Addis Ababa Science and Technology University

Accurate prediction of construction durations is crucial for effective project management, particularly in rapidly urbanizing areas such as Addis Ababa. However, there exists a notable research gap regarding the comparative analysis of advanced machine learning (ML) algorithms against traditional methods for this purpose. This study aims to develop and evaluate various advanced ML algorithms to predict construction completion times in Addis Ababa, with the goal of improving resource allocation and enhancing client satisfaction. Data were collected through surveys administered to multiple construction organizations within the city, which served as the foundation for training, validating, and comparing a range of ML models. The research utilized the caret package in R for model development and assessment, incorporating methodologies such as artificial neural networks (NN), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Classification and Regression Trees (CART). To evaluate variable importance, multivariate visualizations, including correlation and scatter plot matrices, were employed, while performance metrics such as mean absolute error (MAE), root mean square error (RMSE), and R-squared (R²) were utilized for model comparison. The findings indicated that the RF model achieved an RMSE of 74 days and an R² of 0.97, while the KNN model also demonstrated strong performance with an RMSE of 81 days and an R² of 0.97, marking them as the most accurate models for predicting construction durations. In contrast, the NN model exhibited subpar performance, likely due to constraints related to training data and variable selection. As a result, the RF model was further optimized to improve its predictive accuracy. The study concludes that while the RF model proves to be highly effective for predicting construction durations in Addis Ababa, there is a critical need to expand the dataset and incorporate additional variables to enhance the performance of deep learning and other ML algorithms in this field.

Fouling in heat transfer units has a negative economic impact. Many scale‐forming impurities are present in cane molasses generated by sugar cane production technology, including cations of aluminum (Al), calcium (Ca), iron (Fe), magnesium (Mg), potassium (K), sodium (Na), as well as anions of carbonates, sulfites, phosphates, sulfate, silicates, and chlorides. Ca cations, in particular, form insoluble complexes with many other chemical constituents, making them a scale‐forming impurity. The accumulation of Ca ion on the heat exchanger's surface could increase heat transfer resistance and reduce its overall efficiency. This article reviews many types of heat‐transmitting unit fouling and their successive fouling occurrences. Identifying the chemicals found in scale deposits helps to determine which cleaning products will effectively clean heat transfer units and which scale inhibitors will drastically lower scale formation rates. Furthermore, numerous unit operations and unit processes, such as molasses pre‐treatment and pre‐fermentation practice (inoculation of yeast), followed by fermentation practice, and product purification practice distillation, are used to limit deposit formation and boost ethanol production efficiency. Molasses pre‐treatment and treatment such as chemical treatment, heat treatment, acid centrifugation, and mechanical treatment are critical in decreasing scale development during heat exchange operations.

Sela Dingay kaolin deposit is a primary deposit formed mainly through the weathering of rhyolitic ignimbrite. Integrated analyses of geological, mineralogical, geochemical and physical properties were conducted to characterise the industrial applications of Sela Dingay kaolin. X-ray diffraction (XRD) and attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR) were used to characterise the raw kaolin samples. The results show that Sela Dingay kaolin is mainly composed of kaolinite (66–93%) and quartz (9–34%). The higher chemical index of alteration (CIA = 94.2%) and chemical index of weathering (CIW = 96.8%) values suggest that Sela Dingay kaolin was formed by intense chemical weathering under oxidising conditions. Sela Dingay kaolinite deposit consists of high percentage of clay/silt-sized particles, a low shrinkage value (1.1–3.4%), medium-high plastic limit (25–35%), with specific gravity and pH values ranging between 2.47 and 2.56 g/cm ³ and 6.3 and 7.2, respectively. The overall results show that Sela Dingay kaolin is medium to high grade and it can be used for various industrial applications with appropriate beneficiation processes to remove higher amounts of Fe 2 O 3 and quartz. Therefore, the newly discovered kaolin deposit (estimated at 5,136,166 tonnes) in the Sela Dingay area is a potential target for future kaolin exploitation.

