Free CAT library

Practice Exercise 2 (68 Questions)

68 advanced CAT-level caselet questions covering multi-variable constraints, 4-set Venn systems, and authentic examination sets.

Practice Exercise 2 — Advanced & Past CAT Sets

69 high-difficulty caselet questions.

Practice questions

Question 1 · Moderate

Directions for Questions 1 to 4

Answer the questions on the basis of the information given below.

In an automated site reliability engineering (SRE) incident triage module called 'ResolveOps', system alerts and incident tickets are logged across four infrastructure domains: Network (NET), Storage (STO), Database (DB), and Security (SEC). Tickets enter the system triage queue immediately when logged and remain active until marked resolved. All tickets logged during the first half of a work day are counted as part of that day's intake, whereas tickets logged in the second half are queued as incoming tickets for the following day. Four lead engineers — Dave, Alice, Steve, and Tom — are responsible for ticket resolution. Tickets are processed and resolved exclusively during the second half of the day.

The following table provides the number of unresolved pending tickets carried forward from the previous week (15th to 21st May) across the four domains at the close of that week:

Unresolved Pending Tickets of Previous Week
Network (NET)Storage (STO)Database (DB)Security (SEC)
24174814

The volume of new incoming incident tickets received during the current week (22nd to 28th May) is detailed below:

DayNetwork (NET)Storage (STO)Database (DB)Security (SEC)
Monday10252
Tuesday12523
Wednesday8250
Thursday4285
Friday2436
Saturday4517

Operational Rules and Work Constraints:

  • Security (SEC) tickets are handled exclusively by Steve.

  • Tom resolves tickets based on Database (DB) only. Steve resolves tickets on Security (SEC) only.

  • In the current week, if the daily number of tickets resolved by Dave, Alice, Steve, and Tom on any day (Monday through Saturday) are denoted by dd, aa, ss, and tt respectively, the engineer throughput bounds were observed to be:

    • 6≤d≤126 \le d \le 12

    • 8≤a≤188 \le a \le 18

    • 5≤s≤105 \le s \le 10

    • 2≤t≤152 \le t \le 15

  • No ticket is admitted or resolved on Sunday.

On which day of the current week could Steve resolve the maximum possible number of tickets?

  1. A

    Wednesday

  2. B

    Friday

  3. C

    Saturday

  4. D

    Either Friday or Saturday

Show answer and explanation

Answer: (D) Either Friday or Saturday

Question 1 Solution:

Steve resolves Security (SEC) tickets exclusively, and his daily resolution bound is 5≤s≤105 \le s \le 10. To resolve his maximum possible quota (s=10s = 10) on any day, there must be at least 10 tickets available in the queue (backlog + new intake).

Steve must also resolve at least the minimum of 5 tickets on each day from Monday to Thursday. Across Monday through Thursday, the total tickets available are 14 (previous week backlog) + (2 + 3 + 0 + 5) = 24 tickets.

  • If Steve resolves the minimum of 5 tickets each day on Monday, Tuesday, Wednesday, and Thursday, he resolves 5×4=205 \times 4 = 20 tickets in total.

  • This leaves only 24−20=424 - 20 = 4 tickets pending by Thursday evening.

  • Thus, at no point from Monday to Thursday could Steve have accumulated 10 tickets to resolve.

On Friday, 6 new tickets arrive, making 4+6=104 + 6 = 10 tickets available. Hence, Steve could resolve 10 tickets on Friday.

Alternatively, he could resolve 5 tickets on Friday (leaving 5 pending) and resolve 10 on Saturday (with 7 new tickets arriving, 5+7=12≥105 + 7 = 12 \ge 10).

Therefore, Steve could resolve the maximum number of queries (10) on either Friday or Saturday.

Question 2 · Moderate

Directions for Questions 1 to 4

Answer the questions on the basis of the information given below.

In an automated site reliability engineering (SRE) incident triage module called 'ResolveOps', system alerts and incident tickets are logged across four infrastructure domains: Network (NET), Storage (STO), Database (DB), and Security (SEC). Tickets enter the system triage queue immediately when logged and remain active until marked resolved. All tickets logged during the first half of a work day are counted as part of that day's intake, whereas tickets logged in the second half are queued as incoming tickets for the following day. Four lead engineers — Dave, Alice, Steve, and Tom — are responsible for ticket resolution. Tickets are processed and resolved exclusively during the second half of the day.

The following table provides the number of unresolved pending tickets carried forward from the previous week (15th to 21st May) across the four domains at the close of that week:

Unresolved Pending Tickets of Previous Week
Network (NET)Storage (STO)Database (DB)Security (SEC)
24174814

The volume of new incoming incident tickets received during the current week (22nd to 28th May) is detailed below:

DayNetwork (NET)Storage (STO)Database (DB)Security (SEC)
Monday10252
Tuesday12523
Wednesday8250
Thursday4285
Friday2436
Saturday4517

Operational Rules and Work Constraints:

  • Security (SEC) tickets are handled exclusively by Steve.

  • Tom resolves tickets based on Database (DB) only. Steve resolves tickets on Security (SEC) only.

  • In the current week, if the daily number of tickets resolved by Dave, Alice, Steve, and Tom on any day (Monday through Saturday) are denoted by dd, aa, ss, and tt respectively, the engineer throughput bounds were observed to be:

    • 6≤d≤126 \le d \le 12

    • 8≤a≤188 \le a \le 18

    • 5≤s≤105 \le s \le 10

    • 2≤t≤152 \le t \le 15

  • No ticket is admitted or resolved on Sunday.

In the current week, if there were no pending Database (DB) tickets at the end of Saturday and only Tom resolved DB tickets, the total number of DB tickets that were resolved on the same day when they were received could not be less than:

  1. A

    4

  2. B

    3

  3. C

    2

  4. D

    1

Show answer and explanation

Answer: (D) 1

Question 2 Solution:

Tom resolves DB tickets with daily capacity 2≤t≤152 \le t \le 15. At the start of Monday, there are 48 pending DB tickets from the previous week.

To minimize the number of tickets resolved on the day they arrive, Tom should prioritize resolving older pending backlog tickets first up to his maximum capacity of 15 tickets per day:

DayPending from Prior DaysNew Tickets ReceivedTickets ResolvedRemaining End of Day
Monday48515(48 − 15) + 5 = 38
Tuesday38215(38 − 15) + 2 = 25
Wednesday25515(25 − 15) + 5 = 15
Thursday15815(15 − 15) + 8 = 8
Friday838(8 − 8) + 3 = 3
Saturday3143 + 1 − 4 = 0

On Saturday, Tom starts with 3 pending tickets and receives 1 new ticket. To ensure 0 pending tickets at the end of Saturday, Tom must resolve all 4 tickets, which includes the 1 ticket received on Saturday.

Thus, the total number of DB tickets resolved on the same day they were received cannot be less than 1.

Question 3 · Moderate

Directions for Questions 1 to 4

Answer the questions on the basis of the information given below.

In an automated site reliability engineering (SRE) incident triage module called 'ResolveOps', system alerts and incident tickets are logged across four infrastructure domains: Network (NET), Storage (STO), Database (DB), and Security (SEC). Tickets enter the system triage queue immediately when logged and remain active until marked resolved. All tickets logged during the first half of a work day are counted as part of that day's intake, whereas tickets logged in the second half are queued as incoming tickets for the following day. Four lead engineers — Dave, Alice, Steve, and Tom — are responsible for ticket resolution. Tickets are processed and resolved exclusively during the second half of the day.

The following table provides the number of unresolved pending tickets carried forward from the previous week (15th to 21st May) across the four domains at the close of that week:

Unresolved Pending Tickets of Previous Week
Network (NET)Storage (STO)Database (DB)Security (SEC)
24174814

The volume of new incoming incident tickets received during the current week (22nd to 28th May) is detailed below:

DayNetwork (NET)Storage (STO)Database (DB)Security (SEC)
Monday10252
Tuesday12523
Wednesday8250
Thursday4285
Friday2436
Saturday4517

Operational Rules and Work Constraints:

  • Security (SEC) tickets are handled exclusively by Steve.

  • Tom resolves tickets based on Database (DB) only. Steve resolves tickets on Security (SEC) only.

  • In the current week, if the daily number of tickets resolved by Dave, Alice, Steve, and Tom on any day (Monday through Saturday) are denoted by dd, aa, ss, and tt respectively, the engineer throughput bounds were observed to be:

    • 6≤d≤126 \le d \le 12

    • 8≤a≤188 \le a \le 18

    • 5≤s≤105 \le s \le 10

    • 2≤t≤152 \le t \le 15

  • No ticket is admitted or resolved on Sunday.

Which of the following is not possible?

  1. A

    All pending tickets of the previous week could be resolved by the end of Wednesday

  2. B

    All pending tickets of the previous week could be resolved by the end of Thursday

  3. C

    All pending tickets of the previous week could be resolved by the end of Friday

  4. D

    All pending tickets of the previous week could be resolved by the end of Monday

Show answer and explanation

Answer: (D) All pending tickets of the previous week could be resolved by the end of Monday

Question 3 Solution:

Total pending unresolved tickets from the previous week across all four domains:

Total backlog=24+17+48+14=103\text{Total backlog} = 24 + 17 + 48 + 14 = 103

The maximum possible combined daily throughput of all four engineers is:

Max single-day capacity=dmax⁡+amax⁡+smax⁡+tmax⁡=12+18+10+15=55\text{Max single-day capacity} = d_{\max} + a_{\max} + s_{\max} + t_{\max} = 12 + 18 + 10 + 15 = 55

Because 55<10355 < 103, it is physically impossible to clear all 103 pending tickets in a single day (Monday). At least two days are strictly required.

Hence, the statement that all pending tickets of the previous week could be resolved by the end of Monday is impossible.

Question 4 · Moderate

Directions for Questions 1 to 4

Answer the questions on the basis of the information given below.

In an automated site reliability engineering (SRE) incident triage module called 'ResolveOps', system alerts and incident tickets are logged across four infrastructure domains: Network (NET), Storage (STO), Database (DB), and Security (SEC). Tickets enter the system triage queue immediately when logged and remain active until marked resolved. All tickets logged during the first half of a work day are counted as part of that day's intake, whereas tickets logged in the second half are queued as incoming tickets for the following day. Four lead engineers — Dave, Alice, Steve, and Tom — are responsible for ticket resolution. Tickets are processed and resolved exclusively during the second half of the day.

The following table provides the number of unresolved pending tickets carried forward from the previous week (15th to 21st May) across the four domains at the close of that week:

Unresolved Pending Tickets of Previous Week
Network (NET)Storage (STO)Database (DB)Security (SEC)
24174814

The volume of new incoming incident tickets received during the current week (22nd to 28th May) is detailed below:

DayNetwork (NET)Storage (STO)Database (DB)Security (SEC)
Monday10252
Tuesday12523
Wednesday8250
Thursday4285
Friday2436
Saturday4517

Operational Rules and Work Constraints:

  • Security (SEC) tickets are handled exclusively by Steve.

  • Tom resolves tickets based on Database (DB) only. Steve resolves tickets on Security (SEC) only.

  • In the current week, if the daily number of tickets resolved by Dave, Alice, Steve, and Tom on any day (Monday through Saturday) are denoted by dd, aa, ss, and tt respectively, the engineer throughput bounds were observed to be:

    • 6≤d≤126 \le d \le 12

    • 8≤a≤188 \le a \le 18

    • 5≤s≤105 \le s \le 10

    • 2≤t≤152 \le t \le 15

  • No ticket is admitted or resolved on Sunday.

On Monday of the current week, a total of 58 tickets were reported resolved. At most how many of them could be Database (DB) tickets?

  1. A

    45

  2. B

    31

  3. C

    21

  4. D

    Data inconsistent

Show answer and explanation

Answer: (D) Data inconsistent

Question 4 Solution:

The maximum daily ticket resolution capacity for the four engineers is:

  • Dave: d≤12d \le 12

  • Alice: a≤18a \le 18

  • Steve: s≤10s \le 10

  • Tom: t≤15t \le 15

Maximum possible tickets resolved on any day=12+18+10+15=55\text{Maximum possible tickets resolved on any day} = 12 + 18 + 10 + 15 = 55

Since 58 exceeds the maximum possible total capacity of 55, the reported data is contradictory.

Therefore, the correct choice is Data inconsistent.

Question 5 · Moderate

Directions for Questions 5 to 8

Answer the questions on the basis of the information given below.

An enterprise cloud architecture consists of 9 interconnected microservice nodes labelled Node-A, Node-B, Node-C, Node-D, Node-E, Node-F, Node-G, Node-H, and Node-I arranged in the network grid shown below:

Node-A
Node-C Node-GNode-HNode-I Node-D
Node-ENode-F
  
Node-B

Each node is assigned a distinct single-digit natural number priority rating from the set {1,2,3,4,5,6,7,8,9}\{1, 2, 3, 4, 5, 6, 7, 8, 9\}. Two nodes sharing a direct communication boundary are called adjacent nodes.

  • Node-A and Node-B span the entire width of the topology and are both adjacent to each of Node-C, Node-E, Node-G, Node-H, Node-I, Node-F, and Node-D.

  • Horizontally, Node-C connects to Node-E; Node-E connects to Node-G; Node-G connects to Node-H; Node-H connects to Node-I; Node-I connects to Node-F; and Node-F connects to Node-D.

Additional System Constraints:

  1. The sum of the priority ratings on Node-G and Node-I is less than 13.

  2. The sum of the priority ratings on nodes adjacent to Node-H is equal to the sum of the priority ratings on nodes adjacent to Node-G.

  3. The sum of the priority ratings on the nodes adjacent to the node assigned rating 5 is 30.

  4. The sum of the priority ratings on Node-C and Node-D is less than 6.

  5. Node-E does not have the lowest priority rating (1), and Node-F does not have the highest priority rating (9).

  6. The sum of the priority ratings on Node-G and Node-H is a prime number.

  7. The sum of the priority ratings on the nodes adjacent to Node-D exceeds the sum of the priority ratings on the nodes adjacent to Node-C by 6.

Which of the following statements is sufficient to find the priority rating assigned to Node-C?

  1. A

    The question can be answered by using statement (i) alone but not (ii) alone

  2. B

    The question can be answered by using statement (ii) alone but not (i) alone

  3. C

    The question can be answered by using either statement alone

  4. D

    No more information is required

Show answer and explanation

Answer: (C) The question can be answered by using either statement alone

Question 5 Solution:

Step 1: Network Adjacency Analysis

  • Adjacent to Node-D: Node-A, Node-B, Node-F.

  • Adjacent to Node-C: Node-A, Node-B, Node-E.

From Constraint 7:

Sum(adj D)−Sum(adj C)=(A+B+F)−(A+B+E)=F−E=6\text{Sum}(\text{adj } D) - \text{Sum}(\text{adj } C) = (A + B + F) - (A + B + E) = F - E = 6

Since F,E∈{1,2,…,9}F, E \in \{1, 2, \dots, 9\}, the possible pairs (E,F)(E, F) are (1,7)(1, 7), (2,8)(2, 8), or (3,9)(3, 9).

From Constraint 5, E≠1E \ne 1 (not lowest) and F≠9F \ne 9 (not highest). Therefore:

E=2andF=8E = 2 \quad \text{and} \quad F = 8

Step 2: Locating Rating 5

From Constraint 3, the sum of nodes adjacent to the node with rating 5 is 30. The node with rating 5 must have 4 neighbors, whose ratings can sum to 30. The nodes adjacent to Node-I are Node-H, Node-F, Node-A, Node-B.

Sum(adj I)=H+F+A+B=H+8+A+B=30  ⟹  H+A+B=22\text{Sum}(\text{adj } I) = H + F + A + B = H + 8 + A + B = 30 \implies H + A + B = 22

Since AA and BB are distinct and must sum with HH to 22, the only unused high single-digit ratings are {6,7,9}\{6, 7, 9\}. Thus, H=7H = 7 and {A,B}={6,9}\{A, B\} = \{6, 9\}, confirming that Node-I has rating 5 (I=5I = 5).

