LAUTECH Journal of Engineering and Technology

Development of Brain Tumor Classification System using Convolutional Neural Network Model with Explainable Artificial Intelligence

2026-04-13T21:42:41+00:00

Brain tumors are abnormal cell growths in the brain that require accurate and timely diagnosis, where Magnetic Resonance Imaging (MRI) plays a critical role in detecting and characterizing tumor structures. However, accurate interpretation of Magnetic Resonance Imaging (MRI) scans is challenging due to their complexity and the limited availability of expert radiologists. This challenge is further compounded by the lack of interpretability in many existing deep learning-based diagnostic systems. Therefore, the need for an automated and interpretable brain tumor classification system arises, which is the problem this study aims to address. In this research, a brain tumor classification system integrated with Explainable Artificial Intelligence (XAI) was developed using MRI images. The model was designed to classify brain tumors into glioma, meningioma, pituitary tumor, and no-tumor categories while providing visual explanations for its predictions using Grad-Class Activation Mapping (Grad-CAM). The performance of the system was evaluated for each tumor category using accuracy, precision, specificity, recall, F1-Score, false positive rate and also the Receiver Operating Characteristic and Area Under Curve (ROC-AUC). Experimental results show that the developed CNN model achieved an overall classification accuracy of 90.6% with an AUC-ROC value of 0.9892, indicating strong discriminative capability across tumor classes. The Grad-CAM visualizations consistently highlighted tumor-affected regions in the MRI images, confirming that the model based its predictions on clinically relevant anatomical structures. The developed model demonstrated effective classification performance and improved interpretability, making it suitable as a reliable decision-support tool for automated brain tumor diagnosis.

Effect of the Heat-Affected Zone on the Mechanical Properties and Microstructure of TIG-Welded Cast Iron/Aluminum Alloy Joints

2026-04-09T15:33:38+00:00

Dissimilar-metal welding between Gray Cast Iron (ASTM A48M) and Aluminum Alloy 6063 presents a persistent engineering challenge, particularly in the heat-affected zone (HAZ) where rapid thermal cycling drives microstructural changes that govern the mechanical performance of the entire joint. To address this, TIG-welded specimens (200 mm × 200 mm × 30 mm) were fabricated at 240 V and 90 A and subjected to three post-weld cooling conditions natural air cooling, oil quenching, and water quenching to systematically examine how cooling strategy shapes both microstructure and mechanical behaviour. Vickers hardness, Charpy impact, and tensile tests were carried out across the weldment, HAZ, and base metal regions, complemented by optical metallography and scanning electron microscopy (SEM) to characterise the underlying microstructural changes. The findings show clearly that oil quenching drives weldment hardness to its peak in both cast iron (310.7 HV) and aluminum alloy (146.8 HV), while water quenching produces the highest HAZ hardness in cast iron (292.0 HV). Where toughness and tensile strength are concerned, however, natural cooling consistently outperforms both quenching methods across all zones and both materials. At the microstructural level, cooling rate was found to directly govern graphite morphology in cast iron and precipitate distribution in the aluminum alloy two mechanisms that together explain the full range of mechanical property gradients observed. Taken together, these results offer concrete, quantitative guidance for selecting post-weld cooling strategies in dissimilar-metal TIG welding, with direct relevance to automotive and structural engineering applications.

Estimation of Municipal Solid Waste Generation Trends From Selected Dumpsite in Ibadan from 2018-2024

2026-05-08T13:05:33+00:00

Municipal solid waste (MSW) generation encompasses residential, commercial, institutional, construction, demolition, and sanitation waste streams. This study quantified, characterized, and estimated waste generated at four selected dumpsites in Ibadan, namely Awotan, Lapite, Ajakanga, and Aba Eku, between 2018 and 2024. Annual waste quantities were estimated by projecting Ibadan's population with an exponential growth model and applying a per capita generation rate of 0.55 kg per person per day, after which a 70 percent collection efficiency was used to derive the collected fraction. The collected waste was then distributed across the four sites according to the operative policy allocation ratios for each period, while composition was characterized by applying percentage fractions for each waste category drawn from field and literature sources, and all estimates were validated against Oyo State Waste Management Authority records. On this basis, total waste generated and deposited across all sites was estimated at 10.8 million tonnes, with an average annual input of 1.54 million tonnes. Awotan received the highest annual average (420,000 tonnes), followed by Lapite (380,000 tonnes), Ajakanga (310,000 tonnes), and Aba Eku (210,000 tonnes). Organic matter dominated the waste composition (59%), followed by plastics (10%), paper (9%), textiles (6%), metals (5%), glass (4%), rubber (3%), and others (4%). The 2020 Oyo State waste management policy shift significantly altered the distribution of waste across sites. The Oyo State Government is recommended to institute mandatory waste segregation into organic, recyclable, and inert categories to reduce CO and PM?? emissions from open burning.

