KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 12

Some of the metrics are blocked by your
consent settings
ENHANCING PLATFORM CREDIBILITY IN THAI E-COMMERCE: RECALIBRATING INFLATED STAR RATINGS VIA SENTIMENT ANALYSIS
(2026-01-01)
Leenawong, Chartchai
;
Thangthawornkit, Warunada
;
Ritthipakdee, Amarita
This study addresses the erosion of platform credibility in international e-commerce caused by the misalignment between numerical star ratings and actual textual sentiment. While star ratings are a dominant mechanism guiding purchasing decisions in rapidly expanding markets like Southeast Asia, they are frequently inflated due to platform-induced biases and region-specific cultural factors, such as the politeness norms and indirect communication strategies prevalent in Thai culture. To overcome this structural limitation, this research aims to empirically quantify this discrepancy and introduce a transparent, sentiment-informed metric—the "Heart Score"—to recalibrate ratings. A rule-based sentiment analysis framework specifically adapted for Thai-language reviews was applied to a dataset of 14,877 women’s clothing entries from Lazada. By leveraging expert-validated lexicons and linguistic rules for negation and amplification, store-level sentiment ratios were extracted and transformed into recalibrated scores via a normalized linear function with penalty adjustments. Findings confirm substantial sentiment misalignment: although the analyzed stores clustered narrowly with star ratings between 4.92 and 5.00, the recalibrated Heart Scores spanned a significantly wider, more meaningful range of 3.81 to 4.71. The most notable gap of 1.11 points exposed hidden dissatisfaction masked by near-perfect ratings. The proposed framework offers strong managerial insights for platform governance by enhancing review credibility, enabling more accurate product benchmarking, and providing a scalable, interpretable model for restoring trust in low-resource language environments.
Some of the metrics are blocked by your
consent settings
Fuel Costs Optimization for Long-Haul Flight with Refueling Layovers
(2024-01-01)
Leenawong, Chartchai
;
Ritthipakdee, Amarita
This study introduced a mathematical model aimed at optimizing fuel costs for long-haul flights, particularly those requiring refueling. The primary objective was to minimize fuel expenses by considering key factors such as flight routes, aircraft types, refueling points, and refueling quantities. The proposed solution used a 0-1 mixed-integer linear programming (MILP) model, supported by auxiliary variables, to effectively manage the constraints of this optimization problem. The MILP model also considered differences in fuel costs at refueling points, including the departure airport. For validation, a case study was conducted involving a long-haul flight from airport AAA to DDD, with refueling options at airports BBB and CCC. The model effectively determined the most economical flight route, assessed the necessity of refueling, and calculated the required fuel amounts at each refueling location. In summary, this study demonstrated that the proposed model could successfully address the challenges of optimizing fuel costs in long-haul flight.
Some of the metrics are blocked by your
consent settings
Modified K-Means Clustering for Demand-Weighted Locations: A Thailand’s Convenience Store Franchise-Case Study
(2023-03-01)
Leenawong, Chartchai
;
Chaikajonwat, Thanrada
This research applies and modifies K-means clustering analysis from Data Mining to solving the location problem. First, a case study of Thailand’s convenience store franchise in locating distribution centers (DCs) is conducted. Then, the final centroids are served at suggested DC locations. Besides the typical distance, Euclidean, used in K-means, Manhattan, and Chebyshev, is also experimented with. Moreover, due to the stores’ different demands, a modification of the centroid calculation is needed to reflect the center-of-gravity effects. For the proposed centroid calculation, the above three distance metrics incorporating the demands as weights give rise to another three approaches and are thus named Weighted Euclidean, Weighted Manhattan, and Weighted Chebyshev, respectively. Besides the optimal locations, the effectiveness of these six clustering approaches is measured by the expected total distribution cost from DCs to their served stores and the expected Davies– Bouldin index (DBI). Concurrently, the efficiency is measured by the expected number of iterations to the final clusters. All these six clustering approaches are then implemented in the case study of locating eight DCs to distribute to 260 convenience stores in Eastern Thailand. The results show that though all approaches yield locations in close proximity, the Weighted Chebyshev is the most effective one having both the lowest expected distribution cost and lowest expected DBI. In contrast, Euclidean is the most efficient approach, with the lowest expected number of iterations to the final clusters, followed by Weighted Chebyshev. Therefore, the DC locations from Weighted Chebyshev could, ultimately, be chosen for this Thailand’s convenience store franchise.
Some of the metrics are blocked by your
consent settings
Event Forecasting for Thailand’s Car Sales during the COVID-19 Pandemic
(2022-07-01)
Leenawong, Chartchai
;
Chaikajonwat, Thanrada
The COVID-19 pandemic that started in 2020 has affected Thailand’s automotive industry, among many others. During the several stages of the pandemic period, car sales figures fluctuate, and hence are difficult to fit and forecast. Due to the trend present in the sales data, the Holt’s forecasting method appears a reasonable choice. However, the pandemic, or in a more general term, the “event”, requires a subtle method to handle this extra event component. This research proposes a forecasting method based on Holt’s method to better suit the time-series data affected by large-scale events. In addition, when combined with seasonality adjustment, three modified Holt’s-based methods are proposed and implemented on Thailand’s monthly car sales covering the pandemic period. Different flags are carefully assigned to each of the sales data to represent different stages of the pandemic. The results show that Holt’s method with seasonality and events yields the lowest MAPE of 8.64%, followed by 9.47% of Holt’s method with events. Compared to the typical Holt’s MAPE of 16.27%, the proposed methods are proved strongly effective for time-series data containing the event component.