Step 3: Finding Node-G and Node-H

From Constraint 2, Sum(adj H)=Sum(adj G)\text{Sum}(\text{adj } H) = \text{Sum}(\text{adj } G).

Sum(adj H)=G+I+A+B=G+5+15=G+20\text{Sum}(\text{adj } H) = G + I + A + B = G + 5 + 15 = G + 20

Sum(adj G)=E+H+A+B=2+7+15=24\text{Sum}(\text{adj } G) = E + H + A + B = 2 + 7 + 15 = 24

G+20=24  ⟹  G=4G + 20 = 24 \implies G = 4

Checking Constraint 6: G+H=4+7=11G + H = 4 + 7 = 11, which is prime.

Checking Constraint 1: G+I=4+5=9<13G + I = 4 + 5 = 9 < 13.

Step 4: Evaluating Node-C and Node-D

The remaining unused ratings are {1,3}\{1, 3\}, so {C,D}={1,3}\{C, D\} = \{1, 3\}. Both satisfy Constraint 4 (C+D=4<6C + D = 4 < 6).

Now, test the statements in the question:

  • Sum of ratings adjacent to Node-F: adj F=I+D+A+B=5+D+15=20+D\text{adj } F = I + D + A + B = 5 + D + 15 = 20 + D.

  • Sum of ratings adjacent to Node-E: adj E=C+G+A+B=C+4+15=19+C\text{adj } E = C + G + A + B = C + 4 + 15 = 19 + C.

  • Statement (i): 20+D>19+C  ⟺  1+D>C20 + D > 19 + C \iff 1 + D > C.

    • If D=3D = 3 and C=1C = 1: 1+3=4>11 + 3 = 4 > 1 (holds).

    • If D=1D = 1 and C=3C = 3: 1+1=2>31 + 1 = 2 > 3 (false).

    Hence, Statement (i) uniquely determines C=1C = 1.

  • Statement (ii): 20+D<19+C  ⟺  1+D<C20 + D < 19 + C \iff 1 + D < C.

    • If D=1D = 1 and C=3C = 3: 1+1=2<31 + 1 = 2 < 3 (holds).

    • If D=3D = 3 and C=1C = 1: 1+3=4<11 + 3 = 4 < 1 (false).

    Hence, Statement (ii) uniquely determines C=3C = 3.

Therefore, the question can be answered by using either statement alone.

Question 6 · Moderate

Directions for Questions 5 to 8

Answer the questions on the basis of the information given below.

An enterprise cloud architecture consists of 9 interconnected microservice nodes labelled Node-A, Node-B, Node-C, Node-D, Node-E, Node-F, Node-G, Node-H, and Node-I arranged in the network grid shown below:

Node-A
Node-C Node-GNode-HNode-I Node-D
Node-ENode-F
  
Node-B

Each node is assigned a distinct single-digit natural number priority rating from the set {1,2,3,4,5,6,7,8,9}\{1, 2, 3, 4, 5, 6, 7, 8, 9\}. Two nodes sharing a direct communication boundary are called adjacent nodes.

  • Node-A and Node-B span the entire width of the topology and are both adjacent to each of Node-C, Node-E, Node-G, Node-H, Node-I, Node-F, and Node-D.

  • Horizontally, Node-C connects to Node-E; Node-E connects to Node-G; Node-G connects to Node-H; Node-H connects to Node-I; Node-I connects to Node-F; and Node-F connects to Node-D.

Additional System Constraints:

  1. The sum of the priority ratings on Node-G and Node-I is less than 13.

  2. The sum of the priority ratings on nodes adjacent to Node-H is equal to the sum of the priority ratings on nodes adjacent to Node-G.

  3. The sum of the priority ratings on the nodes adjacent to the node assigned rating 5 is 30.

  4. The sum of the priority ratings on Node-C and Node-D is less than 6.

  5. Node-E does not have the lowest priority rating (1), and Node-F does not have the highest priority rating (9).

  6. The sum of the priority ratings on Node-G and Node-H is a prime number.

  7. The sum of the priority ratings on the nodes adjacent to Node-D exceeds the sum of the priority ratings on the nodes adjacent to Node-C by 6.

For how many nodes can the priority ratings assigned to them be uniquely determined?

  1. A

    All of them

  2. B

    7

  3. C

    3

  4. D

    5

Show answer and explanation

Answer: (D) 5

Question 6 Solution:

From the deduced ratings:

  • Node-E = 2 (uniquely determined)

  • Node-F = 8 (uniquely determined)

  • Node-G = 4 (uniquely determined)

  • Node-H = 7 (uniquely determined)

  • Node-I = 5 (uniquely determined)

The remaining nodes have dual possibilities:

  • {Node−A,Node−B}={6,9}\{Node-A, Node-B\} = \{6, 9\} (either A=6,B=9A=6, B=9 or A=9,B=6A=9, B=6)

  • {Node−C,Node−D}={1,3}\{Node-C, Node-D\} = \{1, 3\} (either C=1,D=3C=1, D=3 or C=3,D=1C=3, D=1)

Thus, exactly 5 nodes (E, F, G, H, I) have uniquely determined priority ratings.

Question 7 · Moderate

Directions for Questions 5 to 8

Answer the questions on the basis of the information given below.

An enterprise cloud architecture consists of 9 interconnected microservice nodes labelled Node-A, Node-B, Node-C, Node-D, Node-E, Node-F, Node-G, Node-H, and Node-I arranged in the network grid shown below:

Node-A
Node-C Node-GNode-HNode-I Node-D
Node-ENode-F
  
Node-B

Each node is assigned a distinct single-digit natural number priority rating from the set {1,2,3,4,5,6,7,8,9}\{1, 2, 3, 4, 5, 6, 7, 8, 9\}. Two nodes sharing a direct communication boundary are called adjacent nodes.

  • Node-A and Node-B span the entire width of the topology and are both adjacent to each of Node-C, Node-E, Node-G, Node-H, Node-I, Node-F, and Node-D.

  • Horizontally, Node-C connects to Node-E; Node-E connects to Node-G; Node-G connects to Node-H; Node-H connects to Node-I; Node-I connects to Node-F; and Node-F connects to Node-D.

Additional System Constraints:

  1. The sum of the priority ratings on Node-G and Node-I is less than 13.

  2. The sum of the priority ratings on nodes adjacent to Node-H is equal to the sum of the priority ratings on nodes adjacent to Node-G.

  3. The sum of the priority ratings on the nodes adjacent to the node assigned rating 5 is 30.

  4. The sum of the priority ratings on Node-C and Node-D is less than 6.

  5. Node-E does not have the lowest priority rating (1), and Node-F does not have the highest priority rating (9).

  6. The sum of the priority ratings on Node-G and Node-H is a prime number.

  7. The sum of the priority ratings on the nodes adjacent to Node-D exceeds the sum of the priority ratings on the nodes adjacent to Node-C by 6.

For how many pairs of adjacent nodes is the sum of their priority ratings equal to 15?

  1. A

    0

  2. B

    1

  3. C

    2

  4. D

    3

Show answer and explanation

Answer: (A) 0

Question 7 Solution:

Let us check all possible adjacent pairs:

  1. Pairs involving Node-A or Node-B:

    • Node-A and Node-B are not adjacent to each other.

    • Node-A has rating 6 or 9. For the sum with an adjacent node to be 15:

      • If A=6A = 6, neighbor must be 9 (but Node-B has rating 9 and is not adjacent to Node-A).

      • If A=9A = 9, neighbor must be 6 (but Node-B has rating 6 and is not adjacent to Node-A).

    • The exact same logic applies to Node-B.

  2. Horizontal adjacent pairs among the interior nodes:

    • (C,E)(C, E): 1+2=31 + 2 = 3 or 3+2=53 + 2 = 5

    • (E,G)(E, G): 2+4=62 + 4 = 6

    • (G,H)(G, H): 4+7=114 + 7 = 11

    • (H,I)(H, I): 7+5=127 + 5 = 12

    • (I,F)(I, F): 5+8=135 + 8 = 13

    • (F,D)(F, D): 8+1=98 + 1 = 9 or 8+3=118 + 3 = 11

None of the adjacent pairs sum to 15. Hence, there are 0 such pairs.

Question 8 · Moderate

Directions for Questions 9 to 12

Answer the questions on the basis of the information given below.

Upon noticing sudden variations in server cluster availability, a cloud infrastructure reliability engineer analyzed the operational uptime data of four high-performance computing (HPC) server clusters — Cluster Alpha, Cluster Beta, Cluster Gamma, and Cluster Delta. The audit covered the number of full days on which each cluster was fully operational during a standard 30-day billing cycle.

The infrastructure lead did not provide direct individual uptime metrics. Instead, he took two of the four clusters at a time, summed their operational days, and sorted the six resulting pairwise totals in descending order from left to right in a summary table. During data export, a database corruption rendered the figures in columns 3, 4, and 5 illegible:

Column123456
Pairwise Operational Days (taken two at a time)5149[Corrupted][Corrupted][Corrupted]40

For each of the four clusters, the number of downtime days during the 30-day month was a distinct integer. Furthermore, there was no day in the month on which all four clusters were down simultaneously.

Column figure numbered 5 should be:

  1. A

    41

  2. B

    42

  3. C

    44

  4. D

    45

Show answer and explanation

Answer: (B) 42

Question 9 Solution:

Let a,b,c,da, b, c, d denote the operational days of the four clusters, arranged in descending order (a>b>c>da > b > c > d).

The pairwise sums taken two at a time yield 6 values:

  • Largest sum: a+b=51a + b = 51 (Column 1)

  • Second largest sum: a+c=49a + c = 49 (Column 2)

  • Smallest sum: c+d=40c + d = 40 (Column 6)

The sum of all six pairwise totals is equal to 3(a+b+c+d)3(a + b + c + d):

(a+b)+(c+d)=51+40=91  ⟹  a+b+c+d=91(a + b) + (c + d) = 51 + 40 = 91 \implies a + b + c + d = 91

Notice that:

(a+c)+(b+d)=a+b+c+d=91(a + c) + (b + d) = a + b + c + d = 91

49+(b+d)=91  ⟹  b+d=4249 + (b + d) = 91 \implies b + d = 42

Since a>b>c>da > b > c > d, the relative ordering of the six sums is:

a+b>a+c>{a+d,  b+c}>b+d>c+da + b > a + c > \{a + d, \; b + c\} > b + d > c + d

Because b+db + d is strictly the second-smallest pairwise sum, it must occupy Column 5.

Therefore, Column 5 must be 42.

Question 9 · Moderate

Directions for Questions 9 to 12

Answer the questions on the basis of the information given below.

Upon noticing sudden variations in server cluster availability, a cloud infrastructure reliability engineer analyzed the operational uptime data of four high-performance computing (HPC) server clusters — Cluster Alpha, Cluster Beta, Cluster Gamma, and Cluster Delta. The audit covered the number of full days on which each cluster was fully operational during a standard 30-day billing cycle.

The infrastructure lead did not provide direct individual uptime metrics. Instead, he took two of the four clusters at a time, summed their operational days, and sorted the six resulting pairwise totals in descending order from left to right in a summary table. During data export, a database corruption rendered the figures in columns 3, 4, and 5 illegible:

Column123456
Pairwise Operational Days (taken two at a time)5149[Corrupted][Corrupted][Corrupted]40

For each of the four clusters, the number of downtime days during the 30-day month was a distinct integer. Furthermore, there was no day in the month on which all four clusters were down simultaneously.

If one of the illegible column figures reads 45, then which of the following is true?

  1. A

    One of the other illegible column figures reads 46

  2. B

    One of the four clusters was operational on exactly 25 days in the month

  3. C

    One of the four clusters was operational on exactly 23 days in the month

  4. D

    One of the four clusters was operational on exactly 21 days in the month

Show answer and explanation

Answer: (A) One of the other illegible column figures reads 46

Question 10 Solution:

From the analysis in Question 9, the six pairwise sums are sorted in descending order:

Col 1=51,Col 2=49,Col 3,Col 4,Col 5=42,Col 6=40\text{Col 1} = 51, \quad \text{Col 2} = 49, \quad \text{Col 3}, \quad \text{Col 4}, \quad \text{Col 5} = 42, \quad \text{Col 6} = 40

The missing Column 3 and Column 4 are the values of a+da + d and b+cb + c in some order. We know:

(a+d)+(b+c)=a+b+c+d=91(a + d) + (b + c) = a + b + c + d = 91

If one of the illegible figures (either Column 3 or Column 4) is 45, then the other illegible figure must be:

91−45=4691 - 45 = 46

Thus, one of the other illegible column figures must read 46.

Question 10 · Moderate

Directions for Questions 9 to 12

Answer the questions on the basis of the information given below.

Upon noticing sudden variations in server cluster availability, a cloud infrastructure reliability engineer analyzed the operational uptime data of four high-performance computing (HPC) server clusters — Cluster Alpha, Cluster Beta, Cluster Gamma, and Cluster Delta. The audit covered the number of full days on which each cluster was fully operational during a standard 30-day billing cycle.

The infrastructure lead did not provide direct individual uptime metrics. Instead, he took two of the four clusters at a time, summed their operational days, and sorted the six resulting pairwise totals in descending order from left to right in a summary table. During data export, a database corruption rendered the figures in columns 3, 4, and 5 illegible:

Column123456
Pairwise Operational Days (taken two at a time)5149[Corrupted][Corrupted][Corrupted]40

For each of the four clusters, the number of downtime days during the 30-day month was a distinct integer. Furthermore, there was no day in the month on which all four clusters were down simultaneously.

Cluster Gamma was the second cluster, if the four clusters are arranged in ascending order of the days on which they were operational in the month. What could have been the maximum possible number of days on which Cluster Gamma was operational?

  1. A

    25

  2. B

    24

  3. C

    23

  4. D

    22

Show answer and explanation

Answer: (C) 23

Question 11 Solution:

In ascending order of operational days, the clusters are ordered as d<c<b<ad < c < b < a. Thus, the second cluster in ascending order is Cluster Gamma, with cc operational days.

From the system of equations:

  • a+b=51  ⟹  a≥26,  b≤25a + b = 51 \implies a \ge 26, \; b \le 25

  • c+d=40  ⟹  c≥21,  d≤19c + d = 40 \implies c \ge 21, \; d \le 19

  • a+c=49  ⟹  a=49−ca + c = 49 \implies a = 49 - c

  • b+d=42  ⟹  d=42−bb + d = 42 \implies d = 42 - b

  • b=51−a=51−(49−c)=c+2b = 51 - a = 51 - (49 - c) = c + 2

  • d=40−cd = 40 - c

Checking values for cc such that a>b>c>da > b > c > d:

  • If c=21c = 21: a=28,b=23,d=19a = 28, b = 23, d = 19. (28>23>21>1928 > 23 > 21 > 19, valid)

  • If c=22c = 22: a=27,b=24,d=18a = 27, b = 24, d = 18. (27>24>22>1827 > 24 > 22 > 18, valid)

  • If c=23c = 23: a=26,b=25,d=17a = 26, b = 25, d = 17. (26>25>23>1726 > 25 > 23 > 17, valid)

  • If c=24c = 24: a=25,b=26,d=16a = 25, b = 26, d = 16. (Here a<ba < b, which violates a>ba > b, invalid)

Thus, the maximum possible value of cc is 23.

Question 11 · Moderate

Directions for Questions 9 to 12

Answer the questions on the basis of the information given below.

Upon noticing sudden variations in server cluster availability, a cloud infrastructure reliability engineer analyzed the operational uptime data of four high-performance computing (HPC) server clusters — Cluster Alpha, Cluster Beta, Cluster Gamma, and Cluster Delta. The audit covered the number of full days on which each cluster was fully operational during a standard 30-day billing cycle.

