Practice Exercise 1 (40 Questions)
Chapter 19: Bar Graph - Practice Exercise 1 featuring 40 high-yield, isomorphically transformed questions across 11 complex multidimensional problem sets. Covers grouped bar charts, stacked percentage distributions, min-max optimization, sequential stage processing pipelines, and multi-metric semiconductor evaluations.
Practice Exercise 1 — Modern Bar Graph Mastery
This exercise contains 40 comprehensive, first-principles verified CAT-style questions across 11 multidimensional problem sets:
-
Set 1 (Questions 1 to 4): Autonomous Drone Avionics Certification — Examining candidate pass/fail/re-evaluation pipelines, conditional bounds, absentees, and subsequent academic cycle requirements.
-
Set 2 (Questions 5 to 8): Global Cloud Datacenter Infrastructure — Server hardware failure trends across five sovereign cloud regions comparing GPU accelerators vs. CPU/memory host node faults across 2024 and 2025.
-
Set 3 (Questions 9 to 11): Autonomous AI Vehicle Benchmark Suite — Performance benchmark scores across 25 model checkpoints evaluated in three trials across six environmental stress conditions using min-max optimization.
-
Set 4 (Questions 12 to 15): Global Semiconductor Fabrication Hubs — Four-parameter performance evaluation covering wafer revenue per unit, peak cleanroom power demand, chemical waste generation, and critical safety incidents across four manufacturing clusters.
-
Set 5 (Questions 16 to 19): CleanTech EV Battery Manufacturing — Cross-jurisdiction evaluation of gigafactory export volumes and real Industrial Value-Add (IVA) growth rates.
-
Set 6 (Questions 20 to 23): Enterprise Generative AI Software Adoption — Two-dimensional 100% stacked bar chart analyzing deployment organization tiers and procurement channel distributions over a five-year horizon.
-
Set 7 (Questions 24 to 27): Global Quantum Technology Summit & Expo — Attendee growth dynamics (YoY percentage change relative to baseline) and industry vs. academic researcher composition.
-
Set 8 (Questions 28 to 30): Distributed Edge Pipeline Inspection Swarms — Four-stage sequential packet inspection benchmark across four autonomous defense swarms analyzing throughputs, stage catch-up, and relative positioning.
-
Set 9 (Questions 31 to 33): Frontier AI Research Institute Core Competencies — Set theory and integer optimization on multi-competency deployment profiles of research engineers.
-
Set 10 (Questions 34 to 36): Commercial Aerospace Launch Services — Longitudinal payload launch contract values across five years, base-year indexing, and percentage decline analysis.
-
Set 11 (Questions 37 to 40): Frontier AI & Robotics Research Fellows — Institutional fellow enrollments across Corporate Labs, State Universities, Specialized Institutes, and Frontier AI Centres across three academic cycles.
Practice questions
Question 1 · Foundation
Directions for Questions 1 to 4
Across an autonomous robotics and aerospace engineering institute, five specialized engineering cohorts—ECE, CSE, IT, ME, and EE—participated in a mandatory second-semester Avionics & Embedded Software Reliability Assessment.
The cohort sizes enrolled in each division are:
-
ECE, CSE, ME, and EE: 120 engineers each
-
IT: 80 engineers
(Total enrolled cohort strength across all five divisions = 560 engineers)
The bar chart below illustrates the initial assessment outcomes: the number of engineers who Passed (certified) and Failed (uncertified) in the first examination sitting, as well as the number of uncertified engineers who subsequently Applied for Re-evaluation (technical re-audit).
Key Assessment Rules:
-
Only engineers who initially failed the assessment were eligible to apply for re-evaluation.
-
Following re-evaluation, a portion of applicants secured passing status, while the remainder remained uncertified.
-
All engineers who remain uncertified after re-evaluation, along with all absentees from the initial sitting, are required to clear the examination in the subsequent academic cycle.
-
The initial passed and failed counts together represent the total number of candidates who appeared for the assessment in that division.
-
Any difference between the total enrolled cohort and the appeared number of candidates is due to absentees.
What was the overall pass percentage across all five engineering divisions combined prior to re-evaluation?
Show answer and explanation
Answer: (C) 67.9%
Step-by-Step Solution:
-
Calculate Total Batch Size Enrolled Across All Divisions:
-
-
-
-
-
-
.
-
-
Calculate Total Passed Candidates Before Re-evaluation:
-
-
-
-
-
-
.
-
-
Compute Overall Pass Percentage:
Hence, the correct option is 67.9%.
Question 2 · Foundation
Directions for Questions 1 to 4
Across an autonomous robotics and aerospace engineering institute, five specialized engineering cohorts—ECE, CSE, IT, ME, and EE—participated in a mandatory second-semester Avionics & Embedded Software Reliability Assessment.
The cohort sizes enrolled in each division are:
-
ECE, CSE, ME, and EE: 120 engineers each
-
IT: 80 engineers
(Total enrolled cohort strength across all five divisions = 560 engineers)
The bar chart below illustrates the initial assessment outcomes: the number of engineers who Passed (certified) and Failed (uncertified) in the first examination sitting, as well as the number of uncertified engineers who subsequently Applied for Re-evaluation (technical re-audit).
Key Assessment Rules:
-
Only engineers who initially failed the assessment were eligible to apply for re-evaluation.
-
Following re-evaluation, a portion of applicants secured passing status, while the remainder remained uncertified.
-
All engineers who remain uncertified after re-evaluation, along with all absentees from the initial sitting, are required to clear the examination in the subsequent academic cycle.
-
The initial passed and failed counts together represent the total number of candidates who appeared for the assessment in that division.
-
Any difference between the total enrolled cohort and the appeared number of candidates is due to absentees.
If less than 70% of the engineers who applied for re-evaluation pass after re-evaluation from each division, for which division was the ratio of passed engineers to failed engineers maximum after re-evaluation?
Show answer and explanation
Answer: (D) Cannot be determined
Step-by-Step Solution:
-
Determine the Upper Bound on Additional Passed Candidates (< 70%):
-
ECE: of at most 11 engineers pass.
-
CSE: of at most 8 engineers pass.
-
IT: of at most 12 engineers pass.
-
ME: of at most 18 engineers pass.
-
EE: of at most 12 engineers pass.
-
-
Scenario A (Maximum Additional Passes):
-
ECE: Passed , Failed .
-
CSE: Passed , Failed .
-
IT: Passed , Failed .
-
ME: Passed , Failed .
-
EE: Passed , Failed .
