Sirikasemsuk, Kittiwat

Preferred name

Alternative Name

Sirikasemsuk, K.

Main Affiliation

School of Engineering

kittiwat.si@kmitl.ac.th

Now showing 1 - 10 of 27

Some of the metrics are blocked by your
consent settings
Analysis of two-missing-observation 4 × 4 latin squares using the exact approach
(2018-01-01)
Sirikasemsuk, Kittiwat
;
Leerojanaprapa, Kanogkan
This research deals with the analysis of incomplete Latin square designs using the exact approach. Specifically, the study investigated the 4 × 4 Latin square designs with two missing observations without replication. In the research, the general regression significance test (i.e. the exact approach) was used to derive the estimation formulae of fitted parameters for the full and reduced linear statistical models, thereby simplifying the calculation process. In addition, the proposed exact approach-based formulae facilitate the determination of the treatment sum of squares and the error sum of squares, both of which are subsequently employed in the analysis of variance (ANOVA).
Some of the metrics are blocked by your
consent settings
Full-model regression sum of squares of randomized complete block design having one unrecorded observation
(2019-01-01)
Sirikasemsuk, Kittiwat
;
Leerojanaprapa, Kanogkan
;
Sirikasemsuk, Sirisak
In the classical design of experiment, a randomized complete block design (RCBD) is a helpful experimental design because this design uses a small number of experimental units. The randomized complete block design is comprised of two factors, i.e., a nuisance factor and a potential factor. In many real experiments, some data might be missing or unrecorded. The instant formulae were not provided for an analysis of variance in this case. This paper took into account the randomized complete block design (RCBD) with a treatments and b blocks. In this contribution, the RCBD with an unrecorded value was analyzed by means of the exact scheme with the model comparison approach. The advantages of the exact scheme are that the unrecorded value for the unfilled cell is not estimated and the treatment sum of squares is unbiased. It is important to note that there is no ready-made formula for RCBD in the past. Hence, this research paper provided the mathematical formulae for the fitted parameters and the overall regression sum of squares (ORSS) for the full model of experimental data. It is also noted that the ORSS is imperative for the analysis of the variance by means of the exact scheme.
Some of the metrics are blocked by your
consent settings
Modeling Equivalent Circulating Density During Drilling Operations in the Gulf of Thailand
(2026-03-31)
Leerojanaprapa, Kanogkan
;
Suttaloon, Sudarat
;
Bhundarak, Komn
;
Sirikasemsuk, Kittiwat
Equivalent Circulating Density (ECD) represents the total hydrostatic pressure generated by drilling fluid while in motion. This prevents the internal pressure within the well from exceeding the fracture resistance of the rock, which could lead to lost circulation and an inability to effectively control the wellbore pressure. This research aims to predict ECD in 6.125-inch production section in Gulf of Thailand field by using five machine learning algorithms were utilized, namely Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Networks (ANN), Gradient Boosting (GB), and Extreme Gradient Boosting (XGBoost). Various sensors from the Measure While Drilling (MWD), Logging While Drilling (LWD), and Pressure While Drilling (PWD) tools were used to collect raw data, totaling 38,863 records and 24 variables to predict the ECD value. The dataset was randomly split into 80% for training and validation and 20% for testing. The results indicate that the RF technique outperformed the other models in predicting ECD values, producing the lowest RMSE of 0.031. Therefore, the RF model is most suitable for further development and real-time application in predicting ECD values.
Some of the metrics are blocked by your
consent settings
Optimization of Process Conditions for Hard Disk Drive Assembly for Defect Reduction
(2023-11-01)
Sirikasemsuk, Kittiwat
;
Leerojanaprapa, Kanogkan
A spoiler installed in a hard disk drive can reduce the airflow velocity that causes vibration at an actuator arm and its head. In the case study, a hard disk drive manufacturer identified the defect of a loosely tightened spoiler in the hard disk drive by the spoiler installation machine. This occurred as a result of incorrect screw spacing (also called ‘the screwdriver encoder’). There was no study that specifically specified or created a suitable relationship model between the screwdriver encoder and associated parameters for the installation of a spoiler. Factors affecting the screw spacing were determined using multiple regression analysis, and a mathematical model of the significant factors of the screwdriver encoder was built. Data were collected from the spoiler installation machine and its software. Four factors were identified with the potential to impact the screwdriver encoder as (1) vacuum level for picking up the screw, (2) time spent in tightening the screw, (3) bit angle, and (4) screw torque. The ANOVA results pointed out that the screw torque within the range of experimental values did not affect the screw spacing, and a quadratic regression model was the most appropriate under various statistical criteria. This research demonstrated that a sophisticated regression model, i.e., a cubic model, is not always a good choice for an agent. In addition, the optimal values for the spoiler installation machine were determined. The faults decreased by 0.0096% from 0.055% each month. For choosing the relationship model of additional workpiece screw tightening variables, this research phase can be utilized as a guideline.
Some of the metrics are blocked by your
consent settings
Industry cluster using cluster analysis to support industry cluster policy of Thailand
(2019-06-01)
Leerojanaprapa, Kanogkan
;
Bhundarak, Komn
;
Sirikasemsuk, Kittiwat
This research implies clustering into three clusters for five industries by means of k-Means clustering method. Agro-processing, textiles and clothing, petrochemicals and chemicals, electronics and telecommunications equipment and automotive and parts are selected for this study as they are main target promoting industries for Thailand. There were 23,628 firms in this study. From this study, we can identify different patterns of demanded resources in three different groups. One cluster required low level of resourced demands for all variables while the other two clusters required capital and manpower interchangeable between high and medium level. Only the electronics and telecommunications equipment sector showed high to medium demand for all variables. After all firms were divided into three clusters, we were able to define cluster regions by provinces, for particular types of clusters in order to evaluate the potential of each region and also define supporting policy for those firms to meet their demands following the regional economic development strategy.