Facial Expression-Based Customer Sentiment Analysis for Service Quality Improvement using Deep Learning Techniques

2026-05-12T14:29:38+00:00

In today’s customer-centric economy, understanding and responding to customer sentiment is vital for service excellence. Traditional feedback mechanisms, such as surveys and reviews, are often limited by response bias and delayed insights. Hence, this research presents a deep learning approach for improving service delivery through customer sentiment and facial expression analysis. The study leverages specifically MobileNetV2 and InceptionV3, to classify facial expressions into three sentiment classes: Satisfied, not satisfied and Neutral. A local dataset comprising of 900 annotated facial images of locally sourced dataset was curated to address ethnic bias in existing datasets to ensure local relevance while Facial Expression recognition 2013 dataset comprises 48x48 pixel grayscale photos of faces.The methodology involved rigorous data preprocessing, including grayscale normalization, face alignment, and augmentation. MobileNetV2 and InceptionV3 models were trained and evaluated using stratified 80:20 train-test split with categorical cross-entropy as the loss function. Performance was assessed using metrics such as accuracy, precision, recall, F1-score, and confusion matrix. Inception reported an accuracy of 94.01% precision of 0.89 recall of 0.98 and an f1 score of 0.94. MobileNet on the other hand reported an accuracy of 90.12%, precision of 0.89, recall of 0.91 and f1-score of 0.91. This shows that inception model outperformed mobileNet in terms of accuracy, precision, recall and f1-score. The results demonstrate the feasibility of using facial expression recognition for sentiment tracking in service environments such as banks, schools, and retail outlets.

Optimisation of Performance Parameters of Farm Machinery for Economic Sustainability in Irrigation Scheme (A Case Study of Giriyan Irrigation Scheme, North Central Nigeria).

2026-06-02T14:32:17+00:00

The efficient utilization of farm machinery is key to strategic management for sustainable agricultural mechanization and economic sustainability. This study optimized agricultural input-output performance parameters of farm machinery for economic sustainability at Giriyan irrigation scheme, North Central Nigeria. A survey of structured questionnaires was adopted to obtain level of utilization of farm machinery in the scheme. Field data were obtained through soil analysis and real-time measurements of tractor performance using sensor-based instrumentation. An Artificial Neural Network (ANN) model was developed with nine input variables, two hidden layers (8–5nodes), and five output variables to evaluate utilization index, mechanization index, capacity utilization index, specific fuel consumption, and profitability index. The best topology structure for the ANN model developed was 9-8-5-5. The model achieved high prediction accuracy with coefficients of determination (R²) values of 0.9902, 0.9943, 0.9819, 0.9823 and 0.9979 for utilization index, mechanization index, capacity utilization index, specific fuel consumption, and profitability index respectively. The results revealed that operational cost, rated implement width, tractor power and tillage depth were the most influential parameters for optimizing specific fuel consumption and good profitability of farm machinery investment at Giriyan irrigation scheme of North Central Nigeria. The ANN model demonstrated strong predictive capacity, with training and testing errors of Mean Square Error (MSE) of 0.0528 and 0.0349 respectively. The ANN model provides a good decision-making support tool for farm managers and policy makers to improve farm machinery performance, optimize energy use, and promote sustainable agricultural mechanization in many irrigation schemes of North Central Nigeria.

