Machine Learning Operations) and data pipelines has solved previous bottlenecks. These innovations allow governments to unify siloed legacy data into real-time, high-integrity streams, enabling predictive analytics for national budgeting, fraud detection in tax administration, and real-time structural health monitoring of critical infrastructure. Digital twins and 6D BIM: in civil engineering, the integration of AI with digital twin technology and Building Information Modeling (BIM) allows for the real-time simulation of national assets (power plants, bridges, and transit systems). This results in “self-healing” infrastructure concepts where AI predicts failures before they occur, extending the lifespan of public works.

Domains of national impact. In governance, generative AI for bureaucracy is projected to deliver a 3-4% boost in GDP through administrative efficiency and streamlined public services. In healthcare, biomedical foundation models support diagnosis of rare diseases and predictive health outcomes for infants at a national scale. In national security, AI-integrated security operations centres and copilots enable real-time cyber threat analysis and AI-assisted military logistics and decision-making. In manufacturing, high-tech export optimisation is doubling scientific productivity through AI-powered R&D and automated factory floor management.

The transition to sovereign AI and private cloud models ensures that these innovations remain secure and aligned with national interests. By 2026, the primary challenge has shifted from “novelty” to “practical delivery” - focusing on the engineering pipelines, common standards, and workforce upskilling required to operate AI as a core utility for national progress. The year 2026 marks the “Year of Truth”, where the focus is on intent-driven development: engineers define the desired national outcome, and AI systems autonomously architect the underlying technical solutions.

Sub-Saharan Africa represents the majority of the world population without access to reliable electricity supply, with most of its unelectrified population living in rural areas. It is not viable to extend transmission and distribution facilities to these locations since such communities are far from the existing supply networks, are low-income areas and are sparsely populated. Distributed generation in mini-grids has proven to be a viable pathway for rural electrification. Although significant advances have been made in decentralised energy systems, most deployed systems remain isolated and non-optimal, making them vulnerable to intermittency, demand variation and high operational expenses, which could reduce them to stranded assets.

This study seeks to address these issues by developing a comprehensive framework for the design, optimisation and interconnection of distributed generation within clusters of multi-community mini-grids. Using this network, all involved mini-grids are expected to access a wider range of energy resources, greater operational flexibility and a means of adaptive grid integration without incurring the costs of expanded infrastructure to all settlements.

The methodology is a systematic procedure that begins with evaluation of energy requirement, resources available and features of chosen rural communities. The results of this analysis are used in a two-stage modelling process that combines both system design and economic analysis. The results of that process are then fed into an optimisation framework that tests isolated, interconnected and grid-linked configurations. The process concludes with validation and refinement of the proposed interconnectivity strategy.

It is anticipated that the resultant outcome will provide a scalable framework for rural and peri-urban electrification and facilitate the evolution of isolated micro-grids to grid-compatible distributed energy networks. The preliminary analysis demonstrates the performance of the proposed system during a sample week. The reliability of the system is high with almost zero unmet electrical load, while the sustainability of the system is not adequate: the initial design is DG-poor and import-dependent. To further develop the research, future optimisation should focus on scaling of the PV capacity to achieve feasible renewable penetration. The resulting optimised design will then necessitate the creation of an adaptive grid integration pathway, which entails a bidirectional interface that helps in the necessary imports and exports of power.

Nigeria has witnessed significant growth in solar photovoltaic (PV) installations as part of efforts to improve energy access, reduce dependence on fossil fuels, and support sustainable development. However, the long-term performance and reliability of many solar energy systems have been undermined by inadequate design practices, poor installation standards, insufficient technical capacity, and weak quality assurance mechanisms.

This paper examines the role of engineering quality assurance in improving the sustainability and effectiveness of solar energy projects in Nigeria. Drawing from practical experiences in renewable energy training, certification, and field implementation, the study evaluates common causes of system failures and identifies critical gaps in workforce competence, project supervision, equipment selection, and maintenance practices.

The paper further highlights the contribution of professional certification schemes, technical training programmes, engineering standards, and regulatory oversight in improving project outcomes. A structured Quality Assurance Framework is proposed to guide stakeholders involved in solar energy development, including engineers, contractors, training institutions, regulators, and policymakers.

The study concludes that strengthening quality assurance systems within the renewable energy sector can significantly improve project reliability, enhance investor confidence, reduce lifecycle costs, and accelerate progress toward national energy access and sustainability targets. The findings provide practical recommendations for supporting engineering innovations capable of delivering measurable national impact through clean energy deployment.

Cross River State, in Nigeria's South-South geopolitical zone, faces an acute rural electrification crisis: by official accounts, only three of its eighteen Local Government Areas enjoy consistent public electricity supply. A defining structural peculiarity compounds the challenge. The State is not served by a single distribution company but is fragmented across three: the Southern senatorial district is fed by the Port Harcourt Electricity Distribution Company (PHEDC), the Central district draws from the Enugu Electricity Distribution Company (EEDC), and the Northern district is connected to the Jos Electricity Distribution Company (JEDC).