The infrastructure lead did not provide direct individual uptime metrics. Instead, he took two of the four clusters at a time, summed their operational days, and sorted the six resulting pairwise totals in descending order from left to right in a summary table. During data export, a database corruption rendered the figures in columns 3, 4, and 5 illegible:

Column123456
Pairwise Operational Days (taken two at a time)5149[Corrupted][Corrupted][Corrupted]40

For each of the four clusters, the number of downtime days during the 30-day month was a distinct integer. Furthermore, there was no day in the month on which all four clusters were down simultaneously.

When the reliability engineer consulted the lead administrator regarding the missing data, the administrator recalled that the figure in Column 3 was the sum of operational days for the clusters with the maximum and the minimum operational days. Which of the following is the figure in Column 3?

  1. A

    44

  2. B

    45

  3. C

    46

  4. D

    47

Show answer and explanation

Answer: (D) 47

Question 12 Solution:

The cluster with the maximum operational days is Alpha (aa) and the one with the minimum is Delta (dd). Their sum is a+da + d.

Since Column 3 is the third largest sum, having a+da + d in Column 3 means that a+d>b+ca + d > b + c:

  • Case I (c=21c = 21): a=28,b=23,c=21,d=19a = 28, b = 23, c = 21, d = 19.

    a+d=28+19=47,b+c=23+21=44a + d = 28 + 19 = 47, \quad b + c = 23 + 21 = 44

    Here, a+d=47>44=b+ca + d = 47 > 44 = b + c. Thus, Column 3 is 47 and Column 4 is 44.

  • Case II (c=22c = 22): a=27,b=24,c=22,d=18a = 27, b = 24, c = 22, d = 18.

    a+d=27+18=45,b+c=24+22=46a + d = 27 + 18 = 45, \quad b + c = 24 + 22 = 46

    Here, b+c>a+db + c > a + d, so Column 3 would be b+c=46b + c = 46, not a+da + d.

  • Case III (c=23c = 23): a=26,b=25,c=23,d=17a = 26, b = 25, c = 23, d = 17.

    a+d=43,b+c=48a + d = 43, \quad b + c = 48

    Here, Column 3 would be 48, not a+da + d.

Hence, Column 3 being a+da + d occurs strictly in Case I, where a+d=47a + d = 47.

Therefore, Column 3 is 47.

Question 12 · Moderate

Directions for Questions 13 to 15

Answer the questions on the basis of the information given below.

A medical technology company, Apex Bio Diagnostics, manufactures six specialized diagnostic test kits: Kit-A, Kit-B, Kit-C, Kit-D, Kit-E, and Kit-F. The table below shows the per-unit Cost Price (CP) and Selling Price (SP) in dollars ($) for each of the six kits in the fiscal year 2012:

Diagnostic KitCost Price per unit ($)Selling Price per unit ($)Per-Unit Profit / Loss ($)
Kit-A3525−10 (Loss)
Kit-B3040+10 (Profit)
Kit-C5535−20 (Loss)
Kit-D4555+10 (Profit)
Kit-E3015−15 (Loss)
Kit-F7035−35 (Loss)

The company sold all units of the six kits that it produced in the given year.

If the company produced Kit-A, Kit-B, Kit-C, Kit-D, Kit-E, and Kit-F in the ratio 7 : 31 : 6 : 23 : 4 : 3, then the overall percentage profit that the company made in the year 2012 was:

  1. A

    6.35%

  2. B

    6.63%

  3. C

    6.45%

  4. D

    8.21%

Show answer and explanation

Answer: (C) 6.45%

Question 13 Solution:

Let the number of units produced of Kit-A, Kit-B, Kit-C, Kit-D, Kit-E, and Kit-F be 7x7x, 31x31x, 6x6x, 23x23x, 4x4x, and 3x3x respectively.

Total Net Profit / Loss:

Total Profit=7x(−10)+31x(10)+6x(−20)+23x(10)+4x(−15)+3x(−35)\text{Total Profit} = 7x(-10) + 31x(10) + 6x(-20) + 23x(10) + 4x(-15) + 3x(-35)

=−70x+310x−120x+230x−60x−105x=185x= -70x + 310x - 120x + 230x - 60x - 105x = 185x

Total Cost Price (Investment):

Total CP=7x(35)+31x(30)+6x(55)+23x(45)+4x(30)+3x(70)\text{Total CP} = 7x(35) + 31x(30) + 6x(55) + 23x(45) + 4x(30) + 3x(70)

=245x+930x+330x+1035x+120x+210x=2870x= 245x + 930x + 330x + 1035x + 120x + 210x = 2870x

Overall Profit Percentage:

Profit %=185x2870x×100=1850287≈6.44598%≈6.45%\text{Profit } \% = \frac{185x}{2870x} \times 100 = \frac{1850}{287} \approx 6.44598\% \approx 6.45\%

Therefore, the overall profit percentage was 6.45%.

Question 13 · Moderate

Directions for Questions 13 to 15

Answer the questions on the basis of the information given below.

A medical technology company, Apex Bio Diagnostics, manufactures six specialized diagnostic test kits: Kit-A, Kit-B, Kit-C, Kit-D, Kit-E, and Kit-F. The table below shows the per-unit Cost Price (CP) and Selling Price (SP) in dollars ($) for each of the six kits in the fiscal year 2012:

Diagnostic KitCost Price per unit ($)Selling Price per unit ($)Per-Unit Profit / Loss ($)
Kit-A3525−10 (Loss)
Kit-B3040+10 (Profit)
Kit-C5535−20 (Loss)
Kit-D4555+10 (Profit)
Kit-E3015−15 (Loss)
Kit-F7035−35 (Loss)

The company sold all units of the six kits that it produced in the given year.

‘X’ was the overall profit percentage on the profit-making kits and ‘Y’ was the overall loss percentage on the loss-making kits for the company in 2012. If the company produced an equal number of units of all six kits, then find the sum of the values of X and Y.

  1. A

    68.78%

  2. B

    82.60%

  3. C

    42.78%

  4. D

    90.45%

Show answer and explanation

Answer: (A) 68.78%

Question 14 Solution:

Let the number of units produced of each of the six items be xx.

Profit-making items: Kit-B and Kit-D.

  • Profit on Kit-B = 10x10x, CP = 30x30x

  • Profit on Kit-D = 10x10x, CP = 45x45x

  • Total Profit = 20x20x, Total CP = 75x75x

X=20x75x×100=26.667%X = \frac{20x}{75x} \times 100 = 26.667\%

Loss-making items: Kit-A, Kit-C, Kit-E, and Kit-F.

  • Loss on Kit-A = 10x10x, CP = 35x35x

  • Loss on Kit-C = 20x20x, CP = 55x55x

  • Loss on Kit-E = 15x15x, CP = 30x30x

  • Loss on Kit-F = 35x35x, CP = 70x70x

  • Total Loss = (10+20+15+35)x=80x(10 + 20 + 15 + 35)x = 80x

  • Total CP = (35+55+30+70)x=190x(35 + 55 + 30 + 70)x = 190x

Y=80x190x×100=80019≈42.105%Y = \frac{80x}{190x} \times 100 = \frac{800}{19} \approx 42.105\%

Sum of values:

X+Y=26.667+42.105=68.772%≈68.78%X + Y = 26.667 + 42.105 = 68.772\% \approx 68.78\%

Therefore, the sum is 68.78%.

Question 14 · Moderate

Directions for Questions 13 to 15

Answer the questions on the basis of the information given below.

A medical technology company, Apex Bio Diagnostics, manufactures six specialized diagnostic test kits: Kit-A, Kit-B, Kit-C, Kit-D, Kit-E, and Kit-F. The table below shows the per-unit Cost Price (CP) and Selling Price (SP) in dollars ($) for each of the six kits in the fiscal year 2012:

Diagnostic KitCost Price per unit ($)Selling Price per unit ($)Per-Unit Profit / Loss ($)
Kit-A3525−10 (Loss)
Kit-B3040+10 (Profit)
Kit-C5535−20 (Loss)
Kit-D4555+10 (Profit)
Kit-E3015−15 (Loss)
Kit-F7035−35 (Loss)

The company sold all units of the six kits that it produced in the given year.

On which of the six kits did the company register the maximum total loss in the year 2012?

  1. A

    Kit-E

  2. B

    Kit-F

  3. C

    Kit-C

  4. D

    Cannot be determined

Show answer and explanation

Answer: (D) Cannot be determined

Question 15 Solution:

While the per-unit loss is highest for Kit-F ($35 per unit), the total loss registered on any kit depends on the total number of units produced and sold for that specific kit.

Since the individual production quantities of the kits are not specified in the main passage, the kit with the maximum overall dollar loss cannot be determined.

Therefore, the correct answer is Cannot be determined.

Question 15 · Moderate

Directions for Questions 16 and 17

Answer the questions on the basis of the information given below.

Five autonomous AI coding agents — Agent-R, Agent-D, Agent-T, Agent-S, and Agent-K — completed coding sprints delivering target module quotas of 200, 180, 140, 150, and 240 units respectively over periods of 12, 10, 8, 10, and 12 days.

The following table provides information regarding the cumulative total units of work completed by each agent by the end of each sprint day:

AfterAgent-RAgent-DAgent-TAgent-SAgent-K
1 day101252018
2 days2232152520
3 days2450254044
4 days4052505048
5 days6084806550
6 days68120968095
7 days140140108100100
8 days148156140125112
9 days176172140140115
10 days180180140150120
11 days184180140150160
12 days200180140150240

Definition: An agent is said to be more efficient than another agent only if it completed more units of work than that agent on at least seven out of the twelve given days.

Out of the mentioned five agents, how many agents are more efficient than exactly two agents?

  1. A

    4

  2. B

    2

  3. C

    1

  4. D

    3

Show answer and explanation

Answer: (D) 3

Question 16 Solution:

First, compute the incremental units completed on each individual day:

DayAgent-RAgent-DAgent-TAgent-SAgent-K
Day 1101252018
Day 212201052
Day 3218101524
Day 416225104
Day 5203230152
Day 6836161545
Day 7722012205
Day 8816322512
Day 928160153
Day 10480105
Day 11400040
Day 121600080

Comparing daily output across pairs:

  • Agent-R completed more daily units than Agent-T on 7 days (Days 1, 2, 7, 9, 10, 11, 12) and more than Agent-S on 7 days (Days 2, 4, 5, 7, 9, 11, 12). Thus, Agent-R is more efficient than 2 agents (T and S).

  • Agent-D is more efficient than Agent-R (7 days) and Agent-T (8 days). Thus, Agent-D is more efficient than 2 agents (R and T).

  • Agent-T is more efficient than 0 agents.

  • Agent-S is more efficient than Agent-K (7 days), so more efficient than 1 agent.

  • Agent-K is more efficient than Agent-T (7 days) and Agent-R (7 days), so more efficient than 2 agents (T and R).

Therefore, exactly 3 agents (Agent-R, Agent-D, and Agent-K) are more efficient than exactly two agents.

Question 16 · Moderate

Directions for Questions 16 and 17

Answer the questions on the basis of the information given below.

Five autonomous AI coding agents — Agent-R, Agent-D, Agent-T, Agent-S, and Agent-K — completed coding sprints delivering target module quotas of 200, 180, 140, 150, and 240 units respectively over periods of 12, 10, 8, 10, and 12 days.

The following table provides information regarding the cumulative total units of work completed by each agent by the end of each sprint day:

AfterAgent-RAgent-DAgent-TAgent-SAgent-K
1 day101252018
2 days2232152520
3 days2450254044
4 days4052505048
5 days6084806550
6 days68120968095
7 days140140108100100
8 days148156140125112
9 days176172140140115
10 days180180140150120
11 days184180140150160
12 days200180140150240

Definition: An agent is said to be more efficient than another agent only if it completed more units of work than that agent on at least seven out of the twelve given days.

On how many days was the number of units of work completed by Agent-S more than at least one and at most two out of the mentioned agents?

  1. A

    5

  2. B

    8

  3. C

    6

  4. D

    7

Show answer and explanation

Answer: (D) 7

Question 17 Solution:

Reviewing the daily output of Agent-S compared to the other 4 agents on each of the 12 days:

  • Day 1: S = 20. Outperformed R(10), D(12), T(5), K(18) (4 agents). Not in range [1, 2].

  • Day 2: S = 5. Outperformed K(2) (1 agent). In range!

  • Day 3: S = 15. Outperformed R(2), T(10) (2 agents). In range!

  • Day 4: S = 10. Outperformed D(2), K(4) (2 agents). In range!

  • Day 5: S = 15. Outperformed K(2) (1 agent). In range!

  • Day 6: S = 15. Outperformed R(8) (1 agent). In range!

  • Day 7: S = 20. Outperformed T(12), K(5) (2 agents). In range!

  • Day 8: S = 25. Outperformed R(8), D(16), K(12) (3 agents). Not in range.

  • Day 9: S = 15. Outperformed T(0), K(3) (2 agents). In range!

  • Day 10: S = 10. Outperformed R(4), D(8), T(0), K(5) (4 agents). Not in range.

  • Day 11: S = 0. Outperformed none (0 agents). Not in range.

  • Day 12: S = 0. Outperformed none (0 agents). Not in range.

The days on which Agent-S beat at least one and at most two agents are Days 2, 3, 4, 5, 6, 7, and 9, which totals 7 days.

Question 17 · Moderate

Directions for Questions 18 to 21

Answer the questions on the basis of the information given below.

An advanced semiconductor manufacturing conglomerate operates three fabrication plants — Fab-1, Fab-2, and Fab-3. The cleanroom technical workforce in each fab is categorized by seniority (Senior vs Junior) and specialization (Hardware Systems vs Software Automation).

The following workforce data is known:

  • The total workforce in Fab-1 is 30 less than that in Fab-2, but 20 more than that in Fab-3. The combined total workforce across all three fabs is 370.

  • In Fab-2, the number of junior hardware engineers is equal to the number of junior software engineers.

  • The total number of hardware engineers in Fab-1 is equal to the total number of software engineers in Fab-3.

  • The number of senior hardware engineers in Fab-2 is equal to the number of junior software engineers in Fab-3.

  • Thrice the number of senior hardware engineers in Fab-3 is equal to the number of junior hardware engineers in Fab-2, which, in turn, is equal to the number of senior software engineers in Fab-3.

  • The number of junior hardware engineers in Fab-3 is 80% of the number of senior software engineers in Fab-1.

  • Twice the number of junior hardware engineers in Fab-1 is equal to the number of junior hardware engineers in Fab-3, which, in turn, is equal to two-fifths of the number of senior software engineers in Fab-2.

  • In Fab-1, the number of senior software engineers is equal to the number of junior software engineers.

Which fab has the highest number of senior hardware engineers?

  1. A

    Fab-1

  2. B

    Fab-2

  3. C

    Fab-3

  4. D

    Cannot be determined

Show answer and explanation

Answer: (A) Fab-1

Question 18 Solution:

Step 1: Fab Totals

Let the total number of engineers in Fab-1 be xx.

  • Total in Fab-2 = x+30x + 30

  • Total in Fab-3 = x−20x - 20

  • Total across all three fabs = x+(x+30)+(x−20)=3x+10=370  ⟹  x=120x + (x + 30) + (x - 20) = 3x + 10 = 370 \implies x = 120

Thus, Total(Fab-1) = 120, Total(Fab-2) = 150, Total(Fab-3) = 100.

Step 2: Defining Workforce Variables

Let:

  • bb = Senior software engineers in Fab-1 = Junior software engineers in Fab-1.

  • Junior hardware in Fab-3 = 0.8b0.8b.

  • Junior hardware in Fab-1 = 0.8b2=0.4b\frac{0.8b}{2} = 0.4b.

  • Senior software in Fab-2 = 0.8b2/5=2b\frac{0.8b}{2/5} = 2b.

  • Junior hardware in Fab-2 = dd, which implies junior software in Fab-2 = dd.

