Browsing by Author "Meddage, D. P. P"

Now showing 1 - 6 of 6

Open Access
Explainable Machine Learning (XML) to predict external wind pressure of a low-rise building in urban-like settings
(2022-07) Meddage, D. P. P; Ekanayake, I; Weerasuriya, A; Lewangamage, C. S; Ramanayaka, C. D. E; Miyanawala, T
This study used explainable machine learning (XML), a new branch of Machine Learning (ML), to elucidate how ML models make predictions. Three tree-based regression models, Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boost (XGB), were used to predict the normalized mean (Cp,mean), fluctuating (Cp,rms), minimum (Cp,min), and maximum (Cp,max) external wind pressure coefficients of a low-rise building with fixed dimensions in urban-like settings for several wind incidence angles. Two types of XML were used — first, an intrinsic explainable method, which relies on the DT structure to explain the inner workings of the model, and second, SHAP (SHapley Additive exPlanations), a post-hoc explanation technique used particularly for the structurally complex XGB. The intrinsic explainable method proved incapable of explaining the deep tree structure of the DT, but SHAP provided valuable insights by revealing various degrees of positive and negative contributions of certain geometric parameters, the wind incidence angle, and the density of buildings that surround a low-rise building. SHAP also illustrated the relationships between the above factors and wind pressure, and its explanations were in line with what is generally accepted in wind engineering, thus confirming the causality of the ML model’s predictions.
Open Access
Influence of Crumb Rubber and Coconut Coir on Strength and Durability Characteristics of Interlocking Paving Blocks
(MDPI, 2022-07-13) Gamage, S; Palitha, S; Meddage, D. P. P; Mendis, S; Azamathulla, H. M; Rathnayake, U
Interlocking Paving Blocks (IPB) are, nowadays, a widely used construction material. As a result of the surge in demand for IPBs, alternative materials have been investigated to be used for IPBs. This study investigated the strength and durability characteristics (compressive strength, split tensile strength, density, water absorption, skid resistance, and abrasion resistance) of IPBs in the presence of (waste materials) crumb rubber (CR) and coconut coir fibers (CCF). Both compressive and split tensile strength increased in the presence of CCF to a certain extent. CR-based IPBs showcased an increase in skid resistance that satisfied both SLS 1425 and BS EN 1338 specifications. Abrasion depths of CR-based and CCF-based samples show a comparable increase in values when the respective fraction (CR or CCF) increases. Therefore, this research fills the knowledge gap, highlighting the importance of incorporating waste materials (CR and CCF) for the IPB industry rather than open dumping.
Open Access
Investigating applicability of sawdust and retro-reflective materials as external wall insulation under tropical climatic conditions
(Springer link, 2022-04-19) Dharmasena, P; Meddage, D. P. P; Mendis, S
Buildings require energy to maintain their performance. In consequence, built environments cause a surge in the world’s energy demand. Providing passive measures is an efective method of optimizing operational energy usage. In this study, we propose insulation materials (thermal barrier type and resistive insulation) for the walls of a building. Experiments were performed on small-scale physical models constructed with; (a) no insulation, (b) sawdust–cement mortar, and (c) retrorefective (RR) material for external walls. In addition, regression models were developed to predict indoor air temperature with insulation. Subsequently, associated operational energy-saving and decrease in emissions were estimated for each material. The comparison reveals RR (sawdust–cement mortar) is efective in warm (overcast) climatic conditions. Developed regression models have shown a good agreement with experimental results (R>0.8). Moreover, sawdust–cement mortar (RR) materials contributed a 9% (13.4%) reduction in operational energy and a 9% (13.3%) decrease in CO2 emissions. The project highlights the potential to utilize sawdust—a waste material—and RR material as wall insulation to decrease intense operational energy demand.