In the present work, the vibration behavior of honeycomb cored multi‐walled carbon nanotube (MWCNT) and Ti 3 C 2 T x ‐MXene reinforced sandwich composite plate has been investigated. The elastic properties of two‐phase (Ti 3 C 2 Tx‐MXene‐MWCNT)/epoxy sandwich composite were evaluated by utilizing the Halpin‐Tsai method. Then, glass fiber was incorporated as reinforcement, and the elastic characteristics of the hybridized three phase composite were obtained using the Chamis analytical model. The vibration behavior of sandwich panels was investigated with the help of finite element formulation by obtaining the strain fields using high‐order shear deformation theory. The developed numerical model was experimentally validated, demonstrating its efficacy in predicting the natural frequencies of sandwich composite panels under varying conditions. Nano‐filler reinforcement consistently increased natural frequencies across all vibration modes, regardless of boundary conditions. A parametric study revealed that natural frequency monotonically increased with higher aspect ratio and weight fraction of MWCNTs and Ti 3 C 2 T x ‐MXene. However, the thickness ratio had a significantly greater effect on natural frequency than other parameters. Clamping conditions also affected vibration behavior, with natural frequencies following the order: CFFF < SSSS < SFSF < CFCF < CSCS < CCCC. Regarding the transverse response, the root mean square velocity decreased with increasing MXene/CNT concentration and aspect ratio, attributed to enhanced stiffness and load‐bearing capacity of the hybrid sandwich composite. This study offers valuable insights for effective utilization of different types of nanoparticles in conjunction and design and development of nanoparticle‐reinforced sandwich composites, aiding in the prediction of the vibration behavior of these nanostructures. Highlights Synergetic effect of MWCNT and MXene on the natural frequency are revealed Numerical and experimental techniques are applied to measure the natural frequency Effect of nano‐filler concentration, aspect ratio and boundary conditions are studied CNT exerted more pronounced impact on vibration behavior than the MXene RMA velocity decreased with increase in the concentration and aspect ratio of nano‐fillers

Pristine ZnO (Z), MnO2 (M), CuO (C) photocatalysts and polyvinyl alcohol (PVA)-assisted MnO2–ZnO–CuO (MZC) nanocomposites were synthesized via the sol–gel method. The synthesized samples were characterized using thermal analysis (TGA), X-ray diffraction (XRD), dynamic light scattering (DLS), scanning electron microscopy (SEM), energy dispersive X-ray (EDS), transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM) techniques. The thermal analysis results of the prepared nanomaterial confirmed that the suitable calcination temperature for the synthesis of these nanomaterials is 420 °C. In addition to the morphological and elemental composition, the characteristic diffraction peaks of the MZC nanomaterial were found to align with those of the pristine Z, M, and C photocatalysts. The photocatalytic activities of the synthesized nanomaterials for methylene blue (MB) degradation were evaluated under optimized conditions. The degradation efficiencies of Z, M, C, and MZC were found to be 45%, 57%, 66%, and 93%, respectively, for MB in 180 minutes. The MZC nanocomposite exhibited superior photocatalytic activity compared to the pristine materials, which is attributed to the synergetic effect of the Z, M, and C photocatalysts. The effects of pH, initial dye concentration, and catalyst load were also explored to determine the optimum conditions. The best photocatalytic efficiency was observed at pH 8, with a 130 mg L⁻¹ catalyst load, and a 10 mg L⁻¹ initial dye concentration. The efficiency of the MZC nanocomposite in real textile wastewater was also tested, achieving 80% degradation of pollutants within 180 minutes. Recycling experiments were conducted for four consecutive cycles under optimal conditions. The photodegradation efficiency for the first, second, third, and fourth cycles was 93%, 91%, 90%, and 89%, respectively, demonstrating high consistency in photodegradation performance across the four cycles. Moreover, a Z-scheme photocatalytic mechanism was proposed as a potential mechanism for the MZC photocatalytic system.

This research paper discusses a wind turbine system and its integration in remote locations using a hybrid power optimization approach and a hybrid storage system. Wind turbine systems’ optimization controllers operate MPPT strategies efficiently, optimizing the system’s overall performance. The proposed approach is HTb(P&O/FLC), which combines the P&O and FLC methods. This innovative hybrid MPPT strategy takes advantage of the two methods. It maximizes the wind power thus minimizing stress on the storage system. For storage, batteries are important in isolated renewable energy systems due the interminent renewable sources. Unfortunately, they have some drawbacks as lower energy density, and limited charge-discharge. Besides, supercapacitors (SCs) have advantages as quick charging and discharging and long cycle life and some drawbacks as lower energy density and specific voltage requirements. This paper proposes a hybrid storage system combining batteries and SCs. To manage the two storage technologies, an effective algorithm isproposed focusing on managing their storage systems. The findings indicate that the proposed strategy effectively maintains the state of charge (SOC) of both the batteries and SCs in the intended limits. Simulation under MATLAB and Real-time with RT-LAB were carried out to validate the proposed control approach for the wind turbine system.