This paper examines the technical, economic, institutional and geographic barriers to rural electrification, situating the analysis within Nigeria's new legal framework - the Electricity Act 2023 and the March 2023 constitutional amendment that devolved electricity regulation to the States - and the State-level response that establishes the Cross River State Electricity Regulatory Commission (CRSERC), the Cross River State Electrification Agency (CRSEA), and the Cross River State Independent System Operator (CRSISO).

It argues that decentralised solar mini-grids, solar home systems and small hydropower constitute the most viable pathway to universal access, and proposes a phased short-, medium- and long-term roadmap toward the State's target of 95% access by 2028.

The construction industry is a major contributor to global carbon emissions, particularly through the production of Portland cement. This study evaluates the environmental and economic benefits of geopolymer concrete (GPC) as a sustainable alternative to Portland Limestone Cement (PLC) within Nigeria's construction sector. A mixed-method approach was adopted, combining primary data from 100 construction stakeholders collected via structured questionnaires with secondary data from existing literature.

Findings indicate strong stakeholder agreement regarding the environmental advantages of GPC, with 73-78% affirming its potential for significant CO₂ emission reduction, energy savings, and industrial waste utilisation. Economic perceptions were more varied: while 55% acknowledged long-term maintenance cost savings, only 29% viewed initial costs as favourable. The primary barrier to adoption identified was lack of awareness (36%), followed by limited policy and regulatory frameworks. Development of national standards emerged as the most recommended policy intervention (35%).

Comparative analysis shows that GPC can achieve up to 76% lower CO₂ emissions, 60% reduced energy consumption, and approximately 15% lower material costs relative to PLC. The study recommends targeted policy reforms, professional training programmes, and pilot infrastructure projects to accelerate adoption. Although findings highlight strong sustainability potential, limitations include urban sampling bias. Future research should focus on experimental performance testing and rural implementation feasibility.

The use of bio-based materials as alternatives to oilfield chemicals has gradually gained attention among researchers due to their eco-friendly nature. In this research, the potential of coconut coir dust as a green surfactant was studied to ascertain its efficiency at reducing interfacial tension and recovering residual oil using the mechanism of surfactant flooding in sandstone reservoirs, as well as to determine its adsorption on sandstone reservoirs.

Phase behaviour studies were conducted to determine the stability of the system at 100 °C and its compatibility with divalent ions. The surfactant performance was further evaluated through analysis of the microemulsion system, interfacial tension (IFT), optimal salinity, and critical micelle concentration. Oil displacement experiments were carried out using an oil-wet sandstone core under reservoir conditions. Additionally, the adsorption behaviour of the surfactant on the core was examined at ambient conditions (27 °C) using the Langmuir, Freundlich, Temkin, and Linear isotherm models. The adsorption kinetics were also analysed using pseudo-first-order, pseudo-second-order, intraparticle diffusion, and Elovich models.

Results showed strong fluid-fluid compatibility and the formation of a bi-continuous microemulsion across a range of reservoir temperatures, with a low interfacial tension of 1.86 mN/m. Surfactant flooding achieved an additional oil recovery of 42.5%, highlighting the effectiveness of this natural surfactant in enhancing heavy oil recovery. The Langmuir isotherm model exhibited the highest correlation coefficient (R² = 0.897), indicating that surfactant adsorption onto the sandstone core occurs at specific, homogeneous sites and that at higher surfactant concentrations further adsorption is restricted.

The high R² values obtained from most kinetic models suggest strong adsorption capacity of the sandstone core and a significant affinity for the surfactant, particularly at lower concentrations. This implies that at low surfactant concentrations, adsorption within the reservoir may limit the efficiency of the EOR process. Therefore, injecting the surfactant (coconut coir dust) at higher concentrations is necessary to achieve optimal recovery performance.

Construction is a field that has been challenged over time in terms of low productivity, high rates of safety risks (20% of the workplace deaths in the United States as postulated by the Bureau of Labor Statistics), and labour shortages, estimated at 2.2 million employees by 2030 according to Associated Builders and Contractors. To solve these problems, autonomous machinery - robotic excavators, robotic bricklayers, drones and robots in 3D printing - use artificial intelligence (AI), computer vision and Building Information Modeling (BIM).

This study examines the opportunities of integrating autonomous machinery into BIM and human-robot collaboration (HRC) to make a construction environment, despite being unstructured, more productive, safer, and sustainable. The mixed-method design of the study, including both a comprehensive literature review and a detailed case study analysis combined with simulation-based assessment using the Robot Operating System (ROS), is suggested to introduce a BIM-based HRC model to streamline the use of robots. The reliability of robotic systems was evaluated with the help of Kaplan-Meier survival analysis, and the adoption trends were forecasted using a random forest regression model.

Results show productivity benefits of 25-30%, accident reductions of 40-50%, and material waste reductions of up to 40%, but barriers exist in the form of expensive costs, environmental flexibility, staff opposition and regulatory loopholes. The outcomes of the empirical work agree well with what simulations predicted, yet implementation in real life is 10 per cent behind because of socio-economic factors. The paper suggests high-level AI-based independence, retraining schemes, worldwide HRC building security, and policy models to encourage its use, which can be included in the Construction 5.0 agenda of profitable, robust, and sustainable construction operations.