Open Access
A novel approach to explain the black-box nature of machine learning in compressive strength predictions of concrete using Shapley additive explanations (SHAP)
(Elsevier, 2022-04) Ekanayake, I.U; Meddage, D. P. P; Rathnayake, U
Machine learning (ML) techniques are often employed for the accurate prediction of the compressive strength of concrete. Despite higher accuracy, previous ML models failed to interpret the rationale behind predictions. Model interpretability is essential to appeal to the interest of domain experts. Therefore, overcoming research gaps identified, this research study proposes a way to predict the compressive strength of concrete using supervised ML algorithms (Decision tree, Extra tree, Adaptive boost (AdaBoost), Extreme gradient boost (XGBoost), Light gradient boosting method (LGBM), and Laplacian Kernel Ridge Regression (LKRR). Alternatively, SHapley Additive exPlainations (SHAP) – a novel black-box interpretation approach - was employed to elucidate the predictions. The comparison revealed that tree-based algorithms and LKRR provide acceptable accuracy for compressive strength predictions. Moreover, XGBoost and LKRR algorithms evinced superior performance (R ¼ 0.98). According to SHAP interpretation, XGBoost predictions capture complex relationships among the constituents. On the other hand, SHAP provides unified measures on feature importance and the impact of a variable for a prediction. Interestingly, SHAP interpretations were in accordance with what is generally observed in the compressive behavior of concrete, thus validating the causality of ML predictions.
Open Access
Predicting Bulk Average Velocity with Rigid Vegetation in Open Channels Using Tree-Based Machine Learning: A Novel Approach Using Explainable Artificial Intelligence
(MDPI, 2022-06-10) Meddage, D. P. P; Ekanayake, I. U; Herath, S; Gobirahavan, R; Muttil, N; Rathnayake, U
Predicting the bulk-average velocity (UB) in open channels with rigid vegetation is complicated due to the non-linear nature of the parameters. Despite their higher accuracy, existing regression models fail to highlight the feature importance or causality of the respective predictions. Therefore, we propose a method to predict UB and the friction factor in the surface layer (fS) using tree-based machine learning (ML) models (decision tree, extra tree, and XGBoost). Further, Shapley Additive exPlanation (SHAP) was used to interpret the ML predictions. The comparison emphasized that the XGBoost model is superior in predicting UB (R = 0.984) and fS (R = 0.92) relative to the existing regression models. SHAP revealed the underlying reasoning behind predictions, the dependence of predictions, and feature importance. Interestingly, SHAP adheres to what is generally observed in complex flow behavior, thus, improving trust in predictions
Open Access
A Simplified Mathematical Formulation for Water Quality Index (WQI): A Case Study in the Kelani River Basin, Sri Lanka
(Fluids, 2022-04-23) Meddage, D. P. P; Azamathulla, H; Pandey, M; Rathnayake, U; Makumbura, R. K
Surface water quality is degraded due to industrialization; however, it is one of the widely used sources for water supply systems worldwide. Thus, the polluted water creates significant issues for the health of the end users. However, poor attention and concern can be identified on this important issue in most developing countries, including Sri Lanka. The Kelani River in Sri Lanka is the heart of the water supply of the whole Colombo area and has the water intake for drinking purposes near an industrialized zone (Biyagama). Therefore, this study intends to analyze the effect of industrialization on surface water quality variation of the Kelani River basin in Sri Lanka in terms of the water quality index (WQI). We proposed a regression model to predict the WQI using the water quality parameters. Nine water quality parameters, including pH, total phosphate, electric conductivity, biochemical oxygen demand, temperature, nitrates, dissolved oxygen, chemical oxygen demand, and chlorine evaluated the Kelani River water quality. The proposed regression model was used to examine the water quality of samples obtained at twelve locations from January 2005 to December 2012. The highest WQI values were found in Raggahawatte Ela throughout the 8 years, located near the Biyagama industrial zone. The relationship of industries to water quality in the Kelani River is stated. The surface water quality gradually decreased as a result of development and industrialized activities. Therefore, this work showcases and recommends the importance of introducing necessary actions and considerations for future water management systems.