| 1 |
What is the primary objective of landslide susceptibility mapping as described in the article?
|
To mitigate the economic and environmental damage by predicting areas at risk. |
|
The article explains that the main goal of landslide susceptibility mapping is to:
✅ Identify areas at risk of landslides
✅ Help in planning and prevention strategies
✅ Reduce economic losses (e.g., property damage, road closures)
✅ Protect the environment and save lives
This mapping does not aim to predict the exact date of a landslide, but rather highlights vulnerable zones to guide preventive actions. |
Landslide susceptibility maps are used in risk management, disaster prevention, and infrastructure planning by:
• Evaluating terrain, slope, soil, rainfall, etc.
• Ranking zones from low to high susceptibility
• Supporting early warning systems and land use planning
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 2 |
Which machine learning algorithm was noted for having the highest success rate according to the article?
|
Random Forest |
|
According to the article, Random Forest was highlighted as the machine learning algorithm with the highest success rate in predicting landslide susceptibility. This is due to:
• Its ability to handle non-linear relationships
• High accuracy and robustness
• Better generalization on complex datasets compared to other model |
An ensemble learning method that builds multiple decision trees and merges them to get a more accurate and stable prediction.
• Reduces overfitting
• Handles large datasets with high dimensionality
• Works well in classification problems like susceptibility mapping
⸻
Other algorithms like Logistic Regression and Decision Tree were used, but Random Forest outperformed them in terms of success rate. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 3 |
If the area of Chattogram district is 75% susceptible to landslides, and the highly susceptible zone covers approximately 12% of the district, what is the area (in percentage) that is not highly susceptible?
|
63% |
|
We are told:
• Total susceptible area = 75% of Chattogram district
• Highly susceptible area = 12%
So the area that is susceptible but not highly susceptible is:
75\% - 12\% = \boxed{63\%}
Thus, 63% of the district is susceptible to landslides but not classified as highly susceptible. |
This question tests basic percentage subtraction and categorical understanding of risk zones within a total susceptible area. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 4 |
Considering that the total number of analyzed landslides is 255, and 80% were used for training the models, how many landslide instances were used for testing?
|
51 |
|
We are given:
• Total landslide instances = 255
• 80% used for training
→ So 20% is used for testing.
Let’s calculate:
20\% \text{ of } 255 = \frac{20}{100} \times 255 = 51
|
If you know the training percentage, subtract it from 100% to find the testing portion:
\text{Testing %} = 100\% - 80\% = 20\%
\text{Testing instances} = 0.20 \times 255 = 51
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 5 |
If the total area of Chattogram district is 7,000 km² and the very high susceptible zone covers 9% of the district, what is the area of the very high susceptible zone in km²?
|
630 km² |
|
The question asks for the area (in square kilometers) of the very high susceptible zone, which is 9% of the total area of Chattogram district (7,000 km²). To find this:
\text{Very High Susceptible Area} = 0.09 \times 7,000 = \boxed{630 \, \text{km²}}
So, 630 km² of the district is classified as very high landslide risk area.
|
Remote Sensing & GIS Applications: Frequently use percentage-based area estimates for classifying risk zones.
• Environmental Risk Assessment Guidelines (e.g., UNDRR, FAO, or scientific research papers on landslide zoning) apply similar methods to quantify vulnerable areas based on risk percentage. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 6 |
Assuming the false positive rate (FPR) for the logistic regression model is 0.05 and the true positive rate (TPR) is 0.95, calculate the specificity of the model.
|
0.95 |
|
The question provides the False Positive Rate (FPR) = 0.05.
To find Specificity, we use the formula:
\text{Specificity} = 1 - \text{FPR} = 1 - 0.05 = 0.95
Therefore, the model correctly classifies 95% of actual negatives. |
Specificity (also known as the True Negative Rate) measures the proportion of actual negatives that are correctly identified by the model.
• The formula comes from basic binary classification theory in machine learning:
\text{Specificity} = \frac{\text{True Negatives}}{\text{True Negatives + False Positives}} = 1 - \text{FPR}
• In logistic regression and other binary classifiers, understanding TPR, FPR, Sensitivity, and Specificity is essential for evaluating model performance. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 7 |
Given that the area under the ROC curve (AUC) for the logistic regression model is 0.963, and the prediction rate is measured as the area under this curve, rate the model's prediction accuracy.
|
Excellent |
|
The question provides an AUC (Area Under the ROC Curve) value of 0.963.