-
In this scenario, CSE achieves the highest ratio ().
-
-
Scenario B (Zero Additional Passes in CSE):
-
If no candidate passes re-evaluation in CSE (0 additional passes), CSE's ratio remains:
-
Meanwhile, if ME attains 18 additional passes, ME's ratio is . In this case, ME has the highest ratio.
-
Because the exact distribution of additional passes is not fixed, the division with the maximum ratio cannot be uniquely determined.
-
Hence, the correct option is Cannot be determined.
Question 3 · Foundation
Directions for Questions 1 to 4
Across an autonomous robotics and aerospace engineering institute, five specialized engineering cohorts—ECE, CSE, IT, ME, and EE—participated in a mandatory second-semester Avionics & Embedded Software Reliability Assessment.
The cohort sizes enrolled in each division are:
-
ECE, CSE, ME, and EE: 120 engineers each
-
IT: 80 engineers
(Total enrolled cohort strength across all five divisions = 560 engineers)
The bar chart below illustrates the initial assessment outcomes: the number of engineers who Passed (certified) and Failed (uncertified) in the first examination sitting, as well as the number of uncertified engineers who subsequently Applied for Re-evaluation (technical re-audit).
Key Assessment Rules:
-
Only engineers who initially failed the assessment were eligible to apply for re-evaluation.
-
Following re-evaluation, a portion of applicants secured passing status, while the remainder remained uncertified.
-
All engineers who remain uncertified after re-evaluation, along with all absentees from the initial sitting, are required to clear the examination in the subsequent academic cycle.
-
The initial passed and failed counts together represent the total number of candidates who appeared for the assessment in that division.
-
Any difference between the total enrolled cohort and the appeared number of candidates is due to absentees.
From each division, exactly 50% of the engineers who applied for re-evaluation failed even after re-evaluation. For which division is the percentage of failed engineers with respect to the total enrolled cohort minimum after re-evaluation?
Show answer and explanation
Answer: (B) CSE
Step-by-Step Solution:
If exactly of re-evaluation applicants failed, the other passed, reducing the failed tally by half the number of applicants.
-
Calculate Final Failed Engineers per Division:
-
ECE: Initial failed . Re-eval pass.
Final failed .
Percentage of total cohort () .
-
CSE: Initial failed . Re-eval pass.
Final failed .
Percentage of total cohort () .
-
IT: Initial failed . Re-eval pass.
Final failed .
Percentage of total cohort () .
-
ME: Initial failed . Re-eval pass.
Final failed .
Percentage of total cohort () .
-
EE: Initial failed . Re-eval pass.
Final failed .
Percentage of total cohort () .
-
-
Comparison:
The lowest percentage of failed engineers with respect to total enrolled cohort is in CSE ().
Hence, the correct option is CSE.
Question 4 · Foundation
Directions for Questions 1 to 4
Across an autonomous robotics and aerospace engineering institute, five specialized engineering cohorts—ECE, CSE, IT, ME, and EE—participated in a mandatory second-semester Avionics & Embedded Software Reliability Assessment.
The cohort sizes enrolled in each division are:
-
ECE, CSE, ME, and EE: 120 engineers each
-
IT: 80 engineers
(Total enrolled cohort strength across all five divisions = 560 engineers)
The bar chart below illustrates the initial assessment outcomes: the number of engineers who Passed (certified) and Failed (uncertified) in the first examination sitting, as well as the number of uncertified engineers who subsequently Applied for Re-evaluation (technical re-audit).
Key Assessment Rules:
-
Only engineers who initially failed the assessment were eligible to apply for re-evaluation.
-
Following re-evaluation, a portion of applicants secured passing status, while the remainder remained uncertified.
-
All engineers who remain uncertified after re-evaluation, along with all absentees from the initial sitting, are required to clear the examination in the subsequent academic cycle.
-
The initial passed and failed counts together represent the total number of candidates who appeared for the assessment in that division.
-
Any difference between the total enrolled cohort and the appeared number of candidates is due to absentees.
From each division, exactly 50% of the engineers who applied for re-evaluation failed even after re-evaluation. How many engineers from this batch will have to clear the certification next year?
Show answer and explanation
Answer: (B) 135
Step-by-Step Solution:
The rules specify: "All the failed engineers along with the absentees have to clear the examination next year."
-
Final Failed Candidates (from Question 3):
-
-
-
-
-
-
.
-
-
Absentees per Division:
-
Appeared Initial Passed Initial Failed
-
Absentees Total Enrolled Appeared:
-
ECE:
-
CSE:
-
IT:
-
ME:
-
EE:
-
-
.
-
-
Total Required to Clear Next Year:
Hence, the correct option is 135.
Question 5 · Foundation
Directions for Questions 5 to 8
The bar chart below depicts the hardware failure incidents (in thousands of units) recorded across five sovereign cloud computing regions—APAC-South, EMEA-North, Americas-East, East-Asia, and Americas-West—categorized by GPU Accelerators and CPU/Memory Host Nodes for calendar years 2024 and 2025.
(Note: Total hardware failures in any region equals the sum of GPU accelerator faults and CPU/memory host faults).
In how many cloud regions was the total number of hardware failures in 2025 less than that in 2024?
Show answer and explanation
Answer: (C) 3
Step-by-Step Solution:
Let us tabulate the total hardware failure incidents (in thousands) for each region:
| Cloud Region | 2024 GPU | 2024 CPU | 2024 Total | 2025 GPU | 2025 CPU | 2025 Total | Trend |
|---|---|---|---|---|---|---|---|
| APAC-South | 640 | 920 | 1,560 | 310 | 960 | 1,270 | Decreased |
| EMEA-North | 910 | 980 | 1,890 | 1,050 | 1,020 | 2,070 | Increased |
| Americas-East | 3,250 | 2,500 | 5,750 | 3,460 | 2,220 | 5,680 | Decreased |
| East-Asia | 4,750 | 4,040 | 8,790 | 4,610 | 4,240 | 8,850 | Increased |
| Americas-West | 1,160 | 1,260 | 2,420 | 1,270 | 580 | 1,850 | Decreased |
Total failures in 2025 were less than in 2024 in 3 regions: APAC-South, Americas-East, and Americas-West.
Hence, the correct option is 3.
Question 6 · Foundation
Directions for Questions 5 to 8
The bar chart below depicts the hardware failure incidents (in thousands of units) recorded across five sovereign cloud computing regions—APAC-South, EMEA-North, Americas-East, East-Asia, and Americas-West—categorized by GPU Accelerators and CPU/Memory Host Nodes for calendar years 2024 and 2025.