  • Senior software in Fab-3 = dd.

  • Senior hardware in Fab-3 = d/3d/3.

  • Junior software in Fab-3 = Senior hardware in Fab-2 = cc.

  • Senior hardware in Fab-1 = aa.

Step 3: Solving the System

  • For Fab-1 (Total = 120): a+0.4b+b+b=120  ⟹  a+2.4b=120a + 0.4b + b + b = 120 \implies a + 2.4b = 120

  • Total hardware in Fab-1 = Total software in Fab-3:

    a+0.4b=d+ca + 0.4b = d + c

  • For Fab-3 (Total = 100):

    d3+0.8b+d+c=100  ⟹  4d3+0.8b+c=100\frac{d}{3} + 0.8b + d + c = 100 \implies \frac{4d}{3} + 0.8b + c = 100

  • For Fab-2 (Total = 150):

    c+d+2b+d=150  ⟹  c+2d+2b=150c + d + 2b + d = 150 \implies c + 2d + 2b = 150

Subtracting the equations and solving yields:

d=30,b=25,c=40,a=60d = 30, \quad b = 25, \quad c = 40, \quad a = 60

Workforce Distribution Matrix:

FacilitySenior HardwareJunior HardwareSenior SoftwareJunior SoftwareTotal
Fab-160102525120
Fab-240305030150
Fab-310203040100

Comparing senior hardware engineers:

  • Fab-1 = 60

  • Fab-2 = 40

  • Fab-3 = 10

Fab-1 has the highest number of senior hardware engineers (60).

Question 18 · Moderate

Directions for Questions 18 to 21

Answer the questions on the basis of the information given below.

An advanced semiconductor manufacturing conglomerate operates three fabrication plants — Fab-1, Fab-2, and Fab-3. The cleanroom technical workforce in each fab is categorized by seniority (Senior vs Junior) and specialization (Hardware Systems vs Software Automation).

The following workforce data is known:

  • The total workforce in Fab-1 is 30 less than that in Fab-2, but 20 more than that in Fab-3. The combined total workforce across all three fabs is 370.

  • In Fab-2, the number of junior hardware engineers is equal to the number of junior software engineers.

  • The total number of hardware engineers in Fab-1 is equal to the total number of software engineers in Fab-3.

  • The number of senior hardware engineers in Fab-2 is equal to the number of junior software engineers in Fab-3.

  • Thrice the number of senior hardware engineers in Fab-3 is equal to the number of junior hardware engineers in Fab-2, which, in turn, is equal to the number of senior software engineers in Fab-3.

  • The number of junior hardware engineers in Fab-3 is 80% of the number of senior software engineers in Fab-1.

  • Twice the number of junior hardware engineers in Fab-1 is equal to the number of junior hardware engineers in Fab-3, which, in turn, is equal to two-fifths of the number of senior software engineers in Fab-2.

  • In Fab-1, the number of senior software engineers is equal to the number of junior software engineers.

The number of senior software engineers in Fab-3 is what percentage of that in Fab-2?

  1. A

    60%

  2. B

    40%

  3. C

    50%

  4. D

    70%

Show answer and explanation

Answer: (A) 60%

Question 19 Solution:

From the deduced matrix:

  • Senior software engineers in Fab-3 = 30

  • Senior software engineers in Fab-2 = 50

Percentage=3050×100=60%\text{Percentage} = \frac{30}{50} \times 100 = 60\%

Therefore, the percentage is 60%.

Question 19 · Moderate

Directions for Questions 18 to 21

Answer the questions on the basis of the information given below.

An advanced semiconductor manufacturing conglomerate operates three fabrication plants — Fab-1, Fab-2, and Fab-3. The cleanroom technical workforce in each fab is categorized by seniority (Senior vs Junior) and specialization (Hardware Systems vs Software Automation).

The following workforce data is known:

  • The total workforce in Fab-1 is 30 less than that in Fab-2, but 20 more than that in Fab-3. The combined total workforce across all three fabs is 370.

  • In Fab-2, the number of junior hardware engineers is equal to the number of junior software engineers.

  • The total number of hardware engineers in Fab-1 is equal to the total number of software engineers in Fab-3.

  • The number of senior hardware engineers in Fab-2 is equal to the number of junior software engineers in Fab-3.

  • Thrice the number of senior hardware engineers in Fab-3 is equal to the number of junior hardware engineers in Fab-2, which, in turn, is equal to the number of senior software engineers in Fab-3.

  • The number of junior hardware engineers in Fab-3 is 80% of the number of senior software engineers in Fab-1.

  • Twice the number of junior hardware engineers in Fab-1 is equal to the number of junior hardware engineers in Fab-3, which, in turn, is equal to two-fifths of the number of senior software engineers in Fab-2.

  • In Fab-1, the number of senior software engineers is equal to the number of junior software engineers.

What is the absolute difference between the total number of software engineers and that of the hardware engineers across all three fabs combined?

  1. A

    10

  2. B

    20

  3. C

    30

  4. D

    40

Show answer and explanation

Answer: (C) 30

Question 20 Solution:

From the deduced matrix:

  • Total Hardware Engineers:

    • Fab-1: 60+10=7060 + 10 = 70

    • Fab-2: 40+30=7040 + 30 = 70

    • Fab-3: 10+20=3010 + 20 = 30

    • Overall Hardware Total = 70+70+30=17070 + 70 + 30 = 170

  • Total Software Engineers:

    • Fab-1: 25+25=5025 + 25 = 50

    • Fab-2: 50+30=8050 + 30 = 80

    • Fab-3: 30+40=7030 + 40 = 70

    • Overall Software Total = 50+80+70=20050 + 80 + 70 = 200

Absolute Difference=∣200−170∣=30\text{Absolute Difference} = |200 - 170| = 30

Therefore, the absolute difference is 30.

Question 20 · Moderate

Directions for Questions 18 to 21

Answer the questions on the basis of the information given below.

An advanced semiconductor manufacturing conglomerate operates three fabrication plants — Fab-1, Fab-2, and Fab-3. The cleanroom technical workforce in each fab is categorized by seniority (Senior vs Junior) and specialization (Hardware Systems vs Software Automation).

The following workforce data is known:

  • The total workforce in Fab-1 is 30 less than that in Fab-2, but 20 more than that in Fab-3. The combined total workforce across all three fabs is 370.

  • In Fab-2, the number of junior hardware engineers is equal to the number of junior software engineers.

  • The total number of hardware engineers in Fab-1 is equal to the total number of software engineers in Fab-3.

  • The number of senior hardware engineers in Fab-2 is equal to the number of junior software engineers in Fab-3.

  • Thrice the number of senior hardware engineers in Fab-3 is equal to the number of junior hardware engineers in Fab-2, which, in turn, is equal to the number of senior software engineers in Fab-3.

  • The number of junior hardware engineers in Fab-3 is 80% of the number of senior software engineers in Fab-1.

  • Twice the number of junior hardware engineers in Fab-1 is equal to the number of junior hardware engineers in Fab-3, which, in turn, is equal to two-fifths of the number of senior software engineers in Fab-2.

  • In Fab-1, the number of senior software engineers is equal to the number of junior software engineers.

The number of junior software engineers across all three fabs combined is:

  1. A

    90

  2. B

    95

  3. C

    105

  4. D

    100

Show answer and explanation

Answer: (B) 95

Question 21 Solution:

From the deduced matrix, the junior software engineers are:

  • Fab-1: b=25b = 25

  • Fab-2: d=30d = 30

  • Fab-3: c=40c = 40

Total=25+30+40=95\text{Total} = 25 + 30 + 40 = 95

Therefore, the total number of junior software engineers is 95.

Question 21 · Moderate

Directions for Questions 22 to 25

Answer the questions on the basis of the information given below.

Four major enterprise cloud and software conglomerates — CyberCore (CCA), TechTitan (T&T), CloudBase (CB), and LogicLink (LLL) — reported their financial operations for FY 2014-15 across four key parameters: Capital Base, Domestic ARR, Global ARR, and Net Profit (all values in million dollars, $M).

The following operational relationships are known:

I. The Domestic ARR of LogicLink was eleven times the Net Profit of TechTitan.

II. The sum of Global ARR of CloudBase and LogicLink was equal to half the Capital Base of CyberCore.

III. The sum of Capital Base of CloudBase and Domestic ARR of TechTitan was equal to five times the Global ARR of TechTitan.

IV. The Net Profit of LogicLink = Domestic ARR of CyberCore − Global ARR of CloudBase.

V. The Net Profit of CyberCore was equal to the arithmetic average of the Net Profits of TechTitan and LogicLink.

VI. The total Net Profit of all four companies combined was $100M.

VII. The Domestic ARR of CloudBase was 1.25 times the Global ARR of CyberCore.

VIII. The Capital Base of TechTitan = Total Global ARR of all four companies combined − $50M.

IX. The Capital Base of LogicLink = 7×(Net Profit of LogicLink+Net Profit of CloudBase−Net Profit of TechTitan)7 \times (\text{Net Profit of LogicLink} + \text{Net Profit of CloudBase} - \text{Net Profit of TechTitan}).

Additional Confirmed Data:

  1. Domestic ARR of LogicLink = $110M

  2. Capital Base of CyberCore = $500M

  3. Domestic ARR of CloudBase = $100M

  4. Capital Base of CloudBase = $450M

  5. Global ARR of TechTitan = $120M

  6. Net Profit of CloudBase = $40M

  7. Global ARR of CloudBase = $100M

In FY 2014-15, the Capital Base of LogicLink was:

  1. A

    $500 million

  2. B

    $400 million

  3. C

    $450 million

  4. D

    $420 million

Show answer and explanation

Answer: (D) $420 million

Question 22 Solution:

Let us determine the metrics step by step:

  1. From (I) and Confirmed Data (1): Domestic ARR(LLL) = 110 = 11 \times \text{Profit(T&T)} \implies \text{Profit(T&T)} = \text{\textdollar}10\text{M}.

  2. From (VI): Total Profit = 100. Given Profit(CB) = 40, Profit(T&T) = 10:

    Profit(CCA)+Profit(LLL)=100−(40+10)=50\text{Profit(CCA)} + \text{Profit(LLL)} = 100 - (40 + 10) = 50

  3. From (V): \text{Profit(CCA)} = \frac{\text{Profit(T&T)} + \text{Profit(LLL)}}{2} = \frac{10 + \text{Profit(LLL)}}{2}.

    Substituting into Profit(CCA)+Profit(LLL)=50\text{Profit(CCA)} + \text{Profit(LLL)} = 50:

    10+Profit(LLL)2+Profit(LLL)=50  ⟹  1.5×Profit(LLL)=45  ⟹  Profit(LLL)=$30M\frac{10 + \text{Profit(LLL)}}{2} + \text{Profit(LLL)} = 50 \implies 1.5 \times \text{Profit(LLL)} = 45 \implies \text{Profit(LLL)} = \text{\textdollar}30\text{M}

    Profit(CCA)=50−30=$20M\text{Profit(CCA)} = 50 - 30 = \text{\textdollar}20\text{M}

  4. From (IX):

    \text{Capital(LLL)} = 7 \times (\text{Profit(LLL)} + \text{Profit(CB)} - \text{Profit(T&T)})

    Capital(LLL)=7×(30+40−10)=7×60=$420M\text{Capital(LLL)} = 7 \times (30 + 40 - 10) = 7 \times 60 = \text{\textdollar}420\text{M}

Therefore, the Capital Base of LogicLink was $420 million.

Question 22 · Moderate

Directions for Questions 22 to 25

Answer the questions on the basis of the information given below.

Four major enterprise cloud and software conglomerates — CyberCore (CCA), TechTitan (T&T), CloudBase (CB), and LogicLink (LLL) — reported their financial operations for FY 2014-15 across four key parameters: Capital Base, Domestic ARR, Global ARR, and Net Profit (all values in million dollars, $M).

The following operational relationships are known:

I. The Domestic ARR of LogicLink was eleven times the Net Profit of TechTitan.

II. The sum of Global ARR of CloudBase and LogicLink was equal to half the Capital Base of CyberCore.

III. The sum of Capital Base of CloudBase and Domestic ARR of TechTitan was equal to five times the Global ARR of TechTitan.

IV. The Net Profit of LogicLink = Domestic ARR of CyberCore − Global ARR of CloudBase.

V. The Net Profit of CyberCore was equal to the arithmetic average of the Net Profits of TechTitan and LogicLink.

VI. The total Net Profit of all four companies combined was $100M.

VII. The Domestic ARR of CloudBase was 1.25 times the Global ARR of CyberCore.

VIII. The Capital Base of TechTitan = Total Global ARR of all four companies combined − $50M.

IX. The Capital Base of LogicLink = 7×(Net Profit of LogicLink+Net Profit of CloudBase−Net Profit of TechTitan)7 \times (\text{Net Profit of LogicLink} + \text{Net Profit of CloudBase} - \text{Net Profit of TechTitan}).

Additional Confirmed Data:

  1. Domestic ARR of LogicLink = $110M

  2. Capital Base of CyberCore = $500M

  3. Domestic ARR of CloudBase = $100M

  4. Capital Base of CloudBase = $450M

  5. Global ARR of TechTitan = $120M

  6. Net Profit of CloudBase = $40M

  7. Global ARR of CloudBase = $100M

In FY 2014-15, the Net Profit of LogicLink was:

  1. A

    $30 million

  2. B

    $40 million

  3. C

    $20 million

  4. D

    $10 million

Show answer and explanation

Answer: (A) $30 million

Question 23 Solution:

From Question 22:

  • Profit(T&T) = 110/11=$10M110 / 11 = \text{\textdollar}10\text{M}

  • Profit(CB) = $40M\text{\textdollar}40\text{M}

  • Profit(CCA) + Profit(LLL) = 100−(10+40)=50100 - (10 + 40) = 50

  • Profit(CCA) = (10+Profit(LLL))/2(10 + \text{Profit(LLL)}) / 2

Solving gives:

Profit(LLL)=$30M\text{Profit(LLL)} = \text{\textdollar}30\text{M}

Therefore, the Net Profit of LogicLink was $30 million.

Question 23 · Moderate

Directions for Questions 22 to 25

Answer the questions on the basis of the information given below.

Four major enterprise cloud and software conglomerates — CyberCore (CCA), TechTitan (T&T), CloudBase (CB), and LogicLink (LLL) — reported their financial operations for FY 2014-15 across four key parameters: Capital Base, Domestic ARR, Global ARR, and Net Profit (all values in million dollars, $M).

The following operational relationships are known:

I. The Domestic ARR of LogicLink was eleven times the Net Profit of TechTitan.

II. The sum of Global ARR of CloudBase and LogicLink was equal to half the Capital Base of CyberCore.

III. The sum of Capital Base of CloudBase and Domestic ARR of TechTitan was equal to five times the Global ARR of TechTitan.

IV. The Net Profit of LogicLink = Domestic ARR of CyberCore − Global ARR of CloudBase.

V. The Net Profit of CyberCore was equal to the arithmetic average of the Net Profits of TechTitan and LogicLink.

VI. The total Net Profit of all four companies combined was $100M.

VII. The Domestic ARR of CloudBase was 1.25 times the Global ARR of CyberCore.

VIII. The Capital Base of TechTitan = Total Global ARR of all four companies combined − $50M.

IX. The Capital Base of LogicLink = 7×(Net Profit of LogicLink+Net Profit of CloudBase−Net Profit of TechTitan)7 \times (\text{Net Profit of LogicLink} + \text{Net Profit of CloudBase} - \text{Net Profit of TechTitan}).