Nigeria holds one of the largest gas reserves globally, yet its development outcomes have lagged behind comparable resource economies - a gap driven more by governance than by geology or geography. This paper argues that Liquefied Natural Gas (LNG) development in Lower Southern Nigeria can be structured to avoid the governance failures historically associated with crude oil extraction if it is treated not merely as an export project, but as an integrated development platform.

Drawing on professional experience across LNG and gas value chain operations in Texas and Mexico, alongside long-term observations from Akwa Ibom, Rivers, Delta, Ogun and Lagos States, the paper proposes a strategic seven-pillar governance model. The model is built around traceability, industrial co-location and technology specialisation, community benefit-sharing, domestic value retention, environmental accountability, governance verification, and strategic workforce development.

The framework is applied to comparative industrial-cluster models in Qatar, Singapore, Shenzhen, and South Korea, and aligned with emerging regional infrastructure opportunities including the West Africa Gas Pipeline (WAGP) and the Gulf of Guinea Coastal Road and trans-regional rail initiatives. The central proposition is that LNG facilities can anchor “prosumer” industrial ecosystems - integrated systems in which facilities simultaneously consume and generate energy, feedstock, services, skills and trade.

In this model, LNG is not presented as inherently sustainable; rather, it becomes developmentally effective only when contracts, infrastructure, skills policy and community arrangements are deliberately structured to produce shared value. The paper further contends that future corridor developments in Lower Southern Nigeria should integrate industrial assets with liveable communities, high-quality social infrastructure, technical training systems, applied research and regional service export capacity. The result is a practical and scalable governance framework that aligns LNG development with regional industrialisation, environmental transition and West African trade integration.

The development of bio-inspired soft robots is an emerging research area which focuses on merging the flexibility and safety of biological systems with those of soft robots. Drawing their inspiration from the motion and manipulation principles exhibited in biological organisms, soft robots have shown a lot of promise in applications related to environmental monitoring and surveillance, industrial and agricultural applications, healthcare monitoring, and interactions between humans and robots.

Yet, the extensive use of such soft robotic systems is restricted due to the costly nature of the sensing, actuation, and control systems that are needed in their development. Specifically, in most cases, the use of high-performance computing and precise controls requires specific and costly hardware.

This study focuses on analysing the potential role of cheap, commercial embedded control systems that would allow for the development of bio-inspired soft robotics platforms that are affordable and scalable. The analysis covers recent advances in embedded sensing, microcontroller-based control systems, and soft actuation technologies, with emphasis on resource-constrained implementations. In addition, the paper discusses the way that system design concepts such as distributed sensing and light-weighted control loops can help in increasing efficiency and reducing cost. Some of the limitations in the process, such as non-linear dynamic systems, real-time control, energy consumption, and hardware limitations, have also been evaluated with respect to proposed solutions.

By identifying current trends, opportunities, and implementation barriers, this work highlights pathways toward practical, cost-effective bio-inspired soft robots that can address real-world challenges, particularly in developing and resource-constrained environments.

Maritime transportation underpins the United Kingdom's economic prosperity and international trade, yet increasing pressures from supply chain disruptions, decarbonisation policies, port congestion, adverse weather conditions, and geopolitical events have exposed the vulnerability of maritime logistics systems. While Automatic Identification System (AIS) based maritime intelligence has achieved significant advances in vessel trajectory prediction, anomaly detection, traffic monitoring, and port analytics, existing studies remain predominantly vessel-centric and rarely investigate the broader interactions between maritime operations, economic activity, and environmental sustainability. This limitation hinders the development of intelligent decision-support frameworks capable of addressing emerging challenges associated with maritime resilience and net-zero transitions.

This work introduces a novel Socio-Economic-Aware Maritime Intelligence (SEAMI) framework for analysing interdependencies among vessel operations, economic performance, and environmental outcomes within the United Kingdom's maritime domain. Using AIS-derived vessel trajectories from UK ports between January 2020 and April 2026, combined with economic indicators, port operational statistics, weather observations, and OECD maritime emissions data, the proposed framework investigates how economic activity influences vessel traffic dynamics and how operational behaviour subsequently affects congestion, fuel consumption, and greenhouse gas emissions.

The framework extracts vessel mobility, voyage, and traffic features from AIS data and integrates them with macroeconomic indicators, including Gross Domestic Product (GDP), trade volume, industrial production, and port performance metrics. To capture complex interactions among ports, vessels, economic regions, and environmental systems, a dynamic graph-learning architecture is proposed alongside explainable artificial intelligence and causal discovery techniques. Additionally, this research evaluates conventional machine learning, deep learning, graph neural network, and transformer-based approaches for predicting maritime congestion, traffic density, emissions, and economic impacts. Explainability and causal analyses are employed to identify key drivers of congestion and emissions and to quantify the influence of economic growth, trade activity, and weather conditions on maritime operations.

By moving beyond traditional vessel-behaviour modelling, this work establishes a new paradigm for sustainability-aware maritime intelligence, supporting evidence-based policy development, resilient port operations, and strategic planning within the UK maritime sector. The proposed SEAMI framework contributes to the development of intelligent maritime transportation systems that simultaneously address economic efficiency, environmental sustainability, and transportation resilience.

