Engineering Journal of Research and Development

LAND USE MAPPING WITH R AND MACHINE LEARNING: ANALYZING PASSIVE AND ACTIVE REMOTE SENSING DATA IN KHON KAEN CITY

Wanwisa Inthapasan, Chattichai Waisurasingha, Chutima Waisurasingha — Fri, 29 Mar 2024 00:00:00 +0700

This study focuses on the classification of Sentinel-2, Sentinel-1, and their hybrid satellite imagery using the R programming language and a range of machine learning techniques, including K-nearest neighbors (KNN), Artificial Neural network (ANN), Support Vector Machine (SVM), Random Forest (RF), and Decision Tree (DT). These techniques were applied to both passive and active satellite image data from Khon Kaen city in 2020 to evaluate their classification accuracy. Results showed that Sentinel-2 satellite images could be classified with a peak accuracy of 92%, the highest among the tested data sets. We also found that increasing the volume of training data could potentially enhance the classification accuracy for both Sentinel-2 and Sentinel-1 imagery. The findings underscore the potential of using R for further applications in urban growth analysis, given the right data. This highlights the significant role that machine learning can play in advancing our understanding of urban landscapes through remote sensing data analysis.

KEY SUCCESS FACTORS FOR THE CONSTRUCTION PROJECT OF MASS RAPID TRANSIT IN HEAVY RAIL TYPE

Nuttaya Eanghong, Mongkol Ussavadilokrit, Dundusid Porananond — Fri, 29 Mar 2024 00:00:00 +0700

The construction project of mass rapid transit in heavy rail type is an infrastructure endeavor with high complexity and substantial budget allocation. It involves coordination among various stakeholders and demands consideration of numerous factors for project success. This research aims to explore the success factors in mass rapid transit construction projects and compare them under two contract models: Design-Build and Design-Bid-Build (or Detailed Design). This qualitative research gathered data through interviews with 13 project owners. The quality of research tool has been assessed before data collection. And the research also assessed data saturation through triangulation methods to ensure accuracy and alignment with the research objectives. The research found that the success factors for construction projects of mass rapid transit in heavy rail type consist of a total of 45 factors. Importantly, these factors manifest at different stages throughout the project life cycle. Finally, the comparative analysis of the Design-Build and Design-Bid-Build contract models revealed no significant differences in the majority of factors contributing to the success of construction project of mass rapid transit in heavy rail type.

THE RECYCLING OF SOIL CONTAMINATED WITH IRON RUST FOR FABRICATE CERAMIC TILES

Seree Tuprakay — Fri, 29 Mar 2024 00:00:00 +0700

The purpose of this research is to study the chemical properties of soil contaminated with iron rush. The study aims to determine the optimal mixing ratios of iron rush-contaminated soil with other soil and raw materials used in high-temperature firing for the development of ceramic tiles. The chemical analysis revealed that iron (Fe) was the element present in the highest quantity, with a concentration of 282,715 mg/L (141,357.5 mg/kg of contaminated soil). The composition of the iron-contaminated soil includes aluminum oxide (Al₂O₃), iron oxide (Fe₂O₃), potassium oxide (K₂O), titanium dioxide (TiO₂), manganese dioxide (MnO₂), magnesium oxide (MgO), phosphorus pentoxide (P₂O₅), chloride (Cl), calcium oxide (CaO), chromium oxide (Cr₂O₃), nickel oxide (NiO), copper oxide (CuO), zinc oxide (ZnO), and sulfur trioxide (SO₃). Among these components, Fe₂O₃ was found to be the most abundant, ranging from 72.04% to 78.85 % by molecular weight and loss of ignition (L.O.I.) 14.655%. The average density of the soil 1,632.50 kg/m³, the average moisture content 6.72 %. These values indicate that the density of the soil is similar to that of normal soil, which typically falls within the range of 1,492 kg/m³. The color of the soil is reddish-brown, resembling fired bricks. After the soil was developed into ceramic tiles, the physical properties were evaluated. The highest shrinkage, at 8.97%, was observed in the P60 formula fired at a temperature of 1,100 °C. The color of the tiles after firing varied from dark brown to orange, depending on the mixing ratio of the iron-contaminated soil, which contributes to the red-brown color. Increasing the proportion of iron-contaminated soil resulted in decreased tile strength, while higher firing temperatures increased tile strength. The highest flexural strength value, at 77.07 kg/cm², was obtained from the P80 formula fired at 1,100 °C. The P80 tile formula exhibited water absorption of more than 10% and a modulus of rupture (MOR) of less than 7 N/m², meeting the industrial standards (TSIS 2508-2555).