(Note: Total hardware failures in any region equals the sum of GPU accelerator faults and CPU/memory host faults).
By what percent was the total number of CPU/Memory host faults across the five cloud regions put together in 2025 less than that in 2024?
Show answer and explanation
Answer: (A) 7.01%
Step-by-Step Solution:
-
Calculate Total CPU/Memory Host Faults in 2024 (in '000 units):
-
Calculate Total CPU/Memory Host Faults in 2025 (in '000 units):
-
Compute Percentage Reduction:
Hence, the correct option is 7.01%.
Question 7 · Foundation
Directions for Questions 5 to 8
The bar chart below depicts the hardware failure incidents (in thousands of units) recorded across five sovereign cloud computing regions—APAC-South, EMEA-North, Americas-East, East-Asia, and Americas-West—categorized by GPU Accelerators and CPU/Memory Host Nodes for calendar years 2024 and 2025.
(Note: Total hardware failures in any region equals the sum of GPU accelerator faults and CPU/memory host faults).
What was the absolute difference between the total number of GPU accelerator faults across the five cloud regions put together in 2024 and 2025?
Show answer and explanation
Answer: (B) 10,000
Step-by-Step Solution:
-
Calculate Total GPU Accelerator Faults in 2024 (in '000 units):
-
Calculate Total GPU Accelerator Faults in 2025 (in '000 units):
-
Absolute Difference:
Hence, the correct option is 10,000.
Question 8 · Foundation
Directions for Questions 5 to 8
The bar chart below depicts the hardware failure incidents (in thousands of units) recorded across five sovereign cloud computing regions—APAC-South, EMEA-North, Americas-East, East-Asia, and Americas-West—categorized by GPU Accelerators and CPU/Memory Host Nodes for calendar years 2024 and 2025.
(Note: Total hardware failures in any region equals the sum of GPU accelerator faults and CPU/memory host faults).
For which cloud region was the percentage increase in the total number of hardware failures in 2025 as compared to 2024 the least?
Show answer and explanation
Answer: (A) East-Asia
Step-by-Step Solution:
-
Analyze Regions with an Increase in Total Hardware Failures:
-
APAC-South: (decrease)
-
Americas-East: (decrease)
-
Americas-West: (decrease)
-
EMEA-North: :
-
East-Asia: :
-
-
Conclusion:
Between EMEA-North and East-Asia, the smallest percentage increase was recorded in East-Asia ().
Hence, the correct option is East-Asia.
Question 9 · Foundation
Directions for Questions 9 to 11
The bar chart below exhibits the average benchmark score obtained by 25 autonomous driving AI model checkpoints across three rigorous evaluation trials—Trial-1, Trial-2, and Trial-3.
In each trial, six operational stress test conditions were conducted: Urban Dense, Highway Fog, Night Rain, Mountain Pass, Snow Blizzard, and Construction Zone.
Evaluation Standards:
-
The maximum possible score in any stress test condition across all trials was 100.
-
A model checkpoint was required to score at least 40% marks (i.e. at least 40 points) in a condition to qualify (pass). Scoring strictly less than 40 points constitutes a failure.
The number of model checkpoints that failed in all six stress conditions in Trial-1 cannot be more than:
Show answer and explanation
Answer: (B) 1
Step-by-Step Solution:
-
Thresholds & Constraints:
-
Maximum score .
-
Failing score .
-
Passing score .
-
Number of evaluated models .
-
-
Condition with Highest Average in Trial-1:
-
In Trial-1, Snow Blizzard has the highest average of 96.
-
Total points scored across all 25 models in Snow Blizzard .
-
-
Calculate Maximum Possible Failures in Snow Blizzard:
-
Let be the number of models that failed in Snow Blizzard (each scoring at most 39).
-
The remaining models can score at most 100 points each.
-
Total score .
-
Since total points equal 2,400:
-
Since must be an integer, .
-
-
Deduction:
-
For a model to fail in all six conditions in Trial-1, it must fail in Snow Blizzard.
-
Since at most 1 model could have failed in Snow Blizzard, at most 1 model could have failed across all conditions.
-
Hence, the correct option is 1.
Question 10 · Foundation
Directions for Questions 9 to 11
The bar chart below exhibits the average benchmark score obtained by 25 autonomous driving AI model checkpoints across three rigorous evaluation trials—Trial-1, Trial-2, and Trial-3.
In each trial, six operational stress test conditions were conducted: Urban Dense, Highway Fog, Night Rain, Mountain Pass, Snow Blizzard, and Construction Zone.
Evaluation Standards:
-
The maximum possible score in any stress test condition across all trials was 100.
-
A model checkpoint was required to score at least 40% marks (i.e. at least 40 points) in a condition to qualify (pass). Scoring strictly less than 40 points constitutes a failure.
If no model checkpoint failed in any of the stress conditions in Trial-2, at most how many models could have scored a perfect 100 in at least one condition in Trial-2?
Show answer and explanation
Answer: (D) 25
Step-by-Step Solution:
-
Given Conditions:
-
No model failed in Trial-2 all 25 models scored at least 40 points in every condition.
-
Total models .
-
-
Feasibility Analysis across Conditions:
-
Consider Highway Fog in Trial-2 (Average ):
-
Total points .
-
Suppose 18 models score 100 (total ).
-
The remaining 7 models score a total of points (average ).
-
Thus, 18 models can achieve 100 in Highway Fog.
-
-
Consider Snow Blizzard in Trial-2 (Average ):
-
Total points .
-
Here, the remaining 7 models who did not score 100 in Highway Fog can easily score 100 in Snow Blizzard (requiring points, leaving points distributed among the other 18 models, averaging ).
-
-
-
Conclusion:
- Across the six independent stress conditions, every single one of the 25 models could have scored 100 in at least one condition.
Hence, the correct option is 25.
Question 11 · Foundation
Directions for Questions 9 to 11
The bar chart below exhibits the average benchmark score obtained by 25 autonomous driving AI model checkpoints across three rigorous evaluation trials—Trial-1, Trial-2, and Trial-3.
In each trial, six operational stress test conditions were conducted: Urban Dense, Highway Fog, Night Rain, Mountain Pass, Snow Blizzard, and Construction Zone.
Evaluation Standards:
-
The maximum possible score in any stress test condition across all trials was 100.
-
A model checkpoint was required to score at least 40% marks (i.e. at least 40 points) in a condition to qualify (pass). Scoring strictly less than 40 points constitutes a failure.