Language and communication barriers, especially in multilingual and multicultural settings like Nigeria, discourage adoption and success of IT solutions in organisations although they are not limited by technological factors. Some projects do not produce expected results despite the massive investments in IT systems, and this situation is because of the poor understanding of the system by users, lack of effective communication, and poor documentation. This paper explores the relationship between language proficiency, clarity of communication, cross-cultural communication practices, and quality of documentation and adoption, performance and user satisfaction of IT solutions.

The research design was quantitative, and the surveyed population consisted of 300 IT professionals, end-users, project managers, and stakeholders of different organisations in Imo State, Nigeria. The structured questionnaire was validated as both content and reliability (Cronbach's alpha = 0.87) and used to collect data. Descriptive statistics (means and standard deviations) and inferential statistics (multiple regression analysis) were used to test research questions and null hypotheses at 0.05 level of significance.

It has been found that language proficiency (β = 0.40, p = 0.001) and communication clarity (β = 0.33, p = 0.001) are significant predictors of IT adoption and effectiveness, whereas cross-cultural communication (β = 0.28, p = 0.001) and documentation quality (β = 0.37, p = 0.001) are important predictors of user satisfaction and IT success.

These findings support the use of human-centred communication approaches in IT implementation. The research has practical implications for IT managers and policy makers and emphasises the importance of training, clear documentation and culturally sensitive communication. It is advisable that organisations should adopt these practices into digital transformation and project management plans to increase IT adoption, efficiency, and user satisfaction.

The lack of proper consideration of language inclusivity and effectiveness of communication in engineering innovations can make national development projects less effectively sustainable and less accepted by communities. Although the significance of human-centred approaches is acknowledged, empirical research quantifying the impact of language and communication on project success is still scarce, especially in Nigeria. This paper set out to analyse the degree to which the accessibility of language and effectiveness of communication can impact community acceptance and the sustainability of a project within the context of national development engineering opportunities.

The research design was quantitative and a sample size of 300 stakeholders - including engineers, project managers, government officials, contractors and community representatives in Imo State, Nigeria - was identified using stratified random sampling. A structured questionnaire was used to collect the data, and descriptive statistics (mean, standard deviation, frequency) and inferential statistics (Pearson correlation and regression analysis) were applied at the 0.05 level of significance.

Results indicated moderately high scores of human-centred practices (M = 3.72, SD = 1.10) and language inclusivity (M = 3.71, SD = 1.11). Communication effectiveness, however, was weakly and not significantly related to project sustainability (r = 0.05, p > 0.05), and language accessibility was weakly related to community acceptance (β = 0.08, R² = 0.01, p > 0.05).

The research suggests that multifaceted methods should be used in sustainable engineering projects, combining policies and practices such as language inclusion, participatory involvement, and culturally responsive communication. It recommends adoption of policies, capacity building and structured interactions with the community to improve project acceptance and sustainability.

The construction industry in many emerging economies continues to face operational challenges associated with fragmented communication, poor site documentation, delayed reporting, weak accountability systems, procurement inefficiencies, and inadequate workforce coordination. These challenges frequently contribute to project delays, cost overruns, quality issues, and reduced transparency throughout the project delivery process.

This paper explores the role of digital construction management systems in improving construction site operations, with particular focus on the development framework of SiteVisi, a mobile-based construction monitoring and project visibility platform designed for the Nigerian construction industry. The platform integrates real-time site reporting, project documentation, workforce coordination, procurement tracking, supervision workflows, and communication systems into a centralised digital ecosystem aimed at enhancing operational efficiency and project oversight.

The study highlights how mobile-enabled engineering platforms can improve access to project information, streamline reporting processes, strengthen accountability, and support data-driven decision-making across multiple phases of construction projects. Particular attention is given to the practical realities of implementing digital construction solutions within developing economies where informal operational structures, infrastructure limitations, and limited technology integration remain prevalent.

The paper further examines the broader implications of construction digitisation for national infrastructure development, engineering productivity, project governance, and artisan workforce management. It argues that locally developed construction technology solutions can play a significant role in modernising project delivery systems, improving transparency, and enhancing the efficiency of the built environment sector. SiteVisi is presented as a practical case study demonstrating how engineering-driven digital innovation can contribute to sustainable construction management practices and support the transformation of construction operations in emerging markets.

National infrastructure systems such as power, transportation, water, and industrial networks are increasingly dependent on complex engineering data flows. However, many countries continue to face fragmented data environments, inconsistent standards, and unclear institutional responsibilities that weaken system reliability and public trust. As engineering systems become more digital and interconnected, the absence of trusted data governance frameworks creates operational risks, reduces efficiency, and limits the effective deployment of advanced technologies such as AI, automation, and digital twins.

This paper introduces an Engineering Data Governance Assessment Framework (EDGAF) developed through the analysis of global best practices in data governance, digital public infrastructure, and engineering system management. The framework evaluates governance maturity across five pillars: policy and regulatory alignment; institutional accountability; interoperability and standards; data quality and lifecycle management; and rights-based safeguards.