Hybrid Polymer-Thermal Flooding for Heavy Oil Production: A Critical Review of Experimental and Numerical Advancements

2026-05-06T21:26:38+00:00

The world energy demands are ever increasing as conventional oil deposits continue to depleted and the interest in novel Enhanced Oil Recovery (EOR) procedures. This review considers the hybrid polymer-thermal EOR that aims to combine the advantages of thermal EOR to use a lower viscosity with the property of increasing the power of sweeps in polymer flooding. The traditional polymers such as partially hydrolyzed polyacryl amide (HPAM) are vulnerable to degradation by high temperature and high salinity conditions of reservoirs. New developments such as the use of nanocomposite systems on natural biopolymers (xanthan gum, guar gum, gum arabic) enhanced with silica, alumina or magnesium nanoparticles are also promising. Laboratory experiments (viscometry, FTIR, SEM, TGA, core flooding, and ECLIPSE simulations) show that under simulated reservoir conditions, the viscosity of the oil is better retained and leads to higher oil recovery. Gum arabic nanocomposites demonstrate a certain potential even with a reduced base viscosity. It is a synthesis of existing knowledge on polymer-thermal EOR mechanisms, identifies gaps in the current knowledge, and suggests future research directions including material innovation and development of tools to simulation in order to achieve sustainable heavy oil recovery.

The Valorization Potential of Kolanut Pod for Biogas Production: A Physicochemical and Biochemical Characterisation

2026-06-16T11:44:14+00:00

This study evaluated kolanut pod as a potential biomass for biogas production in an anaerobic digestion process. Fresh kolanut pod underwent mechanical pre-treatment process of washing, sun-drying and pulverised with a Janke and Kunkel hammer mill for particle size reduction. The Dry Matter (DM), Organic Dry Matter (ODM), Organic Carbon (OC) and Nitrogen (ON) and calorific value of the residues were determined by APHA standard method. Its morphological structure and elemental composition were determined using the methods of Scanning Electron Microscopy (SEM)/Energy Dispersive X-ray (EDX), The functional group and microbial characterisation were determined using Fourier transform infrared spectrometry (FTIR) and Bergy's Manual of Bacteriology, respectively.

The proximate analysis showed that the dry matter and organic dry matter of kolanut pod are 91% and 83%, respectively while ultimate analysis revealed that its organic carbon, nitrogen and calorific value are 14.25%, 0.46% and 3.6512 kJ/kg, respectively. The kolanut pod showed irregular and heterogeneous surface morphology with well-developed porous structures that appear as little cavities. The kolanut pod also revealed a particle size of 20.36 ?m, height of 2.41 mm with bulk density of 2.12 g/cm³ while the EDX showed presence of essential macro-elements such as Silicon, Si (22.15%), Calcium, Ca (13.96%), Potassium, K (3.29%), Aluminum, Al (1.98%), Magnesium, Mg (2.35%) and Sodium, Na (0.28%). FTIR revealed a characteristic peaks indicative of cellulose, lignin, and phenolic compounds The anaerobic bacterial present in the kolanut pod are Bacillus cereus, Escherichia coli and Proteus mirabilis.

Design and Implementation of an Automated Power Consumption Monitoring System for Multi-Tenant-Based Billing in Residential Buildings

2026-06-26T20:45:57+00:00

This paper presents the design and implementation of an automated power consumption monitoring system tailored for multi-tenant-based billing in residential buildings. Traditional manual metering methods often lead to inaccurate billing, disputes, and reduced incentives for energy conservation. To address these challenges a prototype was built using a transformer-less 5V supply, 50A CT, HLW8032 metering IC, Arduino Nano, IR receiver, and TM1637 7-segment display; firmware (C++, 500Hz sampling) computes RMS/energy, logs to Electrically Erasable Programmable Read-Only Memory (EEPROM) and handles secure resets. Calibration and tests showed current error ? ±0.7%, voltage error ? ±0.4% (power error <1%), 5V regulation within ±0.3% with <100 mVpp ripple, and a 48-hour field trial recorded kWh deviations between ? 0.54% and + 0.48% (? ±0.6%). EEPROM persisted through power interrupts, IR resets decoded ?98% of the time, firmware ran 72hours without issues, and display refresh averaged ~37 ms. The prototype meets the project objectives which is accurate, repeatable tenant-level monitoring and automated billing with low measurement error and robust operation, demonstrating suitability for residential deployment. Recommended next steps are scaling the architecture for larger complexes, adding networked connectivity and secure cloud APIs, improving the user interface (mobile and web dashboard), and incorporating advanced analytics for consumption forecasting and automated energy-saving recommendations