Some of the metrics are blocked by your
consent settings
Event index computation for forecasting case study: Car sales in Thailand
(2020-12-01)
Rattanametawee, Witchaya
;
Leenawong, Chartchai
Due to the impact of special events, both positive and negative, on the sales data, the ordinary Time-series Decomposition (TSD) forecasting model cannot merely capture these effects, even with the added seasonality and trends. Therefore, in this research, a new method for computing the event indices, representing the unusual fluctuations for a certain period in the time series, is proposed in order for it to be incorporated into TSD, alongside the conventional trend, seasonal, and cyclical components. A case study of subcompact car sales monthly data in Thailand during the years 2011-2018 is examined as for that time period contains the 2011 nationwide big flood reflecting the negative impact, as well as the nation’s tax-incentive first-car buyer scheme reflecting the positive impact on the dataset. The mean absolute percentage error (MAPE) is used as an accuracy measure of the proposed forecasting model and it illustrates the promising results in the end.
Some of the metrics are blocked by your
consent settings
Decision Support Model and Software for Consolidated Order Assignment to Delivery Trucks
(2017-11-15)
Leenawong, Chartchai
;
Wattanawalun, Champ
;
Wongsa, Kittipish
;
Lethaisong, Karnkanok
In this research, a decision model for assisting in loading consolidated orders onto delivery trucks to distribute to scattered customers is constructed. Each truck can have more than one drop-off location. The case-study company based in Bangkok produces bubble wraps that can simply be stacked up on top of the companys small trucks. Most customer orders are less than truck-load. Thus, consolidating orders into full trucks according to customer locations is necessary to ensure minimum transportation and in-transit inventory carrying costs. 0-1 integer programming is used in the model construction. Also, a decision support system for dealing with this problem is developed for ease of use to logistics managers and planners.
Some of the metrics are blocked by your
consent settings
Multiple linear regression using gradient descent: A case study on Thailand car sales
(2017-01-01)
Netisopakul, Ponrudee
;
Leenawong, Chartchai
Severe fluctuations in Thailand car sales had enormous impacts on the automobile and related industries. A reliable forecasting model is needed to accurately forecast the car sales for the next production batch. Using ten-year car sales data, this research proposes a machine learning approach using gradient descent (GD) to fitting multiple linear regression for Thailand car sales forecasts. The resulted forecasting accuracy is then compared with that of a normal equation method (NE) as well as that obtained from a statistical package (SP). First, two independent variables (2IVs): Thailand’s Gross Domestic Product and the 12-month Loan Rate are used in the proposed models. Then, dummy seasonal variables (Season) are added to the regression equations. Finally, dummy event flag variables (Event) are added. Totally, five sets of experiments are conducted. The experiment results show that NE produces the same regression equations as SP. Both GD and NE methods yield exactly the same results for 2IVs, but GD yields slightly less prediction accuracy than NE’s in Season and Event experiments. This research concludes that gradient descent has comparable forecasting accuracy to those from other methods. Nevertheless, when the regression contains dummy variables, caution is recommended.
Some of the metrics are blocked by your
consent settings
The effects of special events on regression for subcompact car sales in Thailand
(2016-11-01)
Rattanametawee, Witchaya
;
Leenawong, Chartchai
;
Netisopakul, Ponrudee
This research proposes a method to dealing with multiple linear regression that integrates the seasonality as well as the effects of some special or unanticipated events for sales figures. The method is then applied to the car sales figures in Thailand after having been through the 2011 national big flood and the 2011-2012 government’s initiative tax-incentive program for boosting the automobile industry. Besides Thailand’s Gross Domestic Products (GDP) and the 12-month Loan’s Interest Rate as explanatory variables, seasonal dummy variables along with the proposed special event variables and appropriate event tagging are incorporated. The statistical results obtained from the proposed regression model with seasons and events, compared to the models with neither seasons nor both yields highest adjusted coefficient of determination (R-squre) and accuracy (MAPE).
Some of the metrics are blocked by your
consent settings
Computational models for studying leadership modes
(2010-03-01)
Wattanasiripong, Nisakorn
;
Wang, Karen
;
Leenawong, Chartchai
Computational models for studying the effects of permanent and rotating leadership on the performances of teams that have full interaction among team members are proposed. In each model, the leader performs both leading roles and regular tasks as an ordinary team member. Permanent leadership refers to the situation when a team has only one leader for the entire time the team exists while rotating leadership refers to the situation when every team member rotates for the leader position. Computer simulations are used for examining the effects of the parameters in these models such as the amount of time the team exists for, the learning ability of a member, and the skill level of the leader on the expected team performance. © 2010 Pushpa Publishing House.
Some of the metrics are blocked by your
consent settings
Quantitative models for forecasting vehicle fuel prices in Thailand
(2009-11-16)
Tipyan, Jenjira
;
Leenawong, Chartchai
In this research, quantitative models for forecasting vehicle fuel selling prices at a gas station in Thailand are investigated. Four types of gasoline include Gasoline 95, Gasoline 91, Gasohol 95, and Diesel. The data are drawn from the year 2002 to 2008. Time series and regression methods are used in this research. A composite method is also proposed using three different approaches in assigning the weights. Firstly, the weights minimize the variance of the combined error. Secondly, the weights are taken from a regression. Thirdly, the weights are all equal. The results such as the best model for prices of Gasoline 91 and Gasohol 95 is the combined method of using regression in obtaining the weights are presented. © 2008 IEEE.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results