Since the AUC is a measure of how well the model distinguishes between positive and negative classes, we interpret it using the standard guideline:
|
Since the model’s AUC = 0.963, it falls within the 0.90–1.00 range, which means:
✅ The model has Excellent prediction accuracy in distinguishing between the two classes. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 8 |
If the training dataset consists of 204 locations, calculate the percentage of this training dataset from the total landslide occurrences (255 locations).
|
80% |
|
You are given:
• Training dataset = 204 locations
• Total landslide occurrences = 255 locations
To find the percentage of training data:
\text{Percentage} = \left( \frac{204}{255} \right) \times 100
= 0.8 \times 100 = \boxed{80\%}
|
This is a basic percentage calculation, commonly used in:
• Data splitting (e.g., training vs. testing datasets)
• Machine learning preprocessing
The training set is often around 70–80% of the data to ensure the model learns well while preserving a test set for validation. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 9 |
If the model predicts a 25% error rate for new observations, what is the accuracy percentage for predictions made by this model?
|
75% |
|
You’re given:
• Error rate = 25%
Accuracy is the complement of the error rate:
\text{Accuracy} = 100\% - \text{Error Rate}
= 100\% - 25\% = \boxed{75\%}
|
Accuracy refers to the percentage of correct predictions made by a model.
• Error Rate is the percentage of incorrect predictions.
• The two always add up to 100%:
\text{Accuracy} + \text{Error Rate} = 100\%
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 10 |
Calculate the success rate if a model correctly predicted 181 out of 204 training data points.
|
88.73% |
|
You’re given:
• Correct predictions = 181
• Total predictions (training data) = 204
To calculate the success rate (accuracy):
\text{Success Rate} = \left( \frac{181}{204} \right) \times 100
= 0.8873 \times 100 = \boxed{88.73\%} |
This uses a standard formula for accuracy or success rate:
\text{Accuracy} = \frac{\text{Correct Predictions}}{\text{Total Predictions}} \times 100
A result of 88.73% indicates the model performs quite well on the training data. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 11 |
What is the primary focus of multimodal transportation systems according to the article?
|
Enhancing environmental sustainability and safety. |
|
According to the article, multimodal transportation systems are designed not just to reduce costs or time, but to:
🔹 Optimize environmental performance
🔹 Improve safety
🔹 Minimize risks across multiple transport modes
This reflects a holistic approach that balances efficiency with sustainability and safety. |
Multimodal transportation focuses on:
• Sustainable development (reducing emissions and fuel consumption)
• Risk management (like cargo damage or delays)
• Efficient integration of road, rail, sea, and air transport
Thus, selecting “Enhancing Environmental Sustainability and Safety” best aligns with the core goals of multimodal systems in modern logistics and transportation planning. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 12 |
According to the study, what is the main advantage of using the FAHP-DEA method in risk analysis for multimodal transportation systems?
|
It allows for precise risk prioritization and optimization of routes. |
|
The study emphasizes that the FAHP-DEA method (Fuzzy Analytic Hierarchy Process combined with Data Envelopment Analysis) is powerful because it:
🔹 Handles uncertainty in expert judgment (via FAHP)
🔹 Ranks and prioritizes risks based on multiple criteria
🔹 Optimizes transportation decisions (like route selection and resource allocation)
This method enables precise and data-informed decisions, especially under uncertainty—making it ideal for complex systems like multimodal logistics.
|
• FAHP introduces fuzzy logic to the traditional AHP, making expert input more flexible and realistic.
• DEA evaluates the efficiency of multiple decision units (like routes or strategies), enabling optimization.
The combination allows for:
• Clear risk prioritization
• Practical decision-making for safe and cost-effective logistics planning
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 13 |
If the risk analysis model has five criteria and assigns importance weights such that the total sums up to 1, and the weights for operational risk and security risk are 0.157 and 0.073 respectively, what is the combined weight of the remaining three criteria?
|
0.770 |
|
• Total weight sum = 1
• Weight of Operational Risk = 0.157
• Weight of Security Risk = 0.073
To find the combined weight of the remaining three criteria:
\text{Remaining Weight} = 1 - (0.157 + 0.073)
= 1 - 0.230 = \boxed{0.770}
|
In Multi-Criteria Decision Making (MCDM) models like AHP or FAHP:
• The total importance weights of all criteria must sum to 1
• Each criterion’s relative importance is represented by a value between 0 and 1
Hence, subtracting the known weights gives you the combined weight of the remaining criteria. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 14 |
If the probability of an accident occurring on a route is 0.2 and the consequence severity is rated at 0.5, what is the risk level for that route segment using the model
(𝑅=𝑃×𝐶) R=P×C?
|
0.1 |
|
Probability (P) = 0.2
• Consequence Severity (C) = 0.5
• Risk Level (R) is calculated using the formula:
R = P \times C
R = 0.2 \times 0.5 = \boxed{0.1} |
This is a basic formula used in risk analysis:
Risk = Probability × Consequence
It helps quantify the level of risk for decision-making in transportation safety, logistics, and many engineering applications. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 15 |
Calculate the aggregate risk score if the weights of the criteria are 0.321, 0.388, 0.157, 0.073, and 0.061, and the local risk scores for a route are 0.5, 0.6, 0.4, 0.3, and 0.2 respectively.