The number of model checkpoints that scored more than 90 marks in all the stress conditions in each of the trials cannot be more than:
Show answer and explanation
Answer: (B) 17
Step-by-Step Solution:
-
Bottleneck Condition Analysis:
-
Scoring points in all conditions across every trial requires scoring at least 91 points in every single condition.
-
The number of such models cannot exceed the maximum number of models scoring in the condition with the lowest average score across the entire dataset.
-
-
Locating the Lowest Average:
-
Looking across all three trials, the lowest average is Night Rain in Trial-1 with an average score of 62.
-
Total score for 25 models in Night Rain (Trial-1) .
-
-
Maximizing Checkpoints Scoring :
-
Let be the number of models scoring at least 91.
-
The remaining models score at least 0 points.
-
Total score .
-
Therefore:
-
Thus, at most 17 models could have scored in Night Rain (Trial-1), which caps the count across all conditions to 17.
-
Hence, the correct option is 17.
Question 12 · Foundation
Directions for Questions 12 to 15
The multi-panel visualization below displays operational and economic performance metrics for four premier global semiconductor manufacturing clusters—Hsinchu, Dresden, Austin, and Phoenix—for fiscal years 2024 and 2025.
The evaluated parameters include:
-
Wafer Revenue per Unit ($/unit)
-
Peak Cleanroom Power Demand (MW)
-
Ultrapure Chemical Waste Generated (kL)
-
Critical Safety Incidents (Severity-1 Alerts)
If the year-on-year percentage increase in wafer revenue per unit is denoted by IPC and the year-on-year percentage increase in peak power demand is denoted by IPP, for how many hubs was (IPC – IPP) negative over the period 2024 to 2025?
Show answer and explanation
Answer: (D) 3
Step-by-Step Solution:
Let us compute and for each fabrication hub between 2024 and 2025:
-
Hsinchu:
-
Wafer Revenue:
-
Peak Power Demand:
-
(Negative)
-
-
Dresden:
-
Wafer Revenue:
-
Peak Power Demand:
-
(Negative)
-
-
Austin:
-
Wafer Revenue:
-
Peak Power Demand:
-
(Negative)
-
-
Phoenix:
-
Wafer Revenue:
-
Peak Power Demand:
-
(Positive)
-
was negative for 3 hubs: Hsinchu, Dresden, and Austin.
Hence, the correct option is 3.
Question 13 · Moderate
Directions for Questions 12 to 15
The multi-panel visualization below displays operational and economic performance metrics for four premier global semiconductor manufacturing clusters—Hsinchu, Dresden, Austin, and Phoenix—for fiscal years 2024 and 2025.
The evaluated parameters include:
-
Wafer Revenue per Unit ($/unit)
-
Peak Cleanroom Power Demand (MW)
-
Ultrapure Chemical Waste Generated (kL)
-
Critical Safety Incidents (Severity-1 Alerts)
It is known that chemical waste generated per cleanroom worker increased in 2025 by x% over 2024, with values: Hsinchu: x = 4; Dresden: x = 3; Austin: x = 5; Phoenix: x = 4. Which hub witnessed the maximum percentage increase in its cleanroom workforce in 2025 over 2024?
Show answer and explanation
Answer: (A) Hsinchu
Step-by-Step Solution:
-
Mathematical Model:
where is total chemical waste growth rate and is workforce growth rate.
-
Evaluate for Each Hub:
-
Hsinchu:
-
Waste: ()
-
-
-
Dresden:
-
Waste: ()
-
-
-
Austin:
-
Waste: ()
-
-
-
Phoenix:
-
Waste: ()
-
-
-
-
Comparison:
The maximum percentage increase in cleanroom workforce occurred in Hsinchu ().
Hence, the correct option is Hsinchu.
Question 14 · Moderate
Directions for Questions 12 to 15
The multi-panel visualization below displays operational and economic performance metrics for four premier global semiconductor manufacturing clusters—Hsinchu, Dresden, Austin, and Phoenix—for fiscal years 2024 and 2025.
The evaluated parameters include:
-
Wafer Revenue per Unit ($/unit)
-
Peak Cleanroom Power Demand (MW)
-
Ultrapure Chemical Waste Generated (kL)
-
Critical Safety Incidents (Severity-1 Alerts)
Which semiconductor fabrication hub witnessed the maximum percentage increase in its workforce in 2025 over that in 2024 based only on the primary dataset?
Show answer and explanation
Answer: (D) Cannot be determined
Step-by-Step Solution:
-
The primary dataset provides information strictly on:
-
Wafer revenue per unit
-
Peak power demand
-
Chemical waste generated
-
Critical safety incidents
-
-
Cleanroom workforce numbers, total headcount, or per-capita worker rates are not given in the main passage (the supplementary parameter in Question 13 applied exclusively to Question 13).
-
Without baseline workforce or per-worker metrics, the percentage increase in total workforce cannot be determined from the graphs.
Hence, the correct option is Cannot be determined.
Question 15 · Moderate
Directions for Questions 12 to 15
The multi-panel visualization below displays operational and economic performance metrics for four premier global semiconductor manufacturing clusters—Hsinchu, Dresden, Austin, and Phoenix—for fiscal years 2024 and 2025.
The evaluated parameters include:
-
Wafer Revenue per Unit ($/unit)
-
Peak Cleanroom Power Demand (MW)
-
Ultrapure Chemical Waste Generated (kL)
-
Critical Safety Incidents (Severity-1 Alerts)
Over the given period (2024 and 2025 combined), the total number of critical safety incidents in Phoenix and Dresden put together was what percent of the total critical safety incidents in Austin and Hsinchu put together?
Show answer and explanation
Answer: (A) 65.22%
Step-by-Step Solution:
-
Calculate Incidents for Phoenix and Dresden (2024 + 2025):
-
Phoenix:
-
Dresden:
-
Combined Phoenix + Dresden .
-
-
Calculate Incidents for Austin and Hsinchu (2024 + 2025):
-
Austin:
-
Hsinchu:
-
Combined Austin + Hsinchu .
-
-
Compute Percentage:
Hence, the correct option is 65.22%.
Question 16 · Moderate
Directions for Questions 16 to 19
The two charts below illustrate the CleanTech EV Battery Pack Exports (in billion dollars) from 2021 to 2025, and the Industrial Value-Add (IVA) Annual Growth Rate over the previous year from 2022 to 2025 across four major manufacturing jurisdictions: Sino-Tech, Indo-Pac, ASEAN-South, and Siam-Clean.