Additional Confirmed Data:

  1. Domestic ARR of LogicLink = $110M

  2. Capital Base of CyberCore = $500M

  3. Domestic ARR of CloudBase = $100M

  4. Capital Base of CloudBase = $450M

  5. Global ARR of TechTitan = $120M

  6. Net Profit of CloudBase = $40M

  7. Global ARR of CloudBase = $100M

In FY 2014-15, the Capital Base of CyberCore was greater than the Capital Base of TechTitan by:

  1. A

    10%

  2. B

    20%

  3. C

    25%

  4. D

    50%

Show answer and explanation

Answer: (C) 25%

Question 24 Solution:

  1. From (II): Global(CB) + Global(LLL) = 0.5×Capital(CCA)=0.5×500=2500.5 \times \text{Capital(CCA)} = 0.5 \times 500 = 250.

    Given Global(CB) = 100   ⟹  \implies Global(LLL) = 250−100=$150M250 - 100 = \text{\textdollar}150\text{M}.

  2. From (VII): Domestic(CB) = 100 = 1.25×Global(CCA)  ⟹  Global(CCA)=100/1.25=$80M1.25 \times \text{Global(CCA)} \implies \text{Global(CCA)} = 100 / 1.25 = \text{\textdollar}80\text{M}.

  3. Total Global ARR across all four companies:

    \text{Total Global ARR} = \text{Global(CCA)} + \text{Global(T&T)} + \text{Global(CB)} + \text{Global(LLL)}

    =80+120+100+150=$450M= 80 + 120 + 100 + 150 = \text{\textdollar}450\text{M}

  4. From (VIII): Capital(T&T) = Total Global ARR − 50 = 450−50=$400M450 - 50 = \text{\textdollar}400\text{M}.

  5. Capital(CCA) = $500M\text{\textdollar}500\text{M}.

Percentage Greater=500−400400×100=100400×100=25%\text{Percentage Greater} = \frac{500 - 400}{400} \times 100 = \frac{100}{400} \times 100 = 25\%

Therefore, Capital(CCA) was greater than Capital(T&T) by 25%.

Question 24 · Moderate

Directions for Questions 22 to 25

Answer the questions on the basis of the information given below.

Four major enterprise cloud and software conglomerates — CyberCore (CCA), TechTitan (T&T), CloudBase (CB), and LogicLink (LLL) — reported their financial operations for FY 2014-15 across four key parameters: Capital Base, Domestic ARR, Global ARR, and Net Profit (all values in million dollars, $M).

The following operational relationships are known:

I. The Domestic ARR of LogicLink was eleven times the Net Profit of TechTitan.

II. The sum of Global ARR of CloudBase and LogicLink was equal to half the Capital Base of CyberCore.

III. The sum of Capital Base of CloudBase and Domestic ARR of TechTitan was equal to five times the Global ARR of TechTitan.

IV. The Net Profit of LogicLink = Domestic ARR of CyberCore − Global ARR of CloudBase.

V. The Net Profit of CyberCore was equal to the arithmetic average of the Net Profits of TechTitan and LogicLink.

VI. The total Net Profit of all four companies combined was $100M.

VII. The Domestic ARR of CloudBase was 1.25 times the Global ARR of CyberCore.

VIII. The Capital Base of TechTitan = Total Global ARR of all four companies combined − $50M.

IX. The Capital Base of LogicLink = 7×(Net Profit of LogicLink+Net Profit of CloudBase−Net Profit of TechTitan)7 \times (\text{Net Profit of LogicLink} + \text{Net Profit of CloudBase} - \text{Net Profit of TechTitan}).

Additional Confirmed Data:

  1. Domestic ARR of LogicLink = $110M

  2. Capital Base of CyberCore = $500M

  3. Domestic ARR of CloudBase = $100M

  4. Capital Base of CloudBase = $450M

  5. Global ARR of TechTitan = $120M

  6. Net Profit of CloudBase = $40M

  7. Global ARR of CloudBase = $100M

In FY 2014-15, the total Domestic ARR of all four companies combined as a percentage of total Capital Base of all four companies combined was approximately:

  1. A

    36%

  2. B

    28%

  3. C

    32%

  4. D

    33%

Show answer and explanation

Answer: (B) 28%

Question 25 Solution:

Domestic ARR of all four companies:

  • From (IV): Profit(LLL) = Domestic(CCA) − Global(CB)   ⟹  30=Domestic(CCA)−100  ⟹  Domestic(CCA)=$130M\implies 30 = \text{Domestic(CCA)} - 100 \implies \text{Domestic(CCA)} = \text{\textdollar}130\text{M}.

  • From (III): Capital(CB) + Domestic(T&T) = 5 \times \text{Global(T&T)} \implies 450 + \text{Domestic(T&T)} = 5 \times 120 = 600 \implies \text{Domestic(T&T)} = \text{\textdollar}150\text{M}.

  • Domestic(CB) = $100M\text{\textdollar}100\text{M} (given).

  • Domestic(LLL) = $110M\text{\textdollar}110\text{M} (given).

Total Domestic ARR=130+150+100+110=$490M\text{Total Domestic ARR} = 130 + 150 + 100 + 110 = \text{\textdollar}490\text{M}

Capital Base of all four companies:

  • Capital(CCA) = $500M\text{\textdollar}500\text{M}

  • Capital(T&T) = $400M\text{\textdollar}400\text{M}

  • Capital(CB) = $450M\text{\textdollar}450\text{M}

  • Capital(LLL) = $420M\text{\textdollar}420\text{M}

Total Capital Base=500+400+450+420=$1770M\text{Total Capital Base} = 500 + 400 + 450 + 420 = \text{\textdollar}1770\text{M}

Percentage:

Percentage=4901770×100≈27.68%≈28%\text{Percentage} = \frac{490}{1770} \times 100 \approx 27.68\% \approx 28\%

Therefore, the required percentage is 28%.

Question 25 · CAT Level

Directions for Questions 26 to 29

Answer the questions on the basis of the information given below.

A developer productivity survey analyzed enterprise user adoption across five core feature modules on a cloud platform: Module I (API Analytics), Module II (Cloud Storage), Module III (Data Pipelines), Module IV (Real-time Dashboards), and Module V (Automated Billing).

The Venn diagram data represents the active user count across the overlapping feature modules (in units of 12,000 users):

  • Mod-IV only = 15

  • Mod-IV and Mod-I (Analytics & Dashboards) = 26 + 7 = 33

  • Mod-IV and Mod-V (Dashboards & Billing) = 14 + 9 = 23

  • Mod-II and Mod-IV (Storage & Dashboards) = 28

  • Mod-IV and Mod-III (Dashboards & Pipelines) = 4 + 11 = 15

  • Total user base across the entire Venn distribution = 200 units (representing 200×12,000=2,400,000200 \times 12{,}000 = 2{,}400{,}000 users).

Note: All figures in the distribution table are in multiples of 12,000.

Additional customer segmentation data for Module IV (Real-time Dashboards) provides the breakdown by enterprise seat tier:

  • Tier-1 (10–14 seats): 90° (25%25\%)

  • Tier-2 (14–16 seats): 30° (8.33%8.33\%, or 1/121/12)

  • Tier-3 (16–20 seats): 120° (33.33%33.33\%, or 1/31/3)

  • Tier-4 (20+ seats): 120° (33.33%33.33\%, or 1/31/3)

How many users who use Real-time Dashboards only or Data Pipelines only belong to Tier-3 (16–20 seats)?

  1. A

    280,000

  2. B

    336,000

  3. C

    240,000

  4. D

    Data Insufficient

Show answer and explanation

Answer: (D) Data Insufficient

Question 26 Solution:

While the seat tier distribution is explicitly given for Module IV (Real-time Dashboards), no seat tier distribution is provided for Module III (Data Pipelines).

Consequently, the number of users who use Data Pipelines only belonging to Tier-3 cannot be calculated.

Therefore, the answer is Data Insufficient.

Question 26 · CAT Level

Directions for Questions 26 to 29

Answer the questions on the basis of the information given below.

A developer productivity survey analyzed enterprise user adoption across five core feature modules on a cloud platform: Module I (API Analytics), Module II (Cloud Storage), Module III (Data Pipelines), Module IV (Real-time Dashboards), and Module V (Automated Billing).

The Venn diagram data represents the active user count across the overlapping feature modules (in units of 12,000 users):

  • Mod-IV only = 15

  • Mod-IV and Mod-I (Analytics & Dashboards) = 26 + 7 = 33

  • Mod-IV and Mod-V (Dashboards & Billing) = 14 + 9 = 23

  • Mod-II and Mod-IV (Storage & Dashboards) = 28

  • Mod-IV and Mod-III (Dashboards & Pipelines) = 4 + 11 = 15

  • Total user base across the entire Venn distribution = 200 units (representing 200×12,000=2,400,000200 \times 12{,}000 = 2{,}400{,}000 users).

Note: All figures in the distribution table are in multiples of 12,000.

Additional customer segmentation data for Module IV (Real-time Dashboards) provides the breakdown by enterprise seat tier:

  • Tier-1 (10–14 seats): 90° (25%25\%)

  • Tier-2 (14–16 seats): 30° (8.33%8.33\%, or 1/121/12)

  • Tier-3 (16–20 seats): 120° (33.33%33.33\%, or 1/31/3)

  • Tier-4 (20+ seats): 120° (33.33%33.33\%, or 1/31/3)

How many users who use API Analytics and Real-time Dashboards belong to Tier-2 (14–16 seats)?

  1. A

    33,000

  2. B

    36,000

  3. C

    30,000

  4. D

    None of these

Show answer and explanation

Answer: (A) 33,000

Question 27 Solution:

From the Venn distribution, users adopting both API Analytics (Module I) and Real-time Dashboards (Module IV) total:

Total=(26+7)×12,000=33×12,000=396,000\text{Total} = (26 + 7) \times 12{,}000 = 33 \times 12{,}000 = 396{,}000

For Tier-2 (14–16 seats), the angular sector is 30°:

Proportion=30∘360∘=112\text{Proportion} = \frac{30^\circ}{360^\circ} = \frac{1}{12}

Users in Tier-2=33×12,000×112=33×1,000=33,000\text{Users in Tier-2} = 33 \times 12{,}000 \times \frac{1}{12} = 33 \times 1{,}000 = 33{,}000

Therefore, the number of users is 33,000.

Question 27 · CAT Level

Directions for Questions 26 to 29

Answer the questions on the basis of the information given below.

A developer productivity survey analyzed enterprise user adoption across five core feature modules on a cloud platform: Module I (API Analytics), Module II (Cloud Storage), Module III (Data Pipelines), Module IV (Real-time Dashboards), and Module V (Automated Billing).

The Venn diagram data represents the active user count across the overlapping feature modules (in units of 12,000 users):

  • Mod-IV only = 15

  • Mod-IV and Mod-I (Analytics & Dashboards) = 26 + 7 = 33

  • Mod-IV and Mod-V (Dashboards & Billing) = 14 + 9 = 23

  • Mod-II and Mod-IV (Storage & Dashboards) = 28

  • Mod-IV and Mod-III (Dashboards & Pipelines) = 4 + 11 = 15

  • Total user base across the entire Venn distribution = 200 units (representing 200×12,000=2,400,000200 \times 12{,}000 = 2{,}400{,}000 users).

Note: All figures in the distribution table are in multiples of 12,000.

Additional customer segmentation data for Module IV (Real-time Dashboards) provides the breakdown by enterprise seat tier:

  • Tier-1 (10–14 seats): 90° (25%25\%)

  • Tier-2 (14–16 seats): 30° (8.33%8.33\%, or 1/121/12)

  • Tier-3 (16–20 seats): 120° (33.33%33.33\%, or 1/31/3)

  • Tier-4 (20+ seats): 120° (33.33%33.33\%, or 1/31/3)

How many users who use Real-time Dashboards and Automated Billing belong to Tier-1 (10–14 seats)?

  1. A

    92,000

  2. B

    69,000

  3. C

    72,000

  4. D

    Data Insufficient

Show answer and explanation

Answer: (B) 69,000

Question 28 Solution:

From the Venn distribution, users adopting both Real-time Dashboards (Module IV) and Automated Billing (Module V) total:

Total=(14+9)×12,000=23×12,000=276,000\text{Total} = (14 + 9) \times 12{,}000 = 23 \times 12{,}000 = 276{,}000

For Tier-1 (10–14 seats), the angular sector is 90°:

Proportion=90∘360∘=14\text{Proportion} = \frac{90^\circ}{360^\circ} = \frac{1}{4}

Users in Tier-1=23×12,000×14=23×3,000=69,000\text{Users in Tier-1} = 23 \times 12{,}000 \times \frac{1}{4} = 23 \times 3{,}000 = 69{,}000

Therefore, the number of users is 69,000.

Question 28 · CAT Level

Directions for Questions 26 to 29

Answer the questions on the basis of the information given below.

A developer productivity survey analyzed enterprise user adoption across five core feature modules on a cloud platform: Module I (API Analytics), Module II (Cloud Storage), Module III (Data Pipelines), Module IV (Real-time Dashboards), and Module V (Automated Billing).

The Venn diagram data represents the active user count across the overlapping feature modules (in units of 12,000 users):

  • Mod-IV only = 15

  • Mod-IV and Mod-I (Analytics & Dashboards) = 26 + 7 = 33

  • Mod-IV and Mod-V (Dashboards & Billing) = 14 + 9 = 23

  • Mod-II and Mod-IV (Storage & Dashboards) = 28

  • Mod-IV and Mod-III (Dashboards & Pipelines) = 4 + 11 = 15

  • Total user base across the entire Venn distribution = 200 units (representing 200×12,000=2,400,000200 \times 12{,}000 = 2{,}400{,}000 users).

Note: All figures in the distribution table are in multiples of 12,000.

Additional customer segmentation data for Module IV (Real-time Dashboards) provides the breakdown by enterprise seat tier:

  • Tier-1 (10–14 seats): 90° (25%25\%)

  • Tier-2 (14–16 seats): 30° (8.33%8.33\%, or 1/121/12)

  • Tier-3 (16–20 seats): 120° (33.33%33.33\%, or 1/31/3)

  • Tier-4 (20+ seats): 120° (33.33%33.33\%, or 1/31/3)

If the seat tier distribution of Module IV applies across all five modules, then the total number of users belonging to Tier-4 (20+ seats) is:

  1. A

    80,000

  2. B

    799,920

  3. C

    800,000

  4. D

    670,000

Show answer and explanation

Answer: (C) 800,000

Question 29 Solution:

Total user base across the entire Venn distribution:

Total users=200×12,000=2,400,000\text{Total users} = 200 \times 12{,}000 = 2{,}400{,}000

For Tier-4 (20+ seats), the angular sector is 120°:

Proportion=120∘360∘=13\text{Proportion} = \frac{120^\circ}{360^\circ} = \frac{1}{3}

Total in Tier-4=2,400,000×13=800,000\text{Total in Tier-4} = 2{,}400{,}000 \times \frac{1}{3} = 800{,}000

Therefore, the total number of users in Tier-4 is 800,000.

Question 29 · CAT Level

Directions for Questions 30 to 33

Answer the questions on the basis of the information given below.

In an advanced AI Doctoral Fellowship program comprising 480 research scholars, each scholar was required to choose exactly 3 specialization tracks out of 5 available tracks: Computer Vision (CV), Natural Language Processing (NLP), Reinforcement Learning (RL), Robotics (ROB), and Quantum ML (QML).

Consider the distribution of scholars across the first three tracks: CV, NLP, and RL:

  • Every scholar chose exactly 3 tracks out of 5, which guarantees that every scholar selected at least one of the three tracks {CV, NLP, RL}.

  • Let SS, DD, and TT denote the number of scholars who selected exactly one, exactly two, and all three of the tracks {CV, NLP, RL} respectively.

  • (I) The number of scholars selecting exactly two of these tracks was one-third the sum of those selecting exactly one and all three: 3D=S+T3D = S + T.

  • In the 3-set Venn diagram:

    • Let bb, cc, and dd denote scholars choosing CV only, NLP only, and RL only respectively.

    • Let pp, qq, and rr denote scholars choosing {CV and NLP only}, {NLP and RL only}, and {CV and RL only} respectively.