Productivity in opencast mining, particularly in drill-blast (DB) and surface miner (SM) operations, is crucial for optimizing efficiency and reducing costs. These operations are directly affected by fragmentation, which in turn impacts equipment utilization, loading cycle times, and downstream operations. This study analyzed field data such as rock properties, machine parameters, blast design results, and post-blast fragmentation size (0.15–0.82 m), with 0.45 m identified as the optimal fragmentation size for a 12 m³ shovel bucket. Traditional productivity assessments often use simplistic models that fail to capture the complexities of mining operations. To address this, an explainable machine learning (ML) model was developed, integrating fragmentation size, rock and machine parameters, and geometric factors to evaluate DB and SM operations in opencast coal mines. Various ML techniques, such as artificial neural network (ANN), random forest regression (RFR), gradient boosting regressor (GBT), and support vector regression (SVR), were employed to analyze these parameters. Among these, the RFR model demonstrated the highest accuracy, with a coefficients of determination (R²) of 99.5% for training and 99.2% for testing in DB datasets, and 99.9% for training and 99.5% for testing in SM datasets. Furthermore, the RFR model had the lowest root mean square error, mean absolute error, and mean absolute percentage error of 10.35, 4.788, and 2.1% for DB training datasets, and 5.53, 1.75, and 1.5% for SM training datasets, respectively, underscoring its superior performance. Using SHAP (Shapley Additive exPlanations), the study identified key productivity drivers: SM cycle time, diesel consumption, and coal face length. Fragmentation size, resulting from blasting, was also found to influence shovel efficiency and overall productivity significantly. This paper highlights the effectiveness of ensemble ML models in predicting and analyzing complex productivity dynamics in opencast mining.

Cypermethrin (Cyp), a persistent synthetic pyrethroid insecticide widely used for insect control. The persistence of Cyp creates toxicity to both humans and the environment This study investigates biochar and Bacillus cereus distinct and collective effects on Cyp -contaminated soil during a 90-day incubation. This study also investigates the effects of different concentrations of Cyp (50, 100, ,500 to 1000 mg kg⁻¹) on soil physicochemical and biological activities during a 90-day incubation period. Microbial biomass carbon and soil respiration rates decreased significantly across all cypermethrin concentrations, with the most substantial reductions observed at 1000 mg kg⁻¹. However noticeable variations in soil enzymes and MBC over time during the entire incubation period. On 1st day, the GMean Enz and MBC rate for Cyp treatments (50, 100, ,500 to 1000 mg kg⁻¹) ranged from 0.98 to 0.63, and 9.06, to 5.03, respectively. Under Cyp pollution, microbial biomass carbon exhibited significant decreases, with the highest inhibition (86.2%) at 1000 mg kg⁻¹ on 1st day of incubation. Soil respiration rates dropped 77%, at 1000 mg kg⁻¹, and Integrated biomarker response (IBR) values peaked on day 30, indicating environmental stress. Biochar and Bacillus cereus effectively facilitated the degradation of Cyp, achieving approximately 85% degradation within the first 45 days of the experiment. The combined application of biochar and Bacillus cereus increased soil pH to a neutral level from 5.9, to 7.1, reduced electrical conductivity from 1.41 µS cm− 1 to 1.20 µS cm⁻¹, and elevated cation exchange capacity from 1.54 ± 0.04 to 6.18 C mol kg⁻¹, while also improving organic carbon content to 3.135%. However, the dehydrogenase activity was decresed upto 47% in the combined application and all other enzymes including urasese catlayse and phostasese enzymes with Gmean enzymeatic activities were significantly improved. These findings suggest biochar and bacterial interaction for soil management to enhance soil resilience against pesticide stress.

HSV encephalitis is a common cause of sporadic fatal encephalitis, affecting persons of all ages. It is also a medical emergency because the prognosis is primarily determined by early treatment initiation, necessitating a rapid and accurate diagnosis. CNS vasculitis complicating HSV encephalitis is a rare phenomenon. We present a case of a 28‐year‐old right‐handed male patient diagnosed with HSV meningoencephalitis complicated with vasculitis, the challneges in the diagnosis and treatment response.