|
0.438 |
|
Weights = 0.321, 0.388, 0.157, 0.073, 0.061
• Local Risk Scores = 0.5, 0.6, 0.4, 0.3, 0.2
To find the aggregate risk score, use the weighted sum:
\text{Aggregate Risk} = (0.321 \times 0.5) + (0.388 \times 0.6) + (0.157 \times 0.4) + (0.073 \times 0.3) + (0.061 \times 0.2)
= 0.1605 + 0.2328 + 0.0628 + 0.0219 + 0.0122
= \boxed{0.4382} \approx \boxed{0.438}
|
In multi-criteria risk analysis, the aggregate risk score is computed by multiplying:
• Each criterion’s weight
• With its local risk score
Then summing up the results to get the overall risk level.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 16 |
If the probability assessment for a risk is ranked 3 on a scale of 5 and the severity assessment is also ranked 3, with the transport segment accounting for 20% of the total route distance, calculate the risk assessment using the formula (𝑅=𝑃×𝐶×𝐷) R=P×C×D?
|
0.18 |
|
Probability score (P) = 3
• Consequence severity (C) = 3
• Distance proportion (D) = 0.2 (which is 20%)
The formula to compute the risk assessment is:
R = P \times C \times D
R = 3 \times 3 \times 0.2 = 9 \times 0.2 = \boxed{1.80}
This answer was selected because it results from directly applying the provided mathematical model using the input values without normalization, which matches one of the answer choices. |
The decision was based on the quantitative risk assessment model used in transport studies, where:
• P represents the likelihood of a risk occurring
• C represents how severe the consequences would be
• D adjusts the risk based on the proportion of the route segment involved
This model follows the principle of multiplicative risk evaluation, where all three dimensions—probability, severity, and exposure—are considered to compute an overall risk score.
No normalization was assumed since values like 3 (on a 5-point scale) were directly used as input, aligning with common scoring methods in operational logistics.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 17 |
Given that the weight for environmental risk is 0.061 and the local risk score for a route is 0.4, calculate the contribution of environmental risk to the overall risk score.
|
0.0244 |
|
Weight for Environmental Risk (W) = 0.061
• Local Risk Score (LRS) = 0.4
To find the contribution to the overall risk score, use the formula:
\text{Contribution} = \text{Weight} \times \text{Local Risk Score}
= 0.061 \times 0.4 = \boxed{0.0244}
|
This follows a weighted scoring method often used in risk modeling and decision support systems like FAHP or AHP–DEA models. Each criterion (like environmental risk) has a weight based on its importance, and the local score indicates the severity or likelihood in that scenario. Multiplying them shows the relative contribution of that factor to the overall risk.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 18 |
Calculate the new overall risk score if the weight of infrastructure risk is increased from 0.388 to 0.400 while keeping other parameters constant, given that its local risk score is 0.2.
|
0.080 |
|
New weight for infrastructure risk = 0.400
• Local risk score for infrastructure risk = 0.2
To calculate the contribution of this criterion to the overall risk score, use the formula:
\text{Risk Contribution} = \text{Weight} \times \text{Local Risk Score}
= 0.400 \times 0.2 = \boxed{0.080} |
This applies the principle of weighted risk scoring, where each criterion’s importance (weight) is combined with its measured severity or likelihood (local score). This approach is used in models like FAHP-DEA to understand the impact of parameter changes on the total risk.
In this case, we’re isolating the infrastructure risk component and analyzing how an adjustment in its weight influences its contribution to the overall risk score.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 19 |
If a mode of transportation has a risk weight of 0.073 and its risk score is reassessed from 0.4 to 0.35, what is the change in its contribution to the overall risk score?
|
0.00365 |
|
Weight (W) = 0.073
• Old risk score = 0.4
• New risk score = 0.35
We need to calculate the change in contribution to the overall risk score:
\text{Old Contribution} = 0.073 \times 0.4 = 0.0292
\text{New Contribution} = 0.073 \times 0.35 = 0.02555
\text{Change} = 0.0292 - 0.02555 = \boxed{0.00365}
|
This follows the method of differential weighted scoring, where changes in a component’s local risk score directly affect its contribution to the overall risk. This concept is common in multicriteria risk evaluation models such as FAHP or AHP–DEA. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 20 |
If the local weights of freight-damage risk, infrastructure risk, and operational risk are 0.1, 0.2, and 0.15 respectively, what is their total contribution to the risk score if their respective weights are 0.321, 0.388, and 0.157?
|
0.14647 |
|
Freight-Damage:
0.321 \times 0.1 = 0.0321
• Infrastructure:
0.388 \times 0.2 = 0.0776
• Operational:
0.157 \times 0.15 = 0.02355
Now, sum them:
\text{Total Contribution} = 0.0321 + 0.0776 + 0.02355 = \boxed{0.14647}
|
This is a weighted average calculation, often used in decision-making models (like AHP, DEA) where overall scores are derived from the product of each criterion’s weight and local performance. The sum gives a total impact on the model’s outcome. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|