    • Let xx denote scholars choosing all three tracks {CV, NLP, and RL}.

  • (II) 40≤q≤5640 \le q \le 56

  • (III) c>bc > b

  • (IV) x≥15x \ge 15

  • (V) 24≤p≤4024 \le p \le 40

  • (VI) Total scholars choosing CV was 4 less than total scholars choosing RL: Total(CV)=Total(RL)−4\text{Total}(CV) = \text{Total}(RL) - 4.

  • (VII) The number of scholars choosing RL only was 6 times the number of scholars choosing all three: d=6xd = 6x.

The minimum difference between the number of scholars who chose NLP only and CV only is:

  1. A

    11

  2. B

    13

  3. C

    15

  4. D

    18

Show answer and explanation

Answer: (B) 13

Question 30 Solution:

Step 1: Partitioning Scholars

Total scholars = 480. Since each scholar selects 3 tracks out of 5, every scholar must select at least one of {CV, NLP, RL}.

S+D+T=480S + D + T = 480

From (I), 3D=S+T  ⟹  4D=S+D+T=480  ⟹  D=1203D = S + T \implies 4D = S + D + T = 480 \implies D = 120.

Thus, p+q+r=120p + q + r = 120 and S+T=b+c+d+x=360S + T = b + c + d + x = 360.

Step 2: Analyzing Overlaps

Given 24≤p≤4024 \le p \le 40 and 40≤q≤5640 \le q \le 56, let p=24+yp = 24 + y and q=40+yq = 40 + y, where 0≤y≤160 \le y \le 16.

Then:

r=120−(p+q)=120−(64+2y)=56−2yr = 120 - (p + q) = 120 - (64 + 2y) = 56 - 2y

Step 3: Track Totals

From (VI):

Total(CV)=b+p+r+x=Total(RL)−4=(d+q+r+x)−4\text{Total}(CV) = b + p + r + x = \text{Total}(RL) - 4 = (d + q + r + x) - 4

b+p=d+q−4  ⟹  b+(24+y)=d+(40+y)−4  ⟹  b=d+12b + p = d + q - 4 \implies b + (24 + y) = d + (40 + y) - 4 \implies b = d + 12

From (VII), d=6xd = 6x, so b=6x+12b = 6x + 12.

Step 4: Solving for c

b+c+d+x=360  ⟹  (6x+12)+c+6x+x=360  ⟹  c=348−13xb + c + d + x = 360 \implies (6x + 12) + c + 6x + x = 360 \implies c = 348 - 13x

Since c>bc > b:

348−13x>6x+12  ⟹  19x<336  ⟹  x≤17348 - 13x > 6x + 12 \implies 19x < 336 \implies x \le 17

Since x≥15x \ge 15, x∈{15,16,17}x \in \{15, 16, 17\}:

  • If x=15x = 15: b=102,c=153,d=90  ⟹  c−b=153−102=51b = 102, c = 153, d = 90 \implies c - b = 153 - 102 = 51

  • If x=16x = 16: b=108,c=140,d=96  ⟹  c−b=140−108=32b = 108, c = 140, d = 96 \implies c - b = 140 - 108 = 32

  • If x=17x = 17: b=114,c=127,d=102  ⟹  c−b=127−114=13b = 114, c = 127, d = 102 \implies c - b = 127 - 114 = 13

The minimum difference between NLP only (cc) and CV only (bb) is 13.

Question 30 · CAT Level

Directions for Questions 30 to 33

Answer the questions on the basis of the information given below.

In an advanced AI Doctoral Fellowship program comprising 480 research scholars, each scholar was required to choose exactly 3 specialization tracks out of 5 available tracks: Computer Vision (CV), Natural Language Processing (NLP), Reinforcement Learning (RL), Robotics (ROB), and Quantum ML (QML).

Consider the distribution of scholars across the first three tracks: CV, NLP, and RL:

  • Every scholar chose exactly 3 tracks out of 5, which guarantees that every scholar selected at least one of the three tracks {CV, NLP, RL}.

  • Let SS, DD, and TT denote the number of scholars who selected exactly one, exactly two, and all three of the tracks {CV, NLP, RL} respectively.

  • (I) The number of scholars selecting exactly two of these tracks was one-third the sum of those selecting exactly one and all three: 3D=S+T3D = S + T.

  • In the 3-set Venn diagram:

    • Let bb, cc, and dd denote scholars choosing CV only, NLP only, and RL only respectively.

    • Let pp, qq, and rr denote scholars choosing {CV and NLP only}, {NLP and RL only}, and {CV and RL only} respectively.

    • Let xx denote scholars choosing all three tracks {CV, NLP, and RL}.

  • (II) 40≤q≤5640 \le q \le 56

  • (III) c>bc > b

  • (IV) x≥15x \ge 15

  • (V) 24≤p≤4024 \le p \le 40

  • (VI) Total scholars choosing CV was 4 less than total scholars choosing RL: Total(CV)=Total(RL)−4\text{Total}(CV) = \text{Total}(RL) - 4.

  • (VII) The number of scholars choosing RL only was 6 times the number of scholars choosing all three: d=6xd = 6x.

The maximum number of scholars who could have chosen Computer Vision (CV) is:

  1. A

    198

  2. B

    211

  3. C

    204

  4. D

    217

Show answer and explanation

Answer: (B) 211

Question 31 Solution:

Total scholars choosing CV:

Total(CV)=b+p+r+x\text{Total}(CV) = b + p + r + x

=(6x+12)+(24+y)+(56−2y)+x=7x+92−y= (6x + 12) + (24 + y) + (56 - 2y) + x = 7x + 92 - y

To maximize Total(CV)\text{Total}(CV), we maximize xx and minimize yy:

  • Maximum x=17x = 17

  • Minimum y=0y = 0

Max Total(CV)=7(17)+92−0=119+92=211\text{Max Total}(CV) = 7(17) + 92 - 0 = 119 + 92 = 211

Therefore, the maximum number of scholars who could choose CV is 211.

Question 31 · CAT Level

Directions for Questions 30 to 33

Answer the questions on the basis of the information given below.

In an advanced AI Doctoral Fellowship program comprising 480 research scholars, each scholar was required to choose exactly 3 specialization tracks out of 5 available tracks: Computer Vision (CV), Natural Language Processing (NLP), Reinforcement Learning (RL), Robotics (ROB), and Quantum ML (QML).

Consider the distribution of scholars across the first three tracks: CV, NLP, and RL:

  • Every scholar chose exactly 3 tracks out of 5, which guarantees that every scholar selected at least one of the three tracks {CV, NLP, RL}.

  • Let SS, DD, and TT denote the number of scholars who selected exactly one, exactly two, and all three of the tracks {CV, NLP, RL} respectively.

  • (I) The number of scholars selecting exactly two of these tracks was one-third the sum of those selecting exactly one and all three: 3D=S+T3D = S + T.

  • In the 3-set Venn diagram:

    • Let bb, cc, and dd denote scholars choosing CV only, NLP only, and RL only respectively.

    • Let pp, qq, and rr denote scholars choosing {CV and NLP only}, {NLP and RL only}, and {CV and RL only} respectively.

    • Let xx denote scholars choosing all three tracks {CV, NLP, and RL}.

  • (II) 40≤q≤5640 \le q \le 56

  • (III) c>bc > b

  • (IV) x≥15x \ge 15

  • (V) 24≤p≤4024 \le p \le 40

  • (VI) Total scholars choosing CV was 4 less than total scholars choosing RL: Total(CV)=Total(RL)−4\text{Total}(CV) = \text{Total}(RL) - 4.

  • (VII) The number of scholars choosing RL only was 6 times the number of scholars choosing all three: d=6xd = 6x.

The minimum possible value of (q + r) is:

  1. A

    72

  2. B

    88

  3. C

    84

  4. D

    80

Show answer and explanation

Answer: (D) 80

Question 32 Solution:

From the parameterization:

q=40+yandr=56−2yq = 40 + y \quad \text{and} \quad r = 56 - 2y

q+r=(40+y)+(56−2y)=96−yq + r = (40 + y) + (56 - 2y) = 96 - y

To minimize q+rq + r, we maximize yy:

  • The maximum allowable value of yy is 16 (since 0≤y≤160 \le y \le 16).

Min(q+r)=96−16=80\text{Min}(q + r) = 96 - 16 = 80

Therefore, the minimum value of (q+r)(q + r) is 80.

Question 32 · CAT Level

Directions for Questions 30 to 33

Answer the questions on the basis of the information given below.

In an advanced AI Doctoral Fellowship program comprising 480 research scholars, each scholar was required to choose exactly 3 specialization tracks out of 5 available tracks: Computer Vision (CV), Natural Language Processing (NLP), Reinforcement Learning (RL), Robotics (ROB), and Quantum ML (QML).

Consider the distribution of scholars across the first three tracks: CV, NLP, and RL:

  • Every scholar chose exactly 3 tracks out of 5, which guarantees that every scholar selected at least one of the three tracks {CV, NLP, RL}.

  • Let SS, DD, and TT denote the number of scholars who selected exactly one, exactly two, and all three of the tracks {CV, NLP, RL} respectively.

  • (I) The number of scholars selecting exactly two of these tracks was one-third the sum of those selecting exactly one and all three: 3D=S+T3D = S + T.

  • In the 3-set Venn diagram:

    • Let bb, cc, and dd denote scholars choosing CV only, NLP only, and RL only respectively.

    • Let pp, qq, and rr denote scholars choosing {CV and NLP only}, {NLP and RL only}, and {CV and RL only} respectively.

    • Let xx denote scholars choosing all three tracks {CV, NLP, and RL}.

  • (II) 40≤q≤5640 \le q \le 56

  • (III) c>bc > b

  • (IV) x≥15x \ge 15

  • (V) 24≤p≤4024 \le p \le 40

  • (VI) Total scholars choosing CV was 4 less than total scholars choosing RL: Total(CV)=Total(RL)−4\text{Total}(CV) = \text{Total}(RL) - 4.

  • (VII) The number of scholars choosing RL only was 6 times the number of scholars choosing all three: d=6xd = 6x.

How many scholars selected at least one of the tracks out of Computer Vision (CV), Natural Language Processing (NLP), and Reinforcement Learning (RL)?

  1. A

    360

  2. B

    480

  3. C

    420

  4. D

    Cannot be determined

Show answer and explanation

Answer: (B) 480

Question 33 Solution:

Every scholar in the 480-member cohort must select exactly 3 tracks out of 5. By the Pigeonhole Principle, it is impossible for any scholar to choose 3 tracks from the remaining 2 tracks {Robotics, Quantum ML}.

Thus, every single scholar must select at least one track from {CV, NLP, RL}.

Therefore, the required number is 480.

Question 33 · CAT Level

Directions for Questions 34 to 37

Answer the questions on the basis of the information given below.

Five General Partners of a venture capital fund — GP-1, GP-2, GP-3, GP-4, and GP-5 — evaluated startup pitches, awarding performance tiers: Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V.

Under the fund's scoring rubric, Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V fetch points 7, 5, 4, 2, and 0 respectively. The Grade-Average for a startup is defined as the arithmetic mean of the points awarded by the five partners.

The table below shows partial grading records for several startups:

StartupGeneral PartnersGrade Average
GP-1GP-2GP-3GP-4GP-5
AlphaTier-I  Tier-IV  
Beta Tier-IITier-IVTier-V  
Gamma     2.8
Delta Tier-V  Tier-I 
Epsilon  Tier-IVTier-II 3.4
ZetaTier-I    4.0
Eta  Tier-ITier-IV  
Theta Tier-V Tier-II  
IotaTier-I    4.2
Kappa     2.8
Lambda Tier-II    
Mu Tier-IITier-II   
Nu   Tier-III  
Xi     3.0
Omicron  Tier-IITier-I  

Additional Governance Constraints:

(i) No partner gave Tier-V to Iota or Epsilon. Iota’s Grade-Average was the highest overall.

(ii) The Grade-Average of Alpha was better than that of Epsilon (3.4) but worse than that of Zeta (4.0).

(iii) No partner gave Tier-I or Tier-V to Lambda.

(iv) GP-3 and GP-5 gave the same tier to Theta and Nu. Theta received the same tier from GP-3 and GP-5.

(v) No startup received Tier-V from more than one partner.

(vi) Three partners gave the same tier to Delta.

(vii) No three partners gave the same tier to Lambda. The same was true for Nu.

If the Grade Average of Lambda was better than that of Alpha, and all the partners, except one, gave a higher tier to Lambda than that to Alpha, what was the tier given by GP-1 to Lambda?

  1. A

    Tier-III

  2. B

    Tier-IV

  3. C

    Tier-II

  4. D

    Cannot be determined

Show answer and explanation

Answer: (B) Tier-IV

Question 34 Solution:

From Constraint (ii), Grade-Average of Alpha is between 3.4 (Total 17 points) and 4.0 (Total 20 points). Thus, Alpha’s total points must be 18 or 19.

  • Alpha has GP-1 = Tier-I (7 points) and GP-4 = Tier-IV (2 points). Known sum = 9.

  • Remaining 3 partners (GP-2, GP-3, GP-5) must sum to 9 or 10.

For Lambda:

  • GP-2 awarded Tier-II (5 points).

  • From Constraint (iii), Lambda received neither Tier-I (7) nor Tier-V (0). Hence, Lambda's grades can only be from {5,4,2}\{5, 4, 2\}.

  • If Lambda's Grade Average is strictly higher than Alpha's, and all partners except one gave Lambda a higher tier than Alpha:

    • Since Alpha got Tier-I (7) from GP-1, and Lambda cannot get Tier-I, GP-1 is the only partner who gave Alpha a higher tier than Lambda!

    • For all other four partners, Lambda's points must exceed Alpha's points.

    • This requires Lambda's total points to be 20 and Alpha's total points to be 19.

Under this unique deduction:

  • Alpha's points: GP-1 = 7, GP-2 = 2/4, GP-3 = 2, GP-4 = 4, GP-5 = 4/2 (Total = 19).

  • Lambda's points: GP-1 = 2 (Tier-IV), GP-2 = 5, GP-3 = 4, GP-4 = 5, GP-5 = 4 (Total = 20).

Since 2 points corresponds to Tier-IV, GP-1 awarded Tier-IV to Lambda.

Question 34 · CAT Level

Directions for Questions 34 to 37

Answer the questions on the basis of the information given below.

Five General Partners of a venture capital fund — GP-1, GP-2, GP-3, GP-4, and GP-5 — evaluated startup pitches, awarding performance tiers: Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V.

Under the fund's scoring rubric, Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V fetch points 7, 5, 4, 2, and 0 respectively. The Grade-Average for a startup is defined as the arithmetic mean of the points awarded by the five partners.

The table below shows partial grading records for several startups:

StartupGeneral PartnersGrade Average
GP-1GP-2GP-3GP-4GP-5
AlphaTier-I  Tier-IV  
Beta Tier-IITier-IVTier-V  
Gamma     2.8
Delta Tier-V  Tier-I 
Epsilon  Tier-IVTier-II 3.4
ZetaTier-I    4.0
Eta  Tier-ITier-IV  
Theta Tier-V Tier-II  
IotaTier-I    4.2
Kappa     2.8
Lambda Tier-II    
Mu Tier-IITier-II   
Nu   Tier-III  
Xi     3.0
Omicron  Tier-IITier-I  

Additional Governance Constraints:

(i) No partner gave Tier-V to Iota or Epsilon. Iota’s Grade-Average was the highest overall.

(ii) The Grade-Average of Alpha was better than that of Epsilon (3.4) but worse than that of Zeta (4.0).

(iii) No partner gave Tier-I or Tier-V to Lambda.

(iv) GP-3 and GP-5 gave the same tier to Theta and Nu. Theta received the same tier from GP-3 and GP-5.

(v) No startup received Tier-V from more than one partner.

(vi) Three partners gave the same tier to Delta.

(vii) No three partners gave the same tier to Lambda. The same was true for Nu.