If the Industrial Value-Add (IVA) of Indo-Pac in 2021 was 220 billion, what was the approximate difference between Sino-Tech’s IVA and Indo-Pac’s IVA in 2022?
Show answer and explanation
Answer: (B) $166.5 billion
Step-by-Step Solution:
-
Identify 2022 Growth Rates:
-
Sino-Tech IVA growth rate in 2022
-
Indo-Pac IVA growth rate in 2022
-
-
Calculate 2022 IVA Values:
-
-
Difference:
Hence, the correct option is $166.5 billion.
Question 17 · Moderate
Directions for Questions 16 to 19
The two charts below illustrate the CleanTech EV Battery Pack Exports (in billion dollars) from 2021 to 2025, and the Industrial Value-Add (IVA) Annual Growth Rate over the previous year from 2022 to 2025 across four major manufacturing jurisdictions: Sino-Tech, Indo-Pac, ASEAN-South, and Siam-Clean.
If the IVA of ASEAN-South in 2021 was $60 billion, then the ratio of Battery Exports to IVA of ASEAN-South in 2022 with respect to that in 2021:
Show answer and explanation
Answer: (A) increased
Step-by-Step Solution:
-
Calculate Ratio in 2021:
-
Battery Exports in 2021
-
IVA in 2021
-
-
-
Calculate Ratio in 2022:
-
Battery Exports in 2022
-
2022 IVA Growth Rate
-
IVA in 2022
-
-
-
Comparison:
Since , the ratio increased.
Hence, the correct option is increased.
Question 18 · Moderate
Directions for Questions 16 to 19
The two charts below illustrate the CleanTech EV Battery Pack Exports (in billion dollars) from 2021 to 2025, and the Industrial Value-Add (IVA) Annual Growth Rate over the previous year from 2022 to 2025 across four major manufacturing jurisdictions: Sino-Tech, Indo-Pac, ASEAN-South, and Siam-Clean.
In which of the following years did Siam-Clean have the least ratio of Battery Exports to IVA? (Given that IVA is always greater than exports)
Show answer and explanation
Answer: (A) 2021
Step-by-Step Solution:
Let be the baseline IVA of Siam-Clean in 2021.
-
Exports Values for Siam-Clean:
-
2021:
-
2022:
-
2025:
-
-
Express IVA in Terms of :
-
2021:
-
2022:
-
2025:
-
-
Compare Exports to IVA Ratios:
-
2021:
-
2022:
-
2025:
-
The ratio was lowest in 2021.
Hence, the correct option is 2021.
Question 19 · Moderate
Directions for Questions 16 to 19
The two charts below illustrate the CleanTech EV Battery Pack Exports (in billion dollars) from 2021 to 2025, and the Industrial Value-Add (IVA) Annual Growth Rate over the previous year from 2022 to 2025 across four major manufacturing jurisdictions: Sino-Tech, Indo-Pac, ASEAN-South, and Siam-Clean.
If the IVA of ASEAN-South in 2021 was ) in 2024.
Show answer and explanation
Answer: (A) 102.37
Step-by-Step Solution:
-
IVA Annual Growth Rates for ASEAN-South:
-
2022:
-
2023:
-
2024:
-
-
Calculate 2024 IVA:
Hence, the correct option is 102.37.
Question 20 · Moderate
Directions for Questions 20 to 23
The charts below display the adoption pattern of enterprise Generative AI software solutions in a premier technology metropolis between 2021 and 2025.
The first chart illustrates the breakdown of deploying organizations across three company size tiers:
-
Tier A: Seed & Early Startup ( seats)
-
Tier B: Mid-Market ( to seats)
-
Tier C: Large Enterprise ( seats)
The second chart illustrates the distribution of contract acquisitions across four procurement channels: Direct Sales, Cloud Marketplace, System Integrators, and Self-Hosted Managed Solutions.
In 2021, if there were 2,000 enterprise deploying accounts in the metropolis and the number of deploying accounts increased by 200 every year during the given period, what was the total number of deploying accounts from Tier C during 2021 to 2025?
Show answer and explanation
Answer: (C) 2,046
Step-by-Step Solution:
-
Calculate Total Deploying Accounts per Year:
-
2021: 2,000 accounts
-
2022: 2,200 accounts
-
2023: 2,400 accounts
-
2024: 2,600 accounts
-
2025: 2,800 accounts
-
-
Calculate Tier C Deploying Accounts per Year:
-
2021: of
-
2022: of
-
2023: of
-
2024: of
-
2025: of
-
-
Sum Over the Five-Year Period:
Hence, the correct option is 2,046.
Question 21 · Moderate
Directions for Questions 20 to 23
The charts below display the adoption pattern of enterprise Generative AI software solutions in a premier technology metropolis between 2021 and 2025.
The first chart illustrates the breakdown of deploying organizations across three company size tiers:
-
Tier A: Seed & Early Startup ( seats)
-
Tier B: Mid-Market ( to seats)
-
Tier C: Large Enterprise ( seats)
The second chart illustrates the distribution of contract acquisitions across four procurement channels: Direct Sales, Cloud Marketplace, System Integrators, and Self-Hosted Managed Solutions.
By what percentage did the difference between the number of deploying accounts acquired via Self-Hosted solutions and those acquired via Cloud Marketplace change between 2021 and 2025? (Assume total deploying accounts remained constant for each year during this period).
Show answer and explanation
Answer: (C) Increased by 118.18%
Step-by-Step Solution:
Let the constant total annual accounts be .
-
Gap in 2021:
-
Cloud Marketplace
-
Self-Hosted
-
Difference of
-
-
Gap in 2025:
-
Cloud Marketplace
-
Self-Hosted
-
Difference of
-
-
Percentage Change in Difference:
The difference increased by 118.18%.
Hence, the correct option is Increased by 118.18%.
Question 22 · Moderate
Directions for Questions 20 to 23
The charts below display the adoption pattern of enterprise Generative AI software solutions in a premier technology metropolis between 2021 and 2025.
The first chart illustrates the breakdown of deploying organizations across three company size tiers:
-
Tier A: Seed & Early Startup ( seats)
-
Tier B: Mid-Market ( to seats)
-
Tier C: Large Enterprise ( seats)
The second chart illustrates the distribution of contract acquisitions across four procurement channels: Direct Sales, Cloud Marketplace, System Integrators, and Self-Hosted Managed Solutions.
Which procurement channel managed to retain the number of its customer accounts at the exact same level for a period of three consecutive years between 2021 and 2025?
