Ahmad Shukri Yahya

Soil moisture content and density prediction using laboratory resistivity experiment

Natural soils are an intimate mixture of solid, liquid and gas phases. This study establishes a correlation for moisture content and density of a soil with its electrical resistivity. In the past, most of the conventional geotechnical site investigation required bulky and heavy equipment to determine the geotechnical parameters necessary for design and construction purposes. Consequentially, time and cost of the project is increased especially when dealing with some difficult site such as on mountainous terrain. This study is based on laboratory soil box resistivity meter observations made on soils mixed with additions of consistent increments of 1-5 % of water to 1500 gram of remolded soils in loose condition. At least 24 repetitive resistivity test observations were made and the moisture content and soil density was determined concurrently for each of the tests. The observations showed that the electrical resistivity variation decreased in a curvilinear manner with increasing percentage of moisture content. A regression equation and coefficient of determination, R2 for moisture content against soil electrical resistivity value was established by moisture content, w = 152.87ρ-0.312 (ρ = soil electrical resistivity) and R2 = 0.7718 respectively. While a regression equation and R2 value for bulk density versus soil electrical resistivity value was observed to be ρbulk = -0.107 ln (ρ) + 1.7249 and 0.7016 respectively. Hence, a viable method is demonstrated where the electrical resistivity value was applicable and has a great potential for geotechnical data prediction of parameters such as moisture content and soil density.

GIS Modeling for Landfill Site Selection via Multi-Criteria Decision Analysis: A Systematic Review

Utilizing multi-criteria decision analysis as an alternative to traditional screening processes for GIS modeling for landfill site selection (LSS) has attracted significant interest in recent years because of its time and cost savings and its ability to achieve better validation and accuracy. This paper surveys the developments in the modeling of LSS using geographic information systems (GIS) on the basis of multi-criteria decision analysis (MCDA) in the past two decades from 1997 to 2014. Emphasis is placed on the third and fifth stages of the overall applied methodology (selection of weights and decision rules), as well as on the efficiency of the LSS models. From the review, the strengths and limitations of using MCDA for LSS modeling via GIS are identified. Moreover, artificial neural networks instead of MCDA can be used as a new approach in the third and fifth stages of LSS models to enhance validation and accuracy.

Quantitative workflow based on NN for weighting criteria in landfill suitability mapping

Our study aims to introduce a new quantitative workflow that integrates neural networks (NNs) and multi criteria decision analysis (MCDA). Existing MCDA workflows reveal a number of drawbacks, because of the reliance on human knowledge in the weighting stage. Thus, new workflow presented to form suitability maps at the regional scale for solid waste planning based on NNs. A feed-forward neural network employed in the workflow. A total of 34 criteria were pre-processed to establish the input dataset for NN modelling. The final learned network used to acquire the weights of the criteria. Accuracies of 95.2% and 93.2% achieved for the training dataset and testing dataset, respectively. The workflow was found to be capable of reducing human interference to generate highly reliable maps. The proposed workflow reveals the applicability of NN in generating landfill suitability maps and the feasibility of integrating them with existing MCDA workflows.

Automating an integrated spatial data-mining model for landfill site selection

An integrated programming environment represents a robust approach to building a valid model for landfill site selection. One of the main challenges in the integrated model is the complicated processing and modelling due to the programming stages and several limitations. An automation process helps avoid the limitations and improve the interoperability between integrated programming environments. This work targets the automation of a spatial data-mining model for landfill site selection by integrating between spatial programming environment (Python-ArcGIS) and non-spatial environment (MATLAB). The model was constructed using neural networks and is divided into nine stages distributed between Matlab and Python-ArcGIS. A case study was taken from the north part of Peninsular Malaysia. 22 criteria were selected to utilise as input data and to build the training and testing datasets. The outcomes show a high-performance accuracy percentage of 98.2% in the testing dataset using 10-fold cross validation. The au...

Sustainable GIS Based-ANN's Solution for Landfill Suitability Analysis

Sustainable suitability analysis for landfill sites is an important and necessary issue for authorities of solid waste planning in the fast growing zones, due to the increasing complexity coming from dealing with various disciplines and requirement and the needy of satisfaction. A combination of geographic information systems including spatial analysis, and artificial neural network ANNs were employed in this study for decision-makers in the sustainable suitability analysis problems in Malaysia and GIS was used to manipulate and present spatial data. The GIS analysis reveals three distinct groupings based on actual conditions of the case study area, environmental factors, economic factors and social factors which are reflection of different factors contributing to the sustainable development. The result shown that ANNs has good information extraction and evaluation functions of the suitability value based on the exact relationship between the input criteria and the output landfill site data with high coefficient of determination (R2) which help decision-makers to analysis sustainable suitability for landfill sites.

Effect of Roadside Development on the Safety of Intersections without Signals

Roadside development plays an important role in the safety of the entire road in general and intersections in particular. Illegal and improper buildouts such as roadside kiosks, bus stops and fire stations are an accident hazard and a hindrance towards sustainable development. This paper presents the first account of analysis of safety evaluation of intersections with respect to roadside development in Malaysia. The data collected on 14 sites was analyzed. The results indicated that the bus stops and restaurants located right infront of the unsignalized intersections were the major contributory factors in decreasing the safety. Moreover it was found that facilities staggered away from the intersection such as another access point are less hazardous. It is recommended that relevant authorities should ensure the removal of such risky roadside developments.

H2S Emissions within Vicinity of Swine Farming Area in Nibong Tebal, Pulau Pinang, Malaysia

Animal farming industries is important in Malaysia because its contribution to the economy. The production causes all the major environmental negative impact such as water pollution from waste water and malodours emanating from farms. The current methods of disposing of manures were no longer adequate or suitable for the new, large and intensive animal farming. Inappropriate technologies, poor maintenance and inadequate dimensions and design of the treatment systems in addition to inappropriate production method are causing serious environmental problems especially odorous emissions. Present of Hydrogen Sulphide (H2S) in the air contribute to the odour pollution. Static monitoring has been done and maximum concentration of H2S is 220.6 ppb. Temperature and relative humidity fluctuations were seem to have influence on concentration of H2S. Linear regression analysis was shown that relative humidity has higher influence on correlation gaseous pollution compared to temperature. Correlation coefficient for H2S and relative humidity was in range 0.675 to 0.881 in the morning and 0.417 to 0.729 in the evening, which are in range of strong correlations.