Predictive Modelling of the Droplet Size Distribution in Swirl Nozzles under Varying Operating Parameters

2026-05-17T15:18:37+00:00

Droplet size distribution is a critical parameter in liquid atomization because it governs spray coverage, evaporation rate, mixing efficiency, and overall process performance. Predicting droplet-size characteristics remains challenging due to the complex interactions among centrifugal forces, liquid-sheet instability, and operating conditions in swirl nozzles. This study presents a predictive modelling approach for estimating droplet size distribution under varying operating parameters. The significance of the model is directly linked to the combined effects of the selected operating parameters, spray pressure, exit orifice diameter, swirl chamber length-to-diameter ratio, and nozzle spraying height, which were systematically varied to evaluate their influence on droplet size distribution. The model evaluation yielded a p-value of 0.00992 and F-value of 22.85 at 95% confidence level, indicating that variations in these parameters collectively have a statistically significant effect on spray characteristics. The coefficient of determination (R²) of 0.4244 shows that the model explains 42.44% of the variation in droplet size distribution, while the coefficient of variation (CV) of 2.90 (less than 10) confirms good reproducibility. Results indicate that increased operating pressure produces finer droplets, whereas nozzle geometry significantly influences distribution uniformity. The predictive model provides a useful tool for spray system design and optimization, reducing reliance on extensive experimental testing and enhancing spray efficiency in practical applications.

Optimization of Process Variables on Yield of Biosurfactant Derived from a Mutant Acinetobacter Sp.

2026-06-22T14:31:19+00:00

Abstract

Microbial Enhanced Oil Recovery (MEOR) involves the utilization of microorganisms to improve oil recovery from hydrocarbon reservoirs. The success of MEOR operation is hinge on the selection and utilization of potent microbial organisms capable of producing high-performance biosurfactants under extreme oil reservoir conditions. This study therefore, developed and optimize of a hyperactive mutant strain of a novel Acinetobacter species isolated from reservoir formation water. To enhance its temperature, pH and salinity tolerance along with metabolic yield, the wild-type strain was subjected to Atmospheric and Room Temperature Plasma (ARTP) using a helium plasma jet at a radio-frequency input power of 120 W. Exposure for 30 seconds resulted in a cellular lethality rate of 90.08% and yielded a positive mutation rate of 60%. Thermal stability evaluations across a temperature range of 45–95 °C over 10 days were assessed based on maximum optical density (OD) 550nm, emulsification indices (E₂₄, E₇₂), oil displacement test (ODT), surface tension (ST), and interfacial tension (IFT) dynamics. The selected hyper-producing isolate was subsequently subjected to multi-objective numerical optimization to evaluate its yield limits across varying pH (7.20–10.52) and salinity (15–35%) levels under extreme thermal stress (95 °C). The Biosurfactant Yield (BY) of 1.858 ml, ODT of 7.985 mm, ST of 49.985 dyne/cm, IFT of 41.362 dyne/cm, and emulsification indices of 20.533% E₂₄, 11.955% E₄₈, and 6.568% E₇₂were obtained. The ARTP mutagenesis provides a highly efficient strategy for generating robust, stress-tolerant bio-agents for tertiary oil recovery applications.

A Theoretical Approach to Designing a Decision Support System for Effective Emergency Response in Oil Spill Logistics

2026-05-07T21:29:55+00:00

The study is based on the design of a comprehensive Decision Support System (DSS) to effectively respond to emergencies in the offshore Niger Delta region, using the case study of the December 20, 2011, Bonga Oil Spill incident. Although there has been a positive change in the preparedness of oil spill responses, the current response logistics is still problematic, which affects efficiency and environmental impact. The study provides an answer to the critical gaps in the current decision support systems by proposing a multi-faceted DSS framework, which integrates the advanced modeling techniques: the Blowout and Spill Occurrence Model (BLOSOM) and the Oil Spill Cleanup Operations Model (OSCOM). Some of the objectives are to simulate the trajectory of oil spills under different environmental conditions, predictive modeling of resource allocation, and the use of visualization tools to assess the effectiveness of response. By using Nigeria-specific information, the research will help improve rapid decision-making, lessen the environmental effects, and reduce community conflicts. This powerful DSS framework can make a considerable enhancement on the oil spill response plans in Nigeria.