Some of the metrics are blocked by your
consent settings
Applying Bayesian network for noncommunicable diseases risk analysis: Implementing national health examination survey in Thailand
(2017-07-02)
Leerojanaprapa, K.
;
Atthirawong, W.
;
Aekplakorn, W.
;
Sirikasemsuk, K.
We propose using a Bayesian network to capture and understand the dependency risk factors affecting the prevalence of chronic diseases. By applying a Bayesian network model, we can visualize interdependencies between risks and their effects on the Noncommunicable disease (NCD) prevalence. By using a Bayesian network to model the prevalence of diabetes, we can define the top three risks as family history of diabetes, obesity, and age. Furthermore, the risk classification results can help to determine the managing strategy. For the Thai population, problems arising from family history of diabetes and obesity can be met by employing a transfer strategy. Age (especially ages of 35-59) and the risk incurred by low intake of fruits and vegetables should use a reduction or mitigation strategy. Finally, those at risk as a result of their area of residence (in urban areas) and socio-economic factors within the 4<sup>th</sup> quantile and low level of physical activity should apply a retain strategy.
Some of the metrics are blocked by your
consent settings
Improving Automotive Tire Defect in the Vulcanization Process by Leakage Bladders Issues
(2024-01-01)
Kiatcharoenpol, Tossapol
;
Sawangnimitkul, Penpisut
;
Junmewong, Rattanaporn
;
Sirikasemsuk, Kittiwat
The primary objective of this project is to minimize defects in industrial tires resulting from issues with leaking bladders during the curing process. The project aims to reduce the current monthly defective tire production of 1,378.70 kilograms to a target of 1,240.83 kilograms per month, which represents a 10% reduction in defects. Utilizing the QC story methodology, QC 7-tools, and Quality Control techniques, a thorough examination of data and root cause analysis identified five significant factors contributing to defective tires caused by bladder leakage. These factors include 1) a lack of expertise in bladder inspection, 2) excessive usage of bladders, 3) contaminations on the bladder surface, 4) inadequate bladder lubrication and 5) oxidation occurring inside the bladder. To address these issues, a new operational procedure and training program were introduced, alongside modifications to the curing machine to mitigate the oxidation process. Over a span of two months, these changes resulted in a reduction in defective tires to 735.345 kilograms per month, equivalent to a 46.66% decrease in the problem, and a cost savings of 10,000 USD per year.
Some of the metrics are blocked by your
consent settings
Solving the incomplete data problem in Greco-Latin square experimental design by exact-scheme analysis of variance without data imputation
(2024-01-01)
Sirikasemsuk, Kittiwat
;
Wongsriya, Sirilak
;
Leerojanaprapa, Kanogkan
This study introduced a novel exact-scheme analysis of variance to tackle the challenge of incomplete data within the Greco-Latin square experimental design (GLSED), specifically for scenarios with a single missing observation across any treatment and block level, thus eliminating the need for conventional data imputation methods. This approach innovatively addresses and mitigates the bias in the treatment sum of squares, a significant drawback of traditional missing plot techniques, by providing a precise, exact-scheme-based formula for calculating the treatment sum of squares in fixed-effect GLSED contexts with unrecorded values. Moreover, it offers a method for correcting biased treatment sum of squares values, presenting an adjustment mechanism for instances where the least squares method was previously employed to estimate missing values. This comprehensive strategy not only enhances the methodological accuracy and integrity of GLSED studies but also contributes significantly to the field by offering a solution to navigate the complexities of incomplete datasets without resorting to data imputation, thus improving the rigor and validity of experimental designs in the face of missing data challenges.
Some of the metrics are blocked by your
consent settings
One missing value problem in Latin square design of any order: Exact analysis of variance
(2017-01-01)
Sirikasemsuk, Kittiwat
;
Leerojanaprapa, Kanogkan
This research proposes a simplified exact approach based on the general linear model for solving the K × K Latin square design (LSD) with one replicate and one missing value, given the lack of ready-made mathematical formulas for the sub-variance. Under the proposed scheme, the effects of the potential variable were determined by means of the regression sums of squares under the full and reduced treatment models. The mathematical expressions could be applied to the LSD with one missing value of any order. Moreover, the treatment, row and column sums of squares are unbiased.
Some of the metrics are blocked by your
consent settings
Measure of bullwhip effect in supply chains with first-order bivariate vector autoregression time-series demand model
(2017-02-01)
Sirikasemsuk, Kittiwat
;
Luong, Huynh Trung
With supply chains becoming increasingly global, the issue of bullwhip effect, a phenomenon attributable to demand fluctuation in the upstream section of the supply chains, has received greater attention from many researchers. However, most existing research studies on quantifying the bullwhip effect were conducted under the first-order autoregressive [AR(1)] incoming demand process or its variants as the fundamental demand process, thereby failing to account for the retailer demand dependency. This research work thus examined the bullwhip effect for the first-order bivariate vector autoregression [VAR(1)] demand process in a two-stage supply chain consisting of one supplier and two retailers. The impacts of the correlation parameters of the demand process, the correlation coefficient between the two error terms, and the variances of the error terms on the bullwhip effect were investigated. As such, the measure of the bullwhip effect was established using an analytical approach in which the minimum mean square error (MMSE) forecasting method and the base stock policy were applied to all members of the supply chain. Numerical experiments were then conducted to illustrate the behavior of the bullwhip effect with respect to various parameters of the demand processes to see in which situations the bullwhip effect would be absent. In addition, an evaluation of the inventory variance ratio was analyzed.

Sirikasemsuk, Kittiwat

Filters

Settings

Sort By

Results per page