Show answer and explanation
Answer: (D) Data Insufficient
Step-by-Step Solution:
-
The provided charts present only the percentage distribution of procurement channels each year.
-
The total count of deploying accounts across the metropolis in general is not specified in the main passage (the assumption of 2,000 accounts was specific only to certain questions).
-
Even where percentage shares remain constant over two years (e.g. System Integrators at 18% in 2023, 2024, and 2025), constant percentage does not imply constant absolute volume unless total metropolitan accounts are proven constant across those three years.
-
Hence, the data is insufficient to establish that absolute customer accounts remained unchanged.
Hence, the correct option is Data Insufficient.
Question 23 · Moderate
Directions for Questions 20 to 23
The charts below display the adoption pattern of enterprise Generative AI software solutions in a premier technology metropolis between 2021 and 2025.
The first chart illustrates the breakdown of deploying organizations across three company size tiers:
-
Tier A: Seed & Early Startup ( seats)
-
Tier B: Mid-Market ( to seats)
-
Tier C: Large Enterprise ( seats)
The second chart illustrates the distribution of contract acquisitions across four procurement channels: Direct Sales, Cloud Marketplace, System Integrators, and Self-Hosted Managed Solutions.
In every organization tier, if the distribution across procurement channels matches the overall total, approximately what was the number of deploying accounts in Tier B procuring via Direct Sales in 2021? (Assume 2,000 total accounts in 2021).
Show answer and explanation
Answer: (B) 585
Step-by-Step Solution:
-
Calculate Tier B Accounts in 2021:
-
Total accounts
-
Tier B proportion in
-
Tier B accounts
-
-
Calculate Direct Sales Procurement:
-
Direct Sales proportion across all tiers in
-
Assuming channel distribution is independent of tier:
-
Hence, the correct option is 585.
Question 24 · Moderate
Directions for Questions 24 to 27
The visualization below outlines attendance data for the annual Global Quantum Technology Summit & Expo from 2021 to 2025.
The left panel displays the Year-on-Year (YoY) Percentage Change in Total Attendees relative to the preceding year (with 2021 serving as the baseline). The right panel shows the Industry Practitioner Share (%) of total attendees each year (the remaining attendees being Academic & National Lab Researchers).
If the total number of summit attendees in 2025 was 64,000, what was the number of academic researcher attendees in 2025?
Show answer and explanation
Answer: (B) 22,400
Step-by-Step Solution:
-
Identify Attendee Distribution in 2025:
-
Total attendees in
-
Industry Practitioner share
-
Academic Researcher share
-
-
Calculate Academic Researcher Count:
Hence, the correct option is 22,400.
Question 25 · Moderate
Directions for Questions 24 to 27
The visualization below outlines attendance data for the annual Global Quantum Technology Summit & Expo from 2021 to 2025.
The left panel displays the Year-on-Year (YoY) Percentage Change in Total Attendees relative to the preceding year (with 2021 serving as the baseline). The right panel shows the Industry Practitioner Share (%) of total attendees each year (the remaining attendees being Academic & National Lab Researchers).
Which of the following statements is definitely true?
Show answer and explanation
Answer: (C) There was a growth of 50% in the number of academic researcher attendees from 2023 to 2024.
Step-by-Step Solution:
Let total attendees in 2021 be :
-
2021: Total . Industry () , Academic () .
-
2022: Change Total . Industry () , Academic () .
-
2023: Change Total . Industry () , Academic () .
-
2024: Change Total . Industry () , Academic () .
-
2025: Change Total . Industry () , Academic () .
Evaluate the Statements:
-
Option 3: Academic attendees grew from in 2023 to in 2024:
This statement is definitely true!
-
Option 1: Academic attendees also decreased from to in 2025 (False).
-
Option 2: Industry attendees grew from to : (False).
-
Option 4: Industry attendees decreased from to in 2025 (False).
Hence, the correct option is There was a growth of 50% in the number of academic researcher attendees from 2023 to 2024.
Question 26 · Moderate
Directions for Questions 24 to 27
The visualization below outlines attendance data for the annual Global Quantum Technology Summit & Expo from 2021 to 2025.
The left panel displays the Year-on-Year (YoY) Percentage Change in Total Attendees relative to the preceding year (with 2021 serving as the baseline). The right panel shows the Industry Practitioner Share (%) of total attendees each year (the remaining attendees being Academic & National Lab Researchers).
The difference between the number of industry practitioner attendees and academic researcher attendees in a year is called the “Domain Gap”. For which of the following pairs of years is the Domain Gap equal?
Show answer and explanation
Answer: (D) 2023 and 2025
Step-by-Step Solution:
Using the attendee counts derived in Question 25:
-
2021:
-
2022:
-
2023:
-
2024:
-
2025:
The Domain Gap equals in both 2023 and 2025.
Hence, the correct option is 2023 and 2025.
Question 27 · Moderate
Directions for Questions 24 to 27
The visualization below outlines attendance data for the annual Global Quantum Technology Summit & Expo from 2021 to 2025.
The left panel displays the Year-on-Year (YoY) Percentage Change in Total Attendees relative to the preceding year (with 2021 serving as the baseline). The right panel shows the Industry Practitioner Share (%) of total attendees each year (the remaining attendees being Academic & National Lab Researchers).
In which year was the number of industry practitioner attendees the maximum?
Show answer and explanation
Answer: (B) 2024
Step-by-Step Solution:
Comparing the industry practitioner attendees across the five years:
-
2021:
-
2022:
-
2023:
-
2024:
-
2025:
The maximum number of industry practitioner attendees was achieved in 2024 with .
Hence, the correct option is 2024.
Question 28 · Moderate
Directions for Questions 28 to 30
Four autonomous cyber-defense sentinel swarms—Aegis, Valkyrie, Cerberus, and Sentinel—competed in an edge network security verification benchmark consisting of four sequential packet inspection stages of 600 Megabits each.
In the benchmark, the Stage 1 inspection agents from each swarm start processing simultaneously from the initial telemetry buffer. As soon as the Stage 1 agent completes verification of its 600 Megabits, the Stage 2 agent from that swarm begins immediately from the pipeline buffer, and this continues sequentially through Stage 4.
The swarm with the least composite total time across all four stages wins. The bar graph below depicts the constant processing throughput (in Mbps) recorded for each stage agent across the four swarms.
Which autonomous defense swarm completed the four-stage verification benchmark in the least composite time (winner)?