Effect of Wood Ash and Sawdust Ash Admixtures on the Engineering Properties of Burnt Laterite-Clay Brick from Igboora

2026-06-16T12:05:17+00:00

Abstract: An attempt to reduce the amount of environmental wastes and the high cost of conventional stabilizers has led to continuous studies on the economic utilization of ash from agro-wastes for improving the engineering properties of burnt brick. This paper reviewed the effect of addition of sawdust ash and wood ash admixtures to a 70:30 parts by weight laterite-clay mix. The admixtures were added in various combinations of proportions by weight (from 0 to 10%). The natural laterite-clay used was classified according AASHTO and USCS as A-7-5 and MH-OH respectively, and the there was general increase in compressive strength of the brick with increase in admixture and sample D (SDA:7.5% and WA:2.5%) has highest value of 11.45 MN m-2. The sample D with samples B and E were categorized as first class bricks with compressive strength higher than 10.3 MN/m2, therefore, adequate for structural building bricks in Grade NW condition. While samples A and C are also categorized as second class brick with compressive strength higher than 7.0 MN/m2., therefore adequate for general construction bricks work in Grade NW condition. The control sample X is categorized as third class brick with compressive strength between 3.5-7.0 MN/m2 (ASTM C62 99, 2000). The 24 hour water absorption capacity generally increased from 15.65% for control sample x to highest value 24.35% for sample D, this was due to change in internal structure during firing.Key words: Burnt brick, engineering properties, laterite-clay

Modelling and Optimization of Corncob-Reinforced Polymer Composite using Response Surface Methodology

2026-06-15T14:11:42+00:00

This study focuses on the development and optimization of corncob-reinforced polymer composites as a sustainable substitute for conventional synthetic materials. Growing environmental concerns and the need to utilize agricultural residues inspired the use of corncob, a lignocellulosic by-product, as a reinforcing filler in an unsaturated polyester matrix. The corncob was processed and characterized for its chemical composition before being incorporated into the polymer at varying fibre loadings and lengths. Response Surface Methodology (RSM) based on a Box–Behnken Design (BBD) was applied to analyze and optimize the influence of fibre loading and fibre length on key mechanical properties, including tensile strength, flexural strength, and impact energy. The results indicated that both variables significantly affected composite performance, with notable interaction and quadratic effects. Analysis of variance (ANOVA) confirmed the validity of the developed quadratic models, with high coefficients of determination (R² > 0.96), demonstrating strong agreement between predicted and experimental results. Optimization revealed that the best mechanical properties—tensile strength (11.74 MPa), flexural strength (20.90 MPa), and impact energy (0.815 J)—were obtained at a fibre loading of 23.54% and fibre length of 1.04 mm. BET analysis showed a high surface area of 266.971 m²/g with a strong correlation coefficient (r = 0.99297), indicating considerable porosity. SEM results confirmed uniform dispersion with localized pores, while EDX analysis revealed dominant silicon content alongside carbon and oxygen. FTIR spectra identified key functional groups, confirming the presence of silicate, carbonyl, alkyl, and hydroxyl functionalities. Overall, the results highlight the potential of corncob as an effective reinforcement material, promoting sustainable material development and efficient agricultural waste utilization.

Proneness of Building Materials to Waste in Oyo State, Nigeria

2026-06-28T14:28:47+00:00

Construction material waste poses a significant challenge to project cost, environmental sustainability, and efficient resource utilisation in the construction industry, particularly in rapidly developing regions such as Oyo State, Nigeria. This study assessed the proneness of building materials to waste and examined the factors influencing waste in building materials in the study area. These were with a view to enhancing construction project delivery. The study population for this research were Quantity Surveying and Construction firms in Oyo State with sampling frame of 18 and 734, respectively. A Census of Quantity Surveying firms was conducted while a sample size of 75 Construction firms was selected for this study using simple random sampling technique, making a total of 93. Primary data were collected from 16 Quantity Surveying firms and 61 Construction firms via structured questionnaires, making a total of 77 and a response rate of 82.8%. Out of the 77 questionnaires retrieved, 70 were properly filled and used for analysis. Findings from the analysis showed that although all the seven materials investigated were moderately prone to waste, blocks, timber, and cement were the top three materials that are prone to waste mostly due to factors such as handling, design variation, competence, planning, weather.