We hypothesise that weak or inconsistent governance across engineering data ecosystems contributes directly to infrastructure failures, inefficiencies, and reduced resilience. Findings demonstrate how governance gaps - such as unclear ownership, poor data quality controls, and limited interoperability - undermine predictive maintenance, cross-agency collaboration, and responsible AI adoption. The proposed framework offers governments, utilities, and engineering organisations a practical pathway for strengthening national infrastructure through trusted, transparent, and well-governed engineering data, ultimately enhancing resilience, enabling innovation, and building public trust at national scale.

Many Internet of Things (IoT) systems deployed in developing regions face significant operational challenges due to unstable power supply, intermittent internet connectivity, limited maintenance access, and harsh environmental conditions. These constraints are particularly evident in applications such as flood monitoring, precision agriculture, smart energy metering, pipeline surveillance, and critical infrastructure monitoring across Nigeria. While artificial intelligence (AI) can improve decision-making in such systems, conventional cloud-dependent approaches often increase communication overhead, latency, and energy consumption, reducing their suitability for long-term field deployment.

This study investigates a memory-aware Edge AI framework designed to improve the energy efficiency and reliability of low-power IoT devices. The proposed approach minimises unnecessary data movement between sensors, memory, processors, and cloud services by incorporating local data filtering, event-driven processing, feature extraction, and lightweight inference directly on embedded edge nodes. Rather than transmitting continuous streams of raw sensor data, the system selectively processes and communicates only relevant events or compressed information, thereby reducing communication and computational costs.

A representative Nigerian use case, such as flood-level monitoring or agricultural condition sensing, is considered to evaluate system performance. Key performance metrics include energy consumption, communication bandwidth utilisation, inference latency, memory utilisation, and device operational lifetime. The framework is implemented using resource-constrained microcontroller-based platforms commonly used in IoT deployments and compared against conventional cloud-centric architectures.

The expected outcome is a practical engineering framework that demonstrates how memory-aware Edge AI can extend battery life, reduce communication requirements, improve response times, and enhance deployment reliability in low-resource environments. The work contributes toward the development of scalable and sustainable intelligent monitoring systems capable of supporting national priorities in environmental monitoring, agriculture, energy management, and critical infrastructure protection across Nigeria.

This study analyses road fatalities in Nigeria, from a spatial point of view. Road traffic data from 1990 - 2023, was sourced from the Federal Road Safety Corps (FRSC). Global Moran's I was used to establish if there is any form of cluster in the number of road traffic fatalities per 100,000 population, and AHC clustering algorithm was used to group states with similar road traffic fatalities. Local indicator of spatial association (LISA) was used to establish statistically significant local clusters of road traffic fatalities.

The distribution of the road traffic fatalities in Nigeria shows that the North experiences more road traffic fatalities compared to the Southern part of the country. The result from the global Moran's I gave a positive significant value indicating that neighbouring states have similar high or low road traffic fatalities per 100,000 population. Five zones were identified using the AHC algorithm and the result showed that most states in very high or high fatalities zones per 100,000 population were located close to either Lagos state or Abuja. The LISA cluster map showed that Abuja and its neighbouring states form a hotspot of very high road traffic fatalities per 100,000 population while Rivers, Imo, Abia and Akwa-Ibom State also form a hotspot of low road traffic fatalities per 100,000 population.

The cost of living in Abuja has encouraged migration to more affordable neighbouring states, where road users commute very long distances to the city, thus contributing to the high road traffic fatalities on the highways linking Abuja.

This paper proposes a structured Train-the-Trainers (TTT) STEM deployment model for primary school teachers in Lagos State, Nigeria, implemented through a strategic collaboration with the Association of Professional Women Engineers of Nigeria (APWEN), Lagos Chapter. Based on the urgent need to bridge significant gaps in foundational STEM pedagogical capacity across Lagos State, the framework offers a scalable, district-based approach to teacher empowerment, innovation culture development, and girl-child inclusion.

Structured across three implementation phases spanning three years and aligned with Lagos State's six educational districts, the model integrates teacher professional development, APWEN mentorship networks, locally adaptable STEM kits, school-based innovation clubs, and a robust monitoring and evaluation system. Expected outcomes include improved STEM teaching quality, increased student innovation participation, stronger female representation in engineering pathways, and a replicable model for Sub-Saharan Africa.

Nigeria's healthcare system faces persistent challenges including inadequate infrastructure, high burden of infectious and non-communicable disease, limited diagnostic capacity and heavy reliance on imported medical technologies. Biomedical engineers and biomedical scientists provide a critical pathway for addressing these challenges through translational therapeutics that convert laboratory discoveries into clinically deployable solutions.

This paper explores engineering-driven biomedical innovation tailored to Nigeria's healthcare needs, focusing on low-cost diagnostics, point-of-care technologies, locally fabricated medical devices, digital health systems and regenerative medicine. It also highlights the role of biomedical sciences in understanding disease mechanisms relevant to Nigeria's epidemiological profile. Key barriers including limited funding, brain drain, weak industry-academia collaboration and regulatory inefficiencies are discussed. Finally, strategic recommendations are provided to strengthen Nigeria's biomedical innovation ecosystem and improve national health outcomes.