Show answer and explanation
Answer: (A) Aegis Swarm
Step-by-Step Solution:
Each stage comprises 600 Megabits. Total processing time for a swarm equals:
-
Aegis Swarm:
-
Valkyrie Swarm:
-
Cerberus Swarm:
-
Sentinel Swarm:
Aegis Swarm recorded the lowest composite processing time ( seconds) and won the benchmark.
Hence, the correct option is Aegis Swarm.
Question 29 · CAT Level
Directions for Questions 28 to 30
Four autonomous cyber-defense sentinel swarms—Aegis, Valkyrie, Cerberus, and Sentinel—competed in an edge network security verification benchmark consisting of four sequential packet inspection stages of 600 Megabits each.
In the benchmark, the Stage 1 inspection agents from each swarm start processing simultaneously from the initial telemetry buffer. As soon as the Stage 1 agent completes verification of its 600 Megabits, the Stage 2 agent from that swarm begins immediately from the pipeline buffer, and this continues sequentially through Stage 4.
The swarm with the least composite total time across all four stages wins. The bar graph below depicts the constant processing throughput (in Mbps) recorded for each stage agent across the four swarms.
What was the shortest distance (in Megabits processed) from the start of Stage 2 at which the Stage 2 process of Aegis Swarm overtook the Stage 2 process of Valkyrie Swarm?
Show answer and explanation
Answer: (D) Stage 2 processes of Aegis Swarm and Valkyrie Swarm did not meet
Step-by-Step Solution:
-
Stage 1 Completion (Stage 2 Start Times):
-
Valkyrie completes Stage 1 at .
-
Aegis completes Stage 1 at .
-
-
Head Start for Valkyrie's Stage 2 Agent:
-
Valkyrie starts Stage 2 at s.
-
When Aegis starts Stage 2 at s, Valkyrie has been running Stage 2 for .
-
Distance verified by Valkyrie in that time .
-
-
Pursuit & Catch-Up Analysis:
-
Throughput of Aegis Stage 2 agent .
-
Relative throughput .
-
Time required for Aegis to close the 120 Megabit deficit:
-
Megabits that Aegis would need to process in that time:
-
However, Stage 2 consists of only 600 Megabits. Aegis finishes Stage 2 at s, well before catching Valkyrie.
-
Therefore, the Stage 2 processes of Aegis Swarm and Valkyrie Swarm did not meet.
Hence, the correct option is Stage 2 processes of Aegis Swarm and Valkyrie Swarm did not meet.
Question 30 · CAT Level
Directions for Questions 28 to 30
Four autonomous cyber-defense sentinel swarms—Aegis, Valkyrie, Cerberus, and Sentinel—competed in an edge network security verification benchmark consisting of four sequential packet inspection stages of 600 Megabits each.
In the benchmark, the Stage 1 inspection agents from each swarm start processing simultaneously from the initial telemetry buffer. As soon as the Stage 1 agent completes verification of its 600 Megabits, the Stage 2 agent from that swarm begins immediately from the pipeline buffer, and this continues sequentially through Stage 4.
The swarm with the least composite total time across all four stages wins. The bar graph below depicts the constant processing throughput (in Mbps) recorded for each stage agent across the four swarms.
Which of the following statements is definitely true?
Show answer and explanation
Answer: (B) Stage 3 processes of Aegis Swarm and Valkyrie Swarm definitely met once during the execution.
Step-by-Step Solution:
Let us examine the Stage 3 processes of Aegis Swarm and Valkyrie Swarm:
-
Stage 3 Start Times:
-
Valkyrie: Stage 1 takes 40 s, Stage 2 takes 50 s Starts Stage 3 at .
-
Aegis: Stage 1 takes 50 s, Stage 2 takes s Starts Stage 3 at .
-
-
Lead and Catch-Up Dynamics in Stage 3:
-
Valkyrie starts Stage 3 at s with speed .
-
By the time Aegis starts Stage 3 at s, Valkyrie has processed:
-
Aegis processes Stage 3 at speed .
-
Relative speed of Aegis .
-
Time required for Aegis to overtake Valkyrie:
-
Distance processed by Aegis during catch-up:
-
Since , Aegis definitely catches up with and crosses Valkyrie inside Stage 3.
-
Therefore, statement (2) is definitely true.
Hence, the correct option is Stage 3 processes of Aegis Swarm and Valkyrie Swarm definitely met once during the execution.
Question 31 · CAT Level
Directions for Questions 31 to 33
The research engineers at a frontier artificial intelligence laboratory work across five specialized technical core competencies: LLM Pre-training, Reinforcement Learning, Computer Vision, Robotics Manipulation, and Neuro-symbolic Reasoning, such that every engineer is deployed in at least two of the five mentioned competencies.
The bar graph below provides the number of research engineers active in each of the five technical competencies.
Find the absolute difference between the maximum and minimum possible number of research engineers in the laboratory.
Show answer and explanation
Answer: (C) 38
Step-by-Step Solution:
Let denote the number of engineers active in exactly 2, 3, 4, and 5 competencies respectively, and let be the total number of engineers.
-
Calculate Total Competency Deployments ():
-
Maximizing Engineers ():
- To maximize , minimize the competencies per engineer by having everyone work on exactly 2 competencies ().
-
Minimizing Engineers ():
-
The total number of engineers cannot be strictly less than the count working on any single competency, which is 54 (Robotics Manipulation). Thus .
-
Feasibility check for :
-
With 20 engineers working on all 5 competencies ().
-
Remaining deployments needed: LLM: 10, RL: 22, Vision: 18, Robotics: 34.
-
Pairwise assignments easily cover the remaining 34 engineers while keeping everyone in competencies.
-
Hence, .
-
-
-
Absolute Difference:
Hence, the correct option is 38.
Question 32 · CAT Level
Directions for Questions 31 to 33
The research engineers at a frontier artificial intelligence laboratory work across five specialized technical core competencies: LLM Pre-training, Reinforcement Learning, Computer Vision, Robotics Manipulation, and Neuro-symbolic Reasoning, such that every engineer is deployed in at least two of the five mentioned competencies.
The bar graph below provides the number of research engineers active in each of the five technical competencies.
The number of research engineers in the laboratory is the minimum possible. Find the maximum possible number of engineers active in exactly two different competencies.
Show answer and explanation
Answer: (B) 26
Step-by-Step Solution:
Given :
-
...(i)
-
...(ii)
Multiply (i) by 3 and subtract (ii):
Also multiply (i) by 2 and subtract from (ii):
-
Maximizing :
-
To maximize , we must maximize .