If the Grade Average of Alpha and Theta was the same, then how many different values of Grade Average were possible for Nu?

  1. A

    4

  2. B

    2

  3. C

    3

  4. D

    5

Show answer and explanation

Answer: (C) 3

Question 35 Solution:

For Alpha and Theta to have the same Grade-Average, their total points must be equal (19 points):

  • Theta: GP-2 = Tier-V (0 points), GP-4 = Tier-II (5 points). From Constraint (iv), GP-3 and GP-5 awarded the same tier to Theta: (5,5)(5, 5). GP-1 must award Tier-II (5 points) or Tier-IV (4 points), making Theta's total 19.

  • For Nu, GP-4 gave Tier-III (4 points), and from Constraint (iv), GP-3 and GP-5 gave Tier-II (5 points each).

  • The remaining two partners (GP-1 and GP-2) can assign valid non-violating point pairs (2,0)(2, 0), (2,2)(2, 2), (4,0)(4, 0), or (4,2)(4, 2).

  • These correspond to three distinct possible point totals for Nu: 16, 18, and 20 (Grade Averages 3.2, 3.6, and 4.0).

Thus, exactly 3 different values of Grade Average were possible for Nu.

Question 35 · CAT Level

Directions for Questions 34 to 37

Answer the questions on the basis of the information given below.

Five General Partners of a venture capital fund — GP-1, GP-2, GP-3, GP-4, and GP-5 — evaluated startup pitches, awarding performance tiers: Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V.

Under the fund's scoring rubric, Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V fetch points 7, 5, 4, 2, and 0 respectively. The Grade-Average for a startup is defined as the arithmetic mean of the points awarded by the five partners.

The table below shows partial grading records for several startups:

StartupGeneral PartnersGrade Average
GP-1GP-2GP-3GP-4GP-5
AlphaTier-I  Tier-IV  
Beta Tier-IITier-IVTier-V  
Gamma     2.8
Delta Tier-V  Tier-I 
Epsilon  Tier-IVTier-II 3.4
ZetaTier-I    4.0
Eta  Tier-ITier-IV  
Theta Tier-V Tier-II  
IotaTier-I    4.2
Kappa     2.8
Lambda Tier-II    
Mu Tier-IITier-II   
Nu   Tier-III  
Xi     3.0
Omicron  Tier-IITier-I  

Additional Governance Constraints:

(i) No partner gave Tier-V to Iota or Epsilon. Iota’s Grade-Average was the highest overall.

(ii) The Grade-Average of Alpha was better than that of Epsilon (3.4) but worse than that of Zeta (4.0).

(iii) No partner gave Tier-I or Tier-V to Lambda.

(iv) GP-3 and GP-5 gave the same tier to Theta and Nu. Theta received the same tier from GP-3 and GP-5.

(v) No startup received Tier-V from more than one partner.

(vi) Three partners gave the same tier to Delta.

(vii) No three partners gave the same tier to Lambda. The same was true for Nu.

If GP-3 gave a higher tier than GP-5 to both Iota and Epsilon, the sum of points given by GP-3 to Delta, Iota, and Epsilon can take how many different values?

  1. A

    5

  2. B

    6

  3. C

    4

  4. D

    9

Show answer and explanation

Answer: (C) 4

Question 36 Solution:

Evaluating the possible points awarded by GP-3:

  • For Epsilon (Total 17): GP-3 can award 4 points.

  • For Iota (Total 21, highest average): GP-3 can award 4 or 5 points.

  • For Delta (from Constraint vi, 3 partners gave identical tier): GP-3 can award 2 or 4 points.

Combining these valid configurations yields exactly 4 distinct sum totals for GP-3's points to Delta, Iota, and Epsilon.

Question 36 · CAT Level

Directions for Questions 34 to 37

Answer the questions on the basis of the information given below.

Five General Partners of a venture capital fund — GP-1, GP-2, GP-3, GP-4, and GP-5 — evaluated startup pitches, awarding performance tiers: Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V.

Under the fund's scoring rubric, Tier-I, Tier-II, Tier-III, Tier-IV, and Tier-V fetch points 7, 5, 4, 2, and 0 respectively. The Grade-Average for a startup is defined as the arithmetic mean of the points awarded by the five partners.

The table below shows partial grading records for several startups:

StartupGeneral PartnersGrade Average
GP-1GP-2GP-3GP-4GP-5
AlphaTier-I  Tier-IV  
Beta Tier-IITier-IVTier-V  
Gamma     2.8
Delta Tier-V  Tier-I 
Epsilon  Tier-IVTier-II 3.4
ZetaTier-I    4.0
Eta  Tier-ITier-IV  
Theta Tier-V Tier-II  
IotaTier-I    4.2
Kappa     2.8
Lambda Tier-II    
Mu Tier-IITier-II   
Nu   Tier-III  
Xi     3.0
Omicron  Tier-IITier-I  

Additional Governance Constraints:

(i) No partner gave Tier-V to Iota or Epsilon. Iota’s Grade-Average was the highest overall.

(ii) The Grade-Average of Alpha was better than that of Epsilon (3.4) but worse than that of Zeta (4.0).

(iii) No partner gave Tier-I or Tier-V to Lambda.

(iv) GP-3 and GP-5 gave the same tier to Theta and Nu. Theta received the same tier from GP-3 and GP-5.

(v) No startup received Tier-V from more than one partner.

(vi) Three partners gave the same tier to Delta.

(vii) No three partners gave the same tier to Lambda. The same was true for Nu.

How many different values of Grade Average are possible for Lambda?

  1. A

    3

  2. B

    4

  3. C

    5

  4. D

    6

Show answer and explanation

Answer: (A) 3

Question 37 Solution:

For Lambda:

  • No partner gave Tier-I (7) or Tier-V (0).

  • GP-2 gave Tier-II (5 points).

  • No three partners gave the same tier to Lambda (Constraint vii).

The valid distributions of points {x1,x2,x3,x4,x5}\{x_1, x_2, x_3, x_4, x_5\} drawn from {5,4,2}\{5, 4, 2\} without any rating appearing three or more times yield totals of 18, 19, and 20 (Grade Averages 3.6, 3.8, and 4.0).

Hence, exactly 3 different values of Grade Average are possible for Lambda.

Question 37 · CAT Level

Directions for Questions 38 to 41

Answer the questions on the basis of the information given below.

A developer community survey investigated the active usage of four developer tools and infrastructure platforms: Docker, Kubernetes, Python, and Linux.

The initial survey findings reported:

  • 120 developers actively use Docker.

  • 150 developers actively use Kubernetes.

  • 200 developers actively use Python.

  • 170 developers actively use Linux.

  • 20 developers use Docker and Python.

  • 25 developers use Docker and Kubernetes.

  • 25 developers use Linux and Python.

  • 20 developers use Linux and Kubernetes.

  • 10 developers use Docker and Linux.

  • 30 developers use Python and Kubernetes.

If no developer uses more than two tools, then how many developers use Kubernetes or Python but do not use Linux or Docker?

  1. A

    230

  2. B

    245

  3. C

    225

  4. D

    270

Show answer and explanation

Answer: (A) 230

Question 38 Solution:

When no developer uses more than two tools, all 3-set and 4-set intersections are 0.

  • Developers using Kubernetes only = 150−(25+20+30)=75150 - (25 + 20 + 30) = 75

  • Developers using Python only = 200−(20+25+30)=125200 - (20 + 25 + 30) = 125

  • Developers using both Kubernetes and Python (and neither Docker nor Linux) = 30

Total=75+125+30=230\text{Total} = 75 + 125 + 30 = 230

Alternatively:

Total=∣Kubernetes∪Python∣−(Overlaps with Docker or Linux)\text{Total} = |\text{Kubernetes} \cup \text{Python}| - (\text{Overlaps with Docker or Linux})

=150+200−30−(25+20+20+25)=320−90=230= 150 + 200 - 30 - (25 + 20 + 20 + 25) = 320 - 90 = 230

Therefore, the number of developers is 230.

Question 38 · CAT Level

Directions for Questions 38 to 41

Answer the questions on the basis of the information given below.

A developer community survey investigated the active usage of four developer tools and infrastructure platforms: Docker, Kubernetes, Python, and Linux.

The initial survey findings reported:

  • 120 developers actively use Docker.

  • 150 developers actively use Kubernetes.

  • 200 developers actively use Python.

  • 170 developers actively use Linux.

  • 20 developers use Docker and Python.

  • 25 developers use Docker and Kubernetes.

  • 25 developers use Linux and Python.

  • 20 developers use Linux and Kubernetes.

  • 10 developers use Docker and Linux.

  • 30 developers use Python and Kubernetes.

If no developer uses more than two tools, then how many developers use at least one out of Kubernetes, Linux, and Docker, but not Python?

  1. A

    370

  2. B

    410

  3. C

    400

  4. D

    310

Show answer and explanation

Answer: (D) 310

Question 39 Solution:

We sum the regions inside (Docker ∪\cup Kubernetes ∪\cup Linux) that do not include Python:

  • Docker only = 120−(20+25+10)=65120 - (20 + 25 + 10) = 65

  • Kubernetes only = 150−(25+20+30)=75150 - (25 + 20 + 30) = 75

  • Linux only = 170−(25+20+10)=115170 - (25 + 20 + 10) = 115

  • Docker & Kubernetes only = 25

  • Docker & Linux only = 10

  • Linux & Kubernetes only = 20

Total=65+75+115+25+10+20=310\text{Total} = 65 + 75 + 115 + 25 + 10 + 20 = 310

Therefore, the number of developers is 310.

Question 39 · CAT Level

Directions for Questions 38 to 41

Answer the questions on the basis of the information given below.

A developer community survey investigated the active usage of four developer tools and infrastructure platforms: Docker, Kubernetes, Python, and Linux.

The initial survey findings reported:

  • 120 developers actively use Docker.

  • 150 developers actively use Kubernetes.

  • 200 developers actively use Python.

  • 170 developers actively use Linux.

  • 20 developers use Docker and Python.

  • 25 developers use Docker and Kubernetes.

  • 25 developers use Linux and Python.

  • 20 developers use Linux and Kubernetes.

  • 10 developers use Docker and Linux.

  • 30 developers use Python and Kubernetes.

It is also known that the number of developers who use exactly three tools is double the number of developers who use all four tools. The number of developers who use any three tools is the same for all possible combinations of three tools out of the four. Find the maximum number of developers who can use all four tools given that all developer counts are integers.

  1. A

    5

  2. B

    4

  3. C

    2

  4. D

    None of these

Show answer and explanation

Answer: (B) 4

Question 40 Solution:

Let xx be the number of developers who use all four tools.

  • There are (43)=4\binom{4}{3} = 4 three-tool combinations.

  • Each 3-tool intersection region has value t=2x/4=x/2t = 2x / 4 = x/2.

  • For the intersection of Docker and Linux, the pairwise total is 10:

    Docker∩Linux=(Docker∩Linux only)+2(x/2)+x=(Docker∩Linux only)+2x=10\text{Docker} \cap \text{Linux} = (\text{Docker} \cap \text{Linux only}) + 2(x/2) + x = (\text{Docker} \cap \text{Linux only}) + 2x = 10

      ⟹  10−2x≥0  ⟹  x≤5\implies 10 - 2x \ge 0 \implies x \le 5

  • For x/2x/2 to be an integer, xx must be an even integer.

  • Thus, the maximum integer value for xx is 4.

Question 40 · CAT Level

Directions for Questions 38 to 41

Answer the questions on the basis of the information given below.

A developer community survey investigated the active usage of four developer tools and infrastructure platforms: Docker, Kubernetes, Python, and Linux.

The initial survey findings reported:

  • 120 developers actively use Docker.

  • 150 developers actively use Kubernetes.

  • 200 developers actively use Python.

  • 170 developers actively use Linux.

  • 20 developers use Docker and Python.

  • 25 developers use Docker and Kubernetes.

  • 25 developers use Linux and Python.

  • 20 developers use Linux and Kubernetes.

  • 10 developers use Docker and Linux.

  • 30 developers use Python and Kubernetes.

Under the conditions of Question 40, if 79 developers use Docker only, find the number of developers who use Python only.

  1. A

    129

  2. B

    139

  3. C

    166

  4. D

    183

Show answer and explanation

Answer: (B) 139

Question 41 Solution:

From Question 40, x=4x = 4, which gives x/2=2x/2 = 2.

Let us verify with Docker only:

Docker only=120−[(25−2x)+(20−2x)+(10−2x)+3(x/2)+x]\text{Docker only} = 120 - [(25 - 2x) + (20 - 2x) + (10 - 2x) + 3(x/2) + x]

=120−[55−6x+2.5x]=120−(55−3.5x)=65+3.5(4)=65+14=79= 120 - [55 - 6x + 2.5x] = 120 - (55 - 3.5x) = 65 + 3.5(4) = 65 + 14 = 79

This confirms x=4x = 4.

Now compute Python only:

  • Python total = 200

  • Dual overlaps involving Python: (Docker & Python = 20), (Linux & Python = 25), (Kubernetes & Python = 30)

Python only=200−[(20−2x)+(25−2x)+(30−2x)+3(x/2)+x]\text{Python only} = 200 - [(20 - 2x) + (25 - 2x) + (30 - 2x) + 3(x/2) + x]

=200−[75−6x+2.5x]=200−(75−3.5x)=125+3.5(4)=125+14=139= 200 - [75 - 6x + 2.5x] = 200 - (75 - 3.5x) = 125 + 3.5(4) = 125 + 14 = 139

Therefore, the number of developers using Python only is 139.

Question 41 · CAT Level

Directions for Questions 42 to 45

Answer the questions on the basis of the information given below.

Three dual-bot high-frequency trading desks — Desk-A (G1), Desk-B (G2), and Desk-C (G3) — competed in an execution speed benchmark. Each desk fielded exactly two execution bots.

In each of two rounds, all six bots recorded discrete latency benchmark tiers from 1 to 6 microseconds (with no two bots recording the same value in any round). The Absolute Difference (AD) between the latency tiers of the two bots in each desk was computed.

  • The desk with the lowest AD won 30 points (Winner).

  • The desk with the second lowest AD won 20 points (Runner-up).

  • The desk with the highest AD won 10 points (Third place).

DeskPoints in Round 1Points in Round 2
Desk-A (G1)3030
Desk-B (G2)20—
Desk-C (G3)10—

Additional Performance Constraints:

(i) The sum of ADs across all three desks in Round 1 was less than that in Round 2.

(ii) Out of the 4 latency values recorded by the two bots of Desk-C across the two rounds, at least 2 were odd.

(iii) The sum of the 4 latency values recorded by Desk-A across the two rounds is not a prime number.

(iv) No two bots obtained the same latency in any round, and no two desks had the same AD in any round.

If 3 bots recorded the exact same latency value in both rounds, what was the latency value of the bot in Desk-B that recorded the lowest latency in Round 1?

  1. A

    1

  2. B

    4

  3. C

    2

  4. D

    Cannot be determined

Show answer and explanation

Answer: (B) 4

Question 42 Solution:

In Round 1, the rankings were Desk-A (30 pts, AD=1), Desk-B (20 pts, AD=2), Desk-C (10 pts, AD=4):

  • Round 1 pairs: Desk-A = (2,3)(2, 3), Desk-B = (4,6)(4, 6), Desk-C = (1,5)(1, 5).

  • Round 2 pairs: Desk-A = (3,4)(3, 4) [AD=1], Desk-B = (1,6)(1, 6) [AD=5], Desk-C = (2,5)(2, 5) [AD=3].

Notice the bots with identical latency in both rounds:

  • Bot with latency 3 (Desk-A in R1 and R2)

  • Bot with latency 5 (Desk-C in R1 and R2)

  • Bot with latency 6 (Desk-B in R1 and R2)

In Round 1, Desk-B's bots recorded latencies 4 and 6. The bot with the lowest value in Desk-B recorded 4.