The increasing penetration of distributed energy resources (DERs) and electric vehicles (EVs) presents both opportunities and operational challenges for modern active distribution systems. This paper develops a deterministic Active Network Management (ANM) framework to optimally coordinate DERs and EV aggregators within a distribution network. The primary objective is to minimise the total operational cost of the distribution system operator (DSO) while ensuring network security and efficient utilisation of renewable generation.

A deterministic optimal power flow (DOPF) model is formulated, incorporating detailed representations of EV aggregators with diverse temporal availability patterns, alongside network, generation, and operational constraints. The proposed framework is implemented on a modified IEEE-33 bus system to evaluate system performance under coordinated EV charging and discharging strategies. Results demonstrate that EV aggregators significantly enhance operational flexibility, leading to reduced grid dependency, improved voltage profiles, and substantial reductions in renewable energy curtailment and peak demand. In particular, aggregators with availability aligned to renewable generation periods exhibit superior performance.

To further assess the robustness of the model, a comprehensive sensitivity analysis is conducted on key parameters, including curtailment penalty, EV penetration level, state-of-charge (SOC) limits, electricity price signals, and renewable generation capacity. The findings reveal that system performance is highly sensitive to these parameters, with EV penetration and curtailment penalties exerting the most significant impact on cost efficiency and renewable utilisation. Additionally, diminishing returns are observed beyond certain EV capacity levels, while tighter SOC constraints limit flexibility and increase system costs.

The study highlights the critical role of EV aggregators as distributed flexibility resources in enabling cost-effective and reliable operation of DER-rich networks. The proposed framework and insights from sensitivity analysis provide valuable guidance for system operators and policymakers in designing flexible and resilient future distribution systems.

The poultry industry constitutes a significant proportion of Nigeria's national GDP where millions of birds are raised annually, but the industry is not immune to the energy challenge of the country. Though a tropical climate eliminates building heating requirements, intensive livestock production requires year-round heating for animal production. Day-old chicks require 32-35 °C ambient temperature, which creates about a 10-15 °C heating gap relative to the ambient temperature conditions of 20-25 °C. The heating demand is estimated at about ₦35 billion annually across the sector, which at present lacks validated engineering estimation tools, resulting in oversized systems, fuel waste and reduced farm profitability.

The thrust of this study is the validation of a degree-days model for agricultural heating load estimation in a tropical climate. The model was developed with dry-bulb temperatures of 18 Nigerian cities based on the Cumulative Distribution Function (CDF) of ambient temperature. Validation of the model is carried out using 10-year ambient temperature data from Abeokuta and Osogbo in south-west Nigeria, achieving good predictive accuracy (R² > 0.99; MAPE of 3.56 and 5.86 for Osogbo and Abeokuta respectively) on average at 32 °C base temperature for poultry brooding, and annual heating degree days of 1,662.35 °C·day and 2,166.14 °C·day for Abeokuta and Osogbo respectively, which demonstrate substantial year-round heating requirements even in a tropical climate.

Applications include sizing of heating systems, estimation of energy consumption, computation of fuel cost, and renewable energy potential assessment. National-level analysis projects potential annual savings through optimised heating system design. The validated model provides essential engineering tools for farmers, agricultural engineers, and energy planners, directly impacting farm economics, food security, and agricultural sustainability.

Reliable electricity demand forecasting remains a major challenge in Nigeria's power sector. Demand estimates are often based on historical averages and fixed assumptions that do not reflect changes in consumption across different regions, seasons, or periods of the day. This contributes to poor load planning, unnecessary outages, and inefficient use of the limited electricity available on the national grid.

This study explores the use of machine learning techniques to improve short-term and medium-term electricity demand forecasting within Nigeria's distribution network. Historical load data are combined with factors such as temperature, time of day, day of the week, public holidays, and seasonal variations to develop predictive models. Four machine learning algorithms - Linear Regression, Random Forest, Gradient Boosting, and Artificial Neural Networks - are evaluated and compared.

The models are assessed using Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and R-squared (R²) to determine their forecasting performance. The study also investigates whether models trained using region-specific data produce more accurate forecasts than models developed from aggregated national data.

The aim of this research is to demonstrate that machine learning can provide a practical approach to electricity demand forecasting using data that are already available to electricity distribution companies. Better demand forecasts can support improved load scheduling, reduce energy wastage, enhance operational planning, and contribute to a more stable and efficient electricity supply in Nigeria. The findings are expected to provide useful insights for utilities and policymakers seeking data-driven approaches to power system planning.

Compressed natural gas vehicles (CNGV) have been introduced in Nigeria as a low-carbon transitional technology to support the transport sector in transitioning to cleaner energy sources (ETP, 2022). However, the diffusion of this vehicle technology depends on socio-technical factors such as strong market formation, policy consistency, adequate infrastructure, high vehicle conversion rate, public perception and stakeholder alignment, among others (Ogunlowo, Bristow and Sohail, 2017). The question then becomes: can CNGV effectively support Nigeria's transition towards lower-carbon mobility?

The aim of the study is to assess the readiness of Nigeria's socio-technical systems for the adoption of CNGV by measuring key indicators such as infrastructure readiness, industry readiness, technical expertise, policy strength, public acceptance, consumer trust, and workforce capability (Geels, 2002). Secondary data from national and international sources related to these indicators will be systematically collected and analysed.