The increasing demand for cement has substantially affected the environment, and its manufacturing requires substantial energy usage. However, most countries in the world recently encountered a significant energy problem. So, researchers are exploring the use of agricultural and industrial waste resources with cementitious characteristics to minimize cement manufacturing, cut energy consumption, and contribute to environmental protection. Therefore, this research is performed on roller compacted concrete (RCC) reinforced with 5%, 10%, 15%, and 20% of corn cob ash (CCA) as substitution material with different percentage of cement and 0.25%, 0.50%, 0.75%, and 1% of jute fibre (JF) together for determining the mechanical properties and embodied carbon (EC) by applying response surface methodology (RSM) modelling. The cubical samples were prepared to achieve the targeted strength about 30 MPa at 28 days and then obtained mix proportions were employed for all combinations at various water-cement ratios to maintain roller-compacted concrete’s zero slump. Results showed that at 0.50% JF and 10% CCA, the flexural strength, splitting tensile strength and compressive strengths, and modulus of elasticity of RCC obtained were 5.3 MPa, 3.8 MPa, 32.88 MPa, and 33.11 GPa at 28 days, respectively. Besides, the embodied carbon of RCC is recoded reducing with combined addition of different levels of JF and CCA as compared to control mixture. In addition, the generation of response prediction algorithms was performed using analysis of variance (ANOVA) at a threshold of significance of 95%. The coefficient of determination (R²) readings for the statistical models ranged from 96 to 99%. It is observed that the use of 0.50% of JF along with 10% of CCA as cementitious constituent in RCC provides best outcomes. Therefore, this method is a superior choice for the construction industry.

Objective Maternal obesity and gestational diabetes mellitus (GDM) are becoming major public health concerns in developing countries. Understanding their relationship can help in developing contextually appropriate and targeted prevention strategies and interventions to improve maternal and infant health outcomes. This study aimed to determine the association of maternal overweight and obesity with GDM among pregnant women in Ethiopia. Design Case-control study. Setting The study was conducted in selected public hospitals in Addis Ababa, Ethiopia, from 10 March to 30 July 2020. Participants 159 pregnant women with GDM (cases) and 477 pregnant women without GDM (controls). Outcome measures and data analysis Screening and diagnosis of GDM in pregnant women was done by a physician using the 2013 WHO criteria of 1-hour plasma glucose level of 10.0 mmol/L (180 mg/dL) or 2-hour plasma glucose level of 8.5–11.0 mmol/L (153–199 mg/dL) following a 75 g oral glucose load. Overweight and obesity were measured using mid-upper arm circumference (MUAC). Binary logistic regression with bivariate and multivariable models was done to measure the association of overweight and obesity with GDM. Adjusted ORs (AORs) with a 95% CI were computed, and statistical significance was determined at a value of p=0.05. Results GDM was associated with obesity (MUAC≥31) (AOR 2.80; 95% CI 1.58 to 4.90), previous history of caesarean section (AOR 1.91; 95% CI 1.14 to 3.21) and inadequate Minimum Dietary Diversification Score <5 (AOR 3.55; 95% CI 2.15 to 5.86). The AOR for overweight (MUAC≥28 and MUAC<31) was 1.51 (95% CI 0.71 to 3.21). The odds of developing GDM were 72% lower in pregnant women who were engaging in high-level physical activity (AOR 0.28; 95% CI 0.12 to 0.67). Conclusion Obesity, but not overweight, was significantly associated with the development of GDM. Screening for GDM is recommended for pregnant women with obesity (MUAC≥31) for targeted intervention. Antenatal care providers should provide information for women of childbearing age on maintaining a healthy body weight before and in-between pregnancies and the need for healthy, diversified food and high-level physical activity.

A subset of machine learning algorithm called Deep Reinforcement Learning (DRL) enables computers or agents to learn behavior by taking actions in a given environment through trial and error while observing the rewards. In this learning paradigm, the agent is given a set of actions to chose and is then rewarded or punished depending on the results of those actions. The agent gradually develops the ability to make the best decisions by maximizing its rewards. DRL blends the learning ability of deep neural networks into the decision making capability of reinforcement learning (RL) frameworks in order to seeks and identify the most favorable set of actions. This survey paper studies DRL applications for diverse image processing tasks. It starts by providing an overview of the latest model-free and model-based RL and DRL algorithms. Then, it looks at how DRL is being used for various image processing tasks including image segmentation and classification, object detection, image registration, image denoising, image restoration, and landmark detection. Lastly, the paper discusses the potential uses and challenges of DRL in the proposed area by addressing the research questions. Survey results have showed that DRL is a promising approach for image processing and that it has the potential to solve complex image processing tasks.