COMPARISON OF BIAS CORRECTION USING GSMAP PRODUCTION AND OBSERVED RAINFALL IN KHON KAEN PROVINCE

Patsakorn Laochai, Chalermchai Pawattana — Fri, 29 Mar 2024 00:00:00 +0700

Some areas of Khon Kaen province always face flood problem due to heavy rainfall. Forecasting of rainfall using GSMap products will encourage flood management and warning. Therefore, the objective of this study was to compare the accuracy of the three bias correction methods of Bias collection, Distribution transformation and Spatial bias. The data of 7 stations for both observed and GSMap daily rainfall during May 1, to September 30, 2021 were utilized to test the accuracy. The results of accuracy using R2 index were 0.30, 0.30 and 0.95 for Bias collection, Distribution transformation and Spatial bias methods, respectively. Testing the accuracy using RMSE (mm.) for Bias collection, Distribution transformation and Spatial bias methods reveals values of 7.57, 7.75 and 2.15, respectively. Additionally, Spatial bias method was suitable to apply for bias correction.

SEISMIC SITE CLASSIFICATION OF NAN CITY, NORTHERN THAILAND

Thanop Thitimakorn, Narongsak Rachukarn, Sasikan Kupongsak — Fri, 29 Mar 2024 00:00:00 +0700

The primary goal of this study is to generate the NEHRP soil classification map for Nan City using the average shear wave velocity values (Vs₍₃₀₎) derived from the multichannel analysis of surface wave (MASW) data. The secondary goal is to use the Vs₍₃₀₎ data to create the preliminary site amplification map of the area. For this work, MASW data were acquired at 36 preselected sites in the Nan City area. After generating the NEHRP map, it is found that soil class D is present mostly in the central and southeast part of the area, while soil class C is found mainly in the western, eastern, and southern parts. A major part of the city is located on soil class D. The soil amplification map indicates higher amplification in the central and southeast part of the city, where the soil consisted mainly of soft sediments from the alluvial plain and the river terrace. The western, eastern, and southern parts of Nan City had a relatively low amplification, perhaps because the sediment in this part is relatively thin or the bedrock is shallow. The results of the study imply that the major part of Nan city may experience earthquake ground shaking due to amplification of the soft soils.

SATELLITE DERIVED BATHYMETRY PRODUCTION CASE STUDY AT DEEPWATER PORT OF MAPTAPHUT INDUSTRIAL PARK RAYONG

Thepchai Srinoi, Thirawat Bannakulpiphat, Phisan Santitamnont, Prajuab Riabroy — Fri, 29 Mar 2024 00:00:00 +0700

The satellite-derived bathymetry model (SDB) is an alternative way of making elevation data for large nearshore areas. This is essential for geological study, coastal environment, and transportation management in deepwater ports because the nearshore depth around the port should be deeper than on a typical beach. This study is about finding the bathymetry production method using Sentinel-2 imagery and depth data from a bathymetry boat at Maptaphut Industrial Area Deepwater Port, Rayong. The suitable images were downloaded over a three-month period. This study investigated the model production from Lyzenga and Stumpf's empirical formulas and blue band selection (bands 1 and 2 from the sentinel-2 image) for better model accuracy. The results showed that model accuracy is about 2–5 m, Lyzenga algorithm was probably same with Stumpf algorithm and the blue band in band 2 was better than band 1. The main limitation of model production in this study was that a lot of turbidity covered the water surface, so the depth estimation model was shallower than field depth data, resulting in lower accuracy in these areas

ANALYSIS OF THE RELATIVE IMPORTANT WEIGHT OF FACTORS AFFECTING LABOR PRODUCTIVITY IN CONSTRUCTION PROJECT

Tewakun Chankampom, Korb Srinavin — Fri, 29 Mar 2024 00:00:00 +0700

One of the essential components of the economies of all countries worldwide is the construction industry. Especially developing countries. This is due to the fact that this industry involves relatively high investments. Therefore, it has a significant impact on the country's economy. The construction sector can contribute approximately 15-20% of the gross domestic product. The productivity in construction often reflects labor productivity because it depends on labor and uses a lot of labor. Global construction productivity has declined over the past five decades. As in the past 10 years, Thailand's labor productivity has shown a downward trend. Therefore, having a highly productive workforce at each stage of a construction project will play a key role in project success. This research, therefore, aims to find factors affecting the decline in labor productivity in the construction industry. By collecting data from 336 construction workers and analyzing the relative importance weight (RIW), the results found that the factor that first affects the decline in labor productivity is "rain" with a weight value RIW 58.87 percent, next in line with "Electricity/water supply interruptions", "Construction sites are noisy/too much dust or trash", "Working at heights" and "Frequent rework due to document, drawing or specification errors". which have RIW values of 51.67, 49.05, 46.96, and 46.25 percent, respectively.