The findings of this study will provide empirical insights into Nigeria's readiness for CNGV adoption and inform future research on large-scale adoption pathways. This study is expected to contribute to the existing literature on lower-carbon transportation adoption in developing countries.

Electricity billing in Nigeria remains a major source of consumer dissatisfaction, utility revenue loss, and regulatory concern. Although metering has improved, a large proportion of electricity customers still remain exposed to estimated billing, weak billing evidence, and disputes over consumption records. This paper examines smart metering as a tool for improving transparent energy billing in Nigeria, with attention to its technical, regulatory, financial, and consumer protection implications.

The study adopts a narrative review approach, drawing on regulatory reports, policy documents, World Bank programme materials, and scholarly literature on smart metering, prepaid metering, non-technical losses, and advanced metering infrastructure. The findings show that smart metering can improve billing transparency by enabling accurate measurement, remote data retrieval, tamper detection, prepaid and postpaid audit trails, and stronger complaint resolution. However, implementation is constrained by meter financing, foreign exchange pressure, interoperability risks, weak communication infrastructure, cyber and data privacy concerns, consumer distrust, and uneven enforcement.

The paper argues that smart metering should not be treated only as a hardware deployment exercise. Rather, it should be implemented as part of a wider advanced metering infrastructure strategy that includes customer enumeration, meter data management systems, regulator-facing data platforms, cybersecurity controls, consumer education, and transparent financing models. The study concludes that smart metering can significantly support transparent electricity billing in Nigeria, but only when integrated with institutional reforms that strengthen accountability, trust, and data-driven regulation.

The growing concerns over climate change, environmental degradation, and the depletion of fossil fuel reserves have accelerated the global transition toward renewable energy sources. Biomass has emerged as one of the most promising renewable energy resources due to its availability, sustainability, and potential to reduce greenhouse gas emissions. Despite the increasing interest in biomass-based renewable energy systems, there remains a lack of comprehensive integrated energy-exergy reviews, limited evaluation of their fossil fuel offset potential, and inadequate comparative sustainability assessments across different biomass conversion technologies.

This review examines biomass-derived renewable energy systems with emphasis on their energy and exergy performance, conversion technologies, and capability to offset fossil fuel consumption. The study reviews major biomass conversion methodologies including thermochemical processes such as combustion, gasification, and pyrolysis; biochemical processes such as anaerobic digestion and fermentation; and physicochemical processes such as transesterification for biodiesel production. The thermodynamic performance of these technologies was critically evaluated using findings on energy and exergy analyses from previous studies to determine system efficiencies, irreversibility losses, and overall sustainability.

The effectiveness of biomass renewable energy technologies in reducing dependence on conventional fossil fuels in electricity generation, transportation, domestic heating, and industrial applications was also evaluated. The review demonstrates that integrating advanced biomass conversion systems with exergy-based optimisation strategies can substantially improve energy efficiency, reduce greenhouse gas emissions, and enhance long-term energy security through partial displacement of fossil fuel resources.

In recent times, the rise in the population of people has skyrocketed globally, which has consequently resulted in an increase in waste generated and a high number of vehicles on the road surface. The environmental and public health issues posed by these generated wastes have made researchers lean towards incorporating them in construction materials. This study investigated the potential use of Bean Pod Ash (BPA), a derivative of agro-wastes, as filler in asphaltic concrete mixtures.

From the study, different tests such as flakiness index, elongation index, aggregate impact value, sieve analysis, aggregate crushing value, and specific gravity tests were carried out on aggregate and filler materials to characterise them. Also, ductility test, viscosity test, penetration test, flash and fire point test, and softening point test were conducted on the bitumen to ascertain its alignment with the standard specifications. Furthermore, Marshall Stability, Marshall Flow, and Indirect Tensile Strength tests and their derivatives, such as rutting resistance and moisture susceptibility, were determined on the asphaltic sample specimens.

The outcome of the study showed that the materials used conform to the standards. The Marshall properties of the samples showed improvement as the percentage BPA addition increases, with the 50% replacement showing the highest value for stability and flow. The Marshall quotient ranges from 3.93 kN/mm to 5.06 kN/mm; the highest value occurred in the control mixture (0% BPA), indicating higher stiffness. The ITS values generally improved, reaching the highest value at about 30% BPA with approximately 15.5 kN/m² (dry) and 10.3 kN/m² (wet), and the highest moisture resistance was recorded to be 79% at 40% BPA.

Overall, the results indicate that incorporating Bean Pod Ash influences the volumetric and mechanical properties of asphaltic concrete mixtures. This establishes that Bean Pod Ash can be effectively utilised as a sustainable alternative filler material in asphalt pavement construction without compromising performance requirements.