To resolve the problems such as fatigue or pitting corrosion on a low-cost evaporator tubing in a household refrigerator, the tube material was changed from copper (Cu) to Aluminum (1070 Al) by applying para-metric accelerated life testing (ALT). This systematic strategy included the following: (1) ALT plan used on BX lifetime that will be X percent of the accumulated failure, (2) corrosion effect modeling, (3) ALTs with revision, and (4) estimation whether the design attained the objective BX lifetime. A quantum-transported life-stress model and a sample size producing reliability quantitative (RQ) statements were proposed. A case study investigation for this methodology was employed to enhance the lifetime of a household refrigerator whose evaporator tubing was failing in the field due to aqueous pitting or crevice corrosion. In the first ALT, for RQ statements, this failure was reproduced by ALT equipment and the refrigerator tubing that were placed in a 3.5% saline solution for the accelerated condition. The pitted tubing was determined to be nearly identical to those returned from the marketplace refrigerators. Using a scanning electron microscope (SEM) with Energy Dispersive X-ray (EDX) showed that the trouble with the evaporator tubing came from water condensation and direct chlorine contact on polyvinyl chloride (PVC) tape with water-soluble adhesives layer. Three action plans were evaluated. First, the tape material was changed from PVC to polyethylene (PE) with a water-soluble adhesive. Second, the contraction tube was enlarged from 50 to 200 mm. Third, polyethylene foam pads between enclosure case and the tubing were inserted. In the 2nd ALT, the refrigerator tubing showed no corrosion. Refrigerator thus achieved the goaled lifetime-B1 life of ten years.

Topological indices are derived from the topology of the graph representation of the molecule and can be used to estimate various molecular properties. They are widely used in quantitative structure–property relationship (QSPR) analysis and computational chemistry studies. Flavonoids are natural compounds in sweet potatoes having various health benefts, including antioxidant and anti-infammatory properties. By understanding the factors infuencing the variation in favonoid content, researchers can develop strategies to enhance the nutritional value of sweet Potato. Degree-based topological indices play a vital role in the QSPR analysis of molecular structures by providing molecular descriptors that help understand and predict the properties of these natural compounds. In this study, we initially computed several degree-based topological indices of fve naturally occurring favonoids (Myricetin, Quercetin, Luteolin, kaempferol, and Apigenin). Then, a QSPR analysis is performed by developing regression models. The results show that the considered topological indices have strong correlations with certain experimental physicochemical properties of favonoids, which indicates their strong predictive ability. This confrms that the topological indices are the best alternatives to the laborious, time-consuming, and costly laboratory experiments.

Understanding the microbial ecology of landfills is crucial for improving waste management strategies and utilizing the potential of these microbial communities for biotechnological applications. This study aimed to conduct a comprehensive taxonomic and functional profiling of the microbial community present in the Addis Ababa municipal solid waste dumpsite using a shotgun metagenomics sequencing approach. The taxonomic analysis of the sample revealed the significant presence of bacteria, with the Actinomycetota (56%), Pseudomonadota (23%), Bacillota (3%), and Chloroflexota (3%) phyla being particularly abundant. The most abundant KEGG categories were carbohydrates metabolism, membrane transport, signal transduction, and amino acid metabolism. The biodegradation and metabolism of xenobiotics, as well as terpenoids and polyketides, were also prevalent. Moreover, the Comprehensive Antibiotic Resistance Database (CARD) identified 52 antibiotic resistance gene (ARG) subtypes belonging to 14 different drug classes, with the highest abundances observed for glycopeptide, phosphonic acid, and multidrug resistance genes. Actinomycetota was the dominant phylum harboring ARGs, followed by Pseudomonadota and Chloroflexota. This study offers valuable insights into the taxonomic and functional diversity of the microbial community in the Addis Ababa municipal solid waste dumpsite. It sheds light on the widespread presence of metabolically versatile microbes, antibiotic resistance genes, mobile genetic elements, and pathogenic bacteria. This understanding can contribute to the creation of efficient waste management strategies and the investigation of possible biotechnological uses for these microbial communities.

Electric vehicles (EVs) have become a crucial option for reducing greenhouse gas emissions by decreasing the reliance on fossil fuels and promoting the use of renewable energy. However, many research efforts have overlooked the detailed analysis of EV characteristics and charging networks, leading to scalability issues, inefficiencies in the charging process, and increased energy consumption. To address these challenges, this study introduces the Fuzzy Discrete Squirrel Search (FDSS) algorithm, which is designed to enhance charging capacity and reduce pollution. While EV manufacturing and network simulation models continue to evolve, there is a need for further optimization. This study summarizes key research findings related to EVs, including their market penetration, design methodologies, and innovative technologies. The primary objective of this work is to uncover the hidden benefits of EV charging characteristics using the Discrete Squirrel Search (DSS) optimization algorithm in combination with a fuzzy logic controller. The FDSS algorithm has been selected to improve performance and effectively identify the latent advantages of EV charging. In developing countries, this research is particularly significant, as it addresses critical obstacles such as inadequate charging infrastructure. Additionally, the concept of “vehicle-to-grid” provides a backup power source when renewable energy is unavailable. Our conclusion emphasizes the importance of recognizing the unique characteristics of EVs to enhance their mobility and overall effectiveness.