-
Maximum possible (since Neuro-symbolic Reasoning has only 20 engineers).
-
Substituting into the expression for :
-
Since , we have .
-
Setting (which yields ):
-
Hence, the correct option is 26.
Question 33 · CAT Level
Directions for Questions 31 to 33
The research engineers at a frontier artificial intelligence laboratory work across five specialized technical core competencies: LLM Pre-training, Reinforcement Learning, Computer Vision, Robotics Manipulation, and Neuro-symbolic Reasoning, such that every engineer is deployed in at least two of the five mentioned competencies.
The bar graph below provides the number of research engineers active in each of the five technical competencies.
The number of research engineers in the laboratory is the minimum possible. Find the maximum possible number of engineers active in exactly three different competencies.
Show answer and explanation
Answer: (B) 43
Step-by-Step Solution:
From Question 32:
-
Constraints for Non-Negativity:
-
Since , we require .
-
To maximize , we must minimize .
-
Rewriting: .
-
To minimize with :
-
If , then .
-
Taking and :
-
-
-
Verification:
-
.
-
Total engineers .
-
Total competency deployments .
-
All conditions hold, confirming .
-
Hence, the correct option is 43.
Question 34 · CAT Level
Directions for Questions 34 to 36
The bar chart below illustrates the annual cumulative contract value (in million dollars) secured by a commercial aerospace enterprise for Orbital Satellite Launch Services across five consecutive calendar years: 2020 through 2024.
What is the average value of launch service contracts secured during the five years shown in the diagram?
Show answer and explanation
Answer: (A) $124.06 Million
Step-by-Step Solution:
-
Sum Contract Values across All Five Years ($ Millions):
-
Compute Average:
Hence, the correct option is $124.06 Million.
Question 35 · CAT Level
Directions for Questions 34 to 36
The bar chart below illustrates the annual cumulative contract value (in million dollars) secured by a commercial aerospace enterprise for Orbital Satellite Launch Services across five consecutive calendar years: 2020 through 2024.
Compared to the performance in 2021 (i.e. taking it as the base of 100%), what can you say about the performances in the years 2020, 2021, 2022, 2023, and 2024 respectively, in percentage terms?
Show answer and explanation
Answer: (A) 150, 100, 210, 215, 97
Step-by-Step Solution:
Baseline year is 2021 ().
-
2020:
-
2021:
-
2022:
-
2023:
-
2024:
The resulting index sequence is 150, 100, 210, 215, 97.
Hence, the correct option is 150, 100, 210, 215, 97.
Question 36 · CAT Level
Directions for Questions 34 to 36
The bar chart below illustrates the annual cumulative contract value (in million dollars) secured by a commercial aerospace enterprise for Orbital Satellite Launch Services across five consecutive calendar years: 2020 through 2024.
Which is the year in which the highest percentage decline is seen in the value of contracts secured compared to the preceding year?
Show answer and explanation
Answer: (B) 2024
Step-by-Step Solution:
Declines compared to the preceding year occurred only in 2021 and 2024:
-
2021 (compared to 2020):
-
2024 (compared to 2023):
The highest percentage decline was recorded in 2024 ( drop).
Hence, the correct option is 2024.
Question 37 · CAT Level
Directions for Questions 37 to 40
The bar graph below illustrates the number of advanced research fellows (in hundreds) active across four distinct categories of research institutions—Corporate Labs, State Academic Universities, Specialized Tech Institutes, and Frontier AI Centres—over three consecutive academic cycles: 2022–23, 2023–24, and 2024–25.
What was the total number of research fellows across all four institutional categories in 2023–24?
Show answer and explanation
Answer: (C) 51,500
Step-by-Step Solution:
-
Fellow Counts in 2023–24 (in hundreds):
-
Corporate Labs: 225
-
State Academic Universities: 150
-
Specialized Tech Institutes: 90
-
Frontier AI Centres: 50
-
-
Sum in Hundreds:
-
Convert to Total Number:
Hence, the correct option is 51,500.
Question 38 · CAT Level
Directions for Questions 37 to 40
The bar graph below illustrates the number of advanced research fellows (in hundreds) active across four distinct categories of research institutions—Corporate Labs, State Academic Universities, Specialized Tech Institutes, and Frontier AI Centres—over three consecutive academic cycles: 2022–23, 2023–24, and 2024–25.
The growth rate in fellows at State Academic Universities compared to that at Corporate Labs between 2022–23 and 2023–24 is:
Show answer and explanation
Answer: (C) equal
Step-by-Step Solution:
-
Growth Rate in Corporate Labs (2022–23 to 2023–24):
-
Growth Rate in State Academic Universities (2022–23 to 2023–24):
Both categories registered an identical growth rate of 50%. Therefore, the growth rate is equal.
Hence, the correct option is equal.
Question 39 · CAT Level
Directions for Questions 37 to 40
The bar graph below illustrates the number of advanced research fellows (in hundreds) active across four distinct categories of research institutions—Corporate Labs, State Academic Universities, Specialized Tech Institutes, and Frontier AI Centres—over three consecutive academic cycles: 2022–23, 2023–24, and 2024–25.
The total number of research fellows in 2025–26, assuming a 10% reduction in the total number over the previous year, is:
Show answer and explanation
Answer: (D) None of these
Step-by-Step Solution:
-
Calculate Total Fellows in 2024–25 (in hundreds):
-
Corporate Labs: 320
-
State Academic Universities: 160
-
Specialized Tech Institutes: 130
-
Frontier AI Centres: 100
-
-
Apply 10% Reduction for 2025–26:
-
Check Options:
Since 63,900 is not among the explicit numerical options (65,000, 71,000, 58,000), the answer is None of these.
Hence, the correct option is None of these.
Question 40 · CAT Level
Directions for Questions 37 to 40
The bar graph below illustrates the number of advanced research fellows (in hundreds) active across four distinct categories of research institutions—Corporate Labs, State Academic Universities, Specialized Tech Institutes, and Frontier AI Centres—over three consecutive academic cycles: 2022–23, 2023–24, and 2024–25.
In 2024–25, what percent of research fellows were studying at Frontier AI Centres?
Show answer and explanation
Answer: (C) 14%
Step-by-Step Solution:
-
Data for 2024–25:
-
Frontier AI Centres
-
Total Fellows across all four categories
-
-
Compute Percentage:
Hence, the correct option is 14%.
Try real CAT questions
Study this concept with interactive practice in PrepSoc.
Open the concept library