The rising prevalence of malnutrition and non-communicable diseases (NCDs) in Sub-Saharan Africa necessitates exploring nutrient-dense, locally available food sources. Indigenous Nigerian oleaginous seeds represent an underutilised reservoir of bioactive lipids and functional compounds with significant nutraceutical potential. This review evaluates the nutritional composition, lipid profiles, and health-promoting properties of selected indigenous oilseeds, including Irvingia gabonensis (bush mango), Tetracarpidium conophorum (African walnut), Treculia africana (African breadfruit), Cucumeropsis mannii (egusi melon), and Cyperus esculentus (tiger nut). Available evidence indicates that these seeds are rich in unsaturated fatty acids - particularly oleic and linoleic acids - alongside appreciable levels of phytosterols, tocopherols (vitamin E), phenolic compounds, and dietary fibre.

The fatty acid composition of these seeds suggests cardioprotective, anti-inflammatory, and antioxidant properties, which are critical in the dietary management and prevention of cardiovascular diseases, type 2 diabetes, obesity, and certain cancers. Additionally, their micronutrient density and energy value position them as strategic resources for combating protein-energy malnutrition and lipid deficiencies in vulnerable populations. Beyond direct consumption, these oleaginous biomaterials demonstrate functional properties suitable for incorporation into fortified foods, plant-based dairy alternatives, therapeutic spreads, and nutraceutical formulations.

Despite their promising attributes, commercialisation is constrained by limited compositional standardisation, inadequate processing technologies, poor value chain development, and insufficient clinical validation of health claims. Harnessing their full potential requires multidisciplinary research integrating food science, lipid chemistry, nutrition, and sustainable agribusiness development. Promoting indigenous oilseeds aligns with global Sustainable Development Goals related to zero hunger, good health and well-being, responsible consumption, and climate resilience. Strategic investment in research, processing innovation, and policy support could transform these underutilised resources into globally competitive functional food ingredients while strengthening Nigeria's bioeconomy and food security systems.

Nigeria's iron and steel sector is crucial for industrial growth and economic diversification but is hindered by ineffective project management, weak leadership, and unsustainable practices. This research investigates how project management and leadership can drive sustainable industrial development in the sector.

Employing a mixed-methods approach, the study integrates primary data from surveys and interviews with secondary data from industry reports and academic literature to evaluate practices, assess leadership styles, and propose sustainable strategies. Expected outcomes include a framework for effective project execution and policy recommendations for stakeholders, aligning with Nigeria's industrial goals and global sustainability objectives, such as SDG 9 (Industry, Innovation, and Infrastructure).

Deoiling hydrocyclones are widely used in industry for the treatment of produced water, which is a mixture of oil and water, and is the primary by-product of oil and gas reservoir operation [1]. Produced water is fed tangentially into conventional deoiling hydrocyclones, while the vortex finder inside aids in separating dispersed oil droplets from water [2]. Efficient removal of oil droplets remains a major challenge faced by the industry, dictated by stringent regulatory requirements.

To better understand the complexities involved in separating oil droplets from water, Computational Fluid Dynamics (CFD) based investigations have been carried out in the present study. A Population Balance Model (PBM) has been used to numerically predict the spatio-temporal distribution of oil droplets within the hydrocyclone. A discretised bin-based approach has been employed along with a log-normal distribution of oil droplets, ranging from 5-80 µm in size. Droplet transport, coalescence and breakage phenomena are modelled across nine bins. The Reynolds Stress Model (RSM) is used to capture anisotropic turbulent flow structures.

Detailed hydrodynamic characterisation of produced water and the separation efficiency of the hydrocyclone are the main objectives of this work. It has been observed that as the produced water propagates from inlet towards the underflow, the droplet size distribution shifts from smaller sizes (10-40 µm) to larger sizes (40-70 µm) due to droplet coalescence. It has also been observed that droplet-droplet coalescence is dominant over droplet breakage due to the low turbulent energy dissipation rate within the hydrocyclone. For a median oil droplet size of 28 µm, the predicted separation efficiency was 74%, showing good agreement with Young's [3] experimental findings. Findings of this study enhance our understanding of key mechanisms governing oil droplet separation in produced water and offer a platform for optimisation of deoiling hydrocyclone design and operation.

This investigation studied the effect of groove size and orientation on binary bearing temperature under different operating conditions. Sixteen bearings were locally produced in a foundry with dimensions of 40 mm outer diameter and 10 mm length each, using a B/D ratio < 1 for safe condition. They were grouped based on groove sizes (0.0015, 0.002 and 0.0025 m) and orientation (axial, radial and inclined at angles of 20°, 40° and 60°). The bearings were fixed one after the other in a hub and a shaft inserted through them at each given time, with one end fixed to the jaw of a lathe, and were allowed to run for 5 minutes using 20 ml of oil lubricant which was applied at intervals for various speeds of 40, 85, 125 and 260 rev/min. After the period of five minutes elapsed, a thermocouple clamped at the base centre hole of the bearing read the temperature value.

Results showed that between speeds of 40 to 85 rev/min the sliding surfaces of the bearing and shaft are practically in direct contact and friction is at its highest level, known as boundary lubrication. At speeds above 85 and 125 rev/min, lower friction levels are achieved through the use of mixed lubrication, where the sliding surfaces are partially separated by the lubricant. At this point a thin film is beginning to form, thereby partially separating the shaft from the bearing. At speeds above 125 rev/min, the minimum of the friction coefficient is reached at the critical value of the duty parameter, which was the dividing line between the mixed and hydrodynamic lubrication zones.