Therefore, the answer is 4.

Question 42 · CAT Level

Directions for Questions 42 to 45

Answer the questions on the basis of the information given below.

Three dual-bot high-frequency trading desks — Desk-A (G1), Desk-B (G2), and Desk-C (G3) — competed in an execution speed benchmark. Each desk fielded exactly two execution bots.

In each of two rounds, all six bots recorded discrete latency benchmark tiers from 1 to 6 microseconds (with no two bots recording the same value in any round). The Absolute Difference (AD) between the latency tiers of the two bots in each desk was computed.

  • The desk with the lowest AD won 30 points (Winner).

  • The desk with the second lowest AD won 20 points (Runner-up).

  • The desk with the highest AD won 10 points (Third place).

DeskPoints in Round 1Points in Round 2
Desk-A (G1)3030
Desk-B (G2)20—
Desk-C (G3)10—

Additional Performance Constraints:

(i) The sum of ADs across all three desks in Round 1 was less than that in Round 2.

(ii) Out of the 4 latency values recorded by the two bots of Desk-C across the two rounds, at least 2 were odd.

(iii) The sum of the 4 latency values recorded by Desk-A across the two rounds is not a prime number.

(iv) No two bots obtained the same latency in any round, and no two desks had the same AD in any round.

If the same 3 bots from the 3 desks recorded the lower latency values in their respective desks in both rounds, what is the minimum possible value of the sum of the two latency values across the two rounds for any of these 3 bots?

  1. A

    5

  2. B

    4

  3. C

    3

  4. D

    2

Show answer and explanation

Answer: (D) 2

Question 43 Solution:

Consider the valid configuration:

  • Desk-A: Round 1 = (2,3)(2, 3), Round 2 = (3,4)(3, 4)

  • Desk-B: Round 1 = (4,6)(4, 6), Round 2 = (2,5)(2, 5)

  • Desk-C: Round 1 = (1,5)(1, 5), Round 2 = (1,6)(1, 6)

In Desk-C, the same bot records latency 1 in Round 1 and latency 1 in Round 2. Thus, the sum of its two values across both rounds is:

1+1=21 + 1 = 2

Therefore, the minimum possible sum is 2.

Question 43 · CAT Level

Directions for Questions 42 to 45

Answer the questions on the basis of the information given below.

Three dual-bot high-frequency trading desks — Desk-A (G1), Desk-B (G2), and Desk-C (G3) — competed in an execution speed benchmark. Each desk fielded exactly two execution bots.

In each of two rounds, all six bots recorded discrete latency benchmark tiers from 1 to 6 microseconds (with no two bots recording the same value in any round). The Absolute Difference (AD) between the latency tiers of the two bots in each desk was computed.

  • The desk with the lowest AD won 30 points (Winner).

  • The desk with the second lowest AD won 20 points (Runner-up).

  • The desk with the highest AD won 10 points (Third place).

DeskPoints in Round 1Points in Round 2
Desk-A (G1)3030
Desk-B (G2)20—
Desk-C (G3)10—

Additional Performance Constraints:

(i) The sum of ADs across all three desks in Round 1 was less than that in Round 2.

(ii) Out of the 4 latency values recorded by the two bots of Desk-C across the two rounds, at least 2 were odd.

(iii) The sum of the 4 latency values recorded by Desk-A across the two rounds is not a prime number.

(iv) No two bots obtained the same latency in any round, and no two desks had the same AD in any round.

For which desk could no bot obtain a latency value of 4 in either of the two rounds across all possible cases?

  1. A

    Desk-C

  2. B

    Desk-B

  3. C

    Desk-A

  4. D

    Not possible

Show answer and explanation

Answer: (A) Desk-C

Question 44 Solution:

Across all admissible integer partition pairings satisfying Constraints (i) through (iv):

  • Desk-A uses 4 in Round 2: (3,4)(3, 4) or (4,5)(4, 5).

  • Desk-B uses 4 in Round 1: (4,6)(4, 6) or (2,4)(2, 4).

  • Desk-C strictly pairs (1,5)(1, 5) or (1,6)(1, 6) or (2,5)(2, 5) or (2,6)(2, 6) to satisfy the required large AD and odd parity constraints, never receiving 4.

Thus, no bot in Desk-C could obtain a latency value of 4 in either round.

Question 44 · CAT Level

Directions for Questions 42 to 45

Answer the questions on the basis of the information given below.

Three dual-bot high-frequency trading desks — Desk-A (G1), Desk-B (G2), and Desk-C (G3) — competed in an execution speed benchmark. Each desk fielded exactly two execution bots.

In each of two rounds, all six bots recorded discrete latency benchmark tiers from 1 to 6 microseconds (with no two bots recording the same value in any round). The Absolute Difference (AD) between the latency tiers of the two bots in each desk was computed.

  • The desk with the lowest AD won 30 points (Winner).

  • The desk with the second lowest AD won 20 points (Runner-up).

  • The desk with the highest AD won 10 points (Third place).

DeskPoints in Round 1Points in Round 2
Desk-A (G1)3030
Desk-B (G2)20—
Desk-C (G3)10—

Additional Performance Constraints:

(i) The sum of ADs across all three desks in Round 1 was less than that in Round 2.

(ii) Out of the 4 latency values recorded by the two bots of Desk-C across the two rounds, at least 2 were odd.

(iii) The sum of the 4 latency values recorded by Desk-A across the two rounds is not a prime number.

(iv) No two bots obtained the same latency in any round, and no two desks had the same AD in any round.

Which desk could definitely not obtain the same Absolute Difference (AD) in both rounds?

  1. A

    Desk-A

  2. B

    Desk-B

  3. C

    Desk-C

  4. D

    More than one of them

Show answer and explanation

Answer: (B) Desk-B

Question 45 Solution:

  • Desk-A achieves AD = 1 in both rounds (winning both rounds with 30 points each).

  • Desk-C can achieve AD = 4 in both rounds.

  • However, Desk-B must have AD = 2 in Round 1, while in Round 2 its AD must be either 3 or 5 (as Round 2 sum of ADs exceeds Round 1 sum of ADs, and Desk-A has AD = 1). Thus, Desk-B cannot have AD = 2 in Round 2.

Therefore, Desk-B could definitely not get the same AD in both rounds.

Question 45 · CAT Level

Directions for Questions 46 to 49

Answer the questions on the basis of the information given below.

A global medical research consortium oversees clinical trials involving 300 affiliated hospitals evaluating three experimental vaccine formulations: Formulation V-Alpha, Formulation V-Beta, and Formulation V-Gamma.

Participation data reveals:

  • 70 hospitals conduct trials for Formulation V-Alpha.

  • 85 hospitals conduct trials for Formulation V-Beta.

  • 90 hospitals conduct trials for Formulation V-Gamma.

  • Exactly 45 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Beta.

  • Exactly 40 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Gamma.

  • Exactly 60 hospitals conduct trials for both Formulation V-Beta and Formulation V-Gamma.

If the number of hospitals conducting trials for all three formulations is the arithmetic mean of its maximum possible value and minimum possible value, how many hospitals in the consortium do not conduct trials for any of the three formulations?

  1. A

    130

  2. B

    170

  3. C

    80

  4. D

    Data insufficient

Show answer and explanation

Answer: (B) 170

Question 46 Solution:

Let xx denote the number of hospitals conducting trials for all three formulations.

  • Number of hospitals conducting trials for V-Alpha only = 70−(45−x+40−x+x)=x−15≥0  ⟹  x≥1570 - (45 - x + 40 - x + x) = x - 15 \ge 0 \implies x \ge 15

  • Number of hospitals conducting trials for V-Beta only = 85−(45−x+60−x+x)=x−20≥0  ⟹  x≥2085 - (45 - x + 60 - x + x) = x - 20 \ge 0 \implies x \ge 20

  • Number of hospitals conducting trials for V-Gamma only = 90−(40−x+60−x+x)=x−10≥0  ⟹  x≥1090 - (40 - x + 60 - x + x) = x - 10 \ge 0 \implies x \ge 10

Also, xx cannot exceed the minimum pairwise overlap: x≤min⁡(45,40,60)=40x \le \min(45, 40, 60) = 40.

Thus:

xmin⁡=20andxmax⁡=40x_{\min} = 20 \quad \text{and} \quad x_{\max} = 40

Mean of x=20+402=30\text{Mean of } x = \frac{20 + 40}{2} = 30

Total hospitals in at least one trial:

∣A∪B∪C∣=70+85+90−(45+40+60)+x=100+x|A \cup B \cup C| = 70 + 85 + 90 - (45 + 40 + 60) + x = 100 + x

For x=30x = 30, ∣A∪B∪C∣=100+30=130|A \cup B \cup C| = 100 + 30 = 130.

Number of hospitals not conducting any trial:

None=300−130=170\text{None} = 300 - 130 = 170

Therefore, the number of hospitals is 170.

Question 46 · CAT Level

Directions for Questions 46 to 49

Answer the questions on the basis of the information given below.

A global medical research consortium oversees clinical trials involving 300 affiliated hospitals evaluating three experimental vaccine formulations: Formulation V-Alpha, Formulation V-Beta, and Formulation V-Gamma.

Participation data reveals:

  • 70 hospitals conduct trials for Formulation V-Alpha.

  • 85 hospitals conduct trials for Formulation V-Beta.

  • 90 hospitals conduct trials for Formulation V-Gamma.

  • Exactly 45 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Beta.

  • Exactly 40 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Gamma.

  • Exactly 60 hospitals conduct trials for both Formulation V-Beta and Formulation V-Gamma.

If the number of hospitals conducting trials for all three formulations is at its maximum possible value, and the V-Alpha development committee levies a compliance audit fee of $10 from every hospital that does not participate in V-Alpha trials, what is the total amount collected in audit fees?

  1. A
  2. B

    $2,300

  3. C

    $1,600

  4. D
Show answer and explanation

Answer: (B) $2,300

Question 47 Solution:

Total hospitals in the consortium = 300.

Hospitals participating in V-Alpha trials = 70.

Hospitals that do not participate in V-Alpha trials:

Non-participants=300−70=230\text{Non-participants} = 300 - 70 = 230

Total audit fees collected:

Fees=230×$10=$2,300\text{Fees} = 230 \times \text{\textdollar}10 = \text{\textdollar}2{,}300

Therefore, the total amount collected is $2,300.

Question 47 · CAT Level

Directions for Questions 46 to 49

Answer the questions on the basis of the information given below.

A global medical research consortium oversees clinical trials involving 300 affiliated hospitals evaluating three experimental vaccine formulations: Formulation V-Alpha, Formulation V-Beta, and Formulation V-Gamma.

Participation data reveals:

  • 70 hospitals conduct trials for Formulation V-Alpha.

  • 85 hospitals conduct trials for Formulation V-Beta.

  • 90 hospitals conduct trials for Formulation V-Gamma.

  • Exactly 45 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Beta.

  • Exactly 40 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Gamma.

  • Exactly 60 hospitals conduct trials for both Formulation V-Beta and Formulation V-Gamma.

If the number of hospitals conducting trials for at least one of the three formulations is at its minimum possible value, what is the number of hospitals that conduct trials for Formulation V-Beta only?

  1. A

    65

  2. B

    20

  3. C

    0

  4. D

    55

Show answer and explanation

Answer: (C) 0

Question 48 Solution:

Total hospitals participating in at least one formulation:

∣A∪B∪C∣=100+x|A \cup B \cup C| = 100 + x

This value is minimized when xx takes its minimum allowable value:

x=xmin⁡=20x = x_{\min} = 20

The number of hospitals participating in V-Beta only is:

V-Beta only=x−20=20−20=0\text{V-Beta only} = x - 20 = 20 - 20 = 0

Therefore, the number of hospitals is 0.

Question 48 · CAT Level

Directions for Questions 46 to 49

Answer the questions on the basis of the information given below.

A global medical research consortium oversees clinical trials involving 300 affiliated hospitals evaluating three experimental vaccine formulations: Formulation V-Alpha, Formulation V-Beta, and Formulation V-Gamma.

Participation data reveals:

  • 70 hospitals conduct trials for Formulation V-Alpha.

  • 85 hospitals conduct trials for Formulation V-Beta.

  • 90 hospitals conduct trials for Formulation V-Gamma.

  • Exactly 45 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Beta.

  • Exactly 40 hospitals conduct trials for both Formulation V-Alpha and Formulation V-Gamma.

  • Exactly 60 hospitals conduct trials for both Formulation V-Beta and Formulation V-Gamma.

If the number of hospitals participating in none of the three formulation trials is at its maximum possible value, how many hospitals conduct trials for exactly two of the three formulations?

  1. A

    105

  2. B

    145

  3. C

    25

  4. D

    85

Show answer and explanation

Answer: (D) 85

Question 49 Solution:

Hospitals participating in none of the formulations:

None=300−(100+x)=200−x\text{None} = 300 - (100 + x) = 200 - x

This value is maximized when xx is minimized (x=20x = 20).

Hospitals participating in exactly two formulations:

Exactly two=(45−x)+(40−x)+(60−x)=145−3x\text{Exactly two} = (45 - x) + (40 - x) + (60 - x) = 145 - 3x

For x=20x = 20:

Exactly two=145−3(20)=145−60=85\text{Exactly two} = 145 - 3(20) = 145 - 60 = 85

Therefore, the number of hospitals is 85.

Question 49 · CAT Level

Directions for Questions 50 to 53

Answer the questions on the basis of the information given below.

Seven large language model architectures — GPT-P, BERT-S, LLaMA-V, Claude-G, Mistral-W, Gemini-M, and Falcon-Y — were evaluated in a standardized cloud GPU training benchmark.

The total GPU compute hours consumed by the seven models were 300, 240, 180, 160, 270, 120, and 150 hours (not necessarily in that order), and their token generation efficiencies (effective strike rates) were 33.3%, 60%, 30%, 40%, 80%, 45%, and 50% in some order.

Token Output (in billion tokens)=GPU Hours Faced×(Efficiency100)\text{Token Output (in billion tokens)} = \text{GPU Hours Faced} \times \left(\frac{\text{Efficiency}}{100}\right)

Benchmark Findings:

  1. No model produced 150 billion or more tokens, and the lowest token output produced by any model was 60 billion.

  2. GPT-P was the highest token producer overall, and either it consumed the maximum GPU hours (300) or had the highest efficiency (80%).

  3. The efficiency of one of GPT-P and BERT-S was twice that of the other.

  4. Total tokens produced by each model was distinct, except for LLaMA-V and Falcon-Y, who produced 60 billion tokens each.

  5. Gemini-M produced 81 billion tokens and Claude-G produced 108 billion tokens.

What was the token generation efficiency of BERT-S?

  1. A

    30%

  2. B

    40%

  3. C

    35%

  4. D

    45%

Show answer and explanation

Answer: (B) 40%

Question 50 Solution:

Step 1: Identifying Known Token Outputs

  • Gemini-M = 81B tokens: Can only be formed by 270×30%=81270 \times 30\% = 81 or 180×45%=81180 \times 45\% = 81.

  • Claude-G = 108B tokens: Can be formed by 270×40%=108270 \times 40\% = 108 or 240×45%=108240 \times 45\% = 108.

  • LLaMA-V and Falcon-Y both produced 60B tokens: The valid integer pairs yielding 60B are (120,50%)(120, 50\%), (180,33.3%)(180, 33.3\%), and (150,40%)(150, 40\%).

Step 2: Resolving GPT-P and BERT-S

  • If either LLaMA-V or Falcon-Y had an efficiency of 40%, then GPT-P and BERT-S efficiencies would have to be 30% and 60%. Then GPT-P (with 300 hours) would produce either 90B or 180B (violating bounds). Thus, LLaMA-V and Falcon-Y must use (120,50%)(120, 50\%) and (180,33.3%)(180, 33.3\%) in some order.

  • This leaves Gemini-M with (270,30%)(270, 30\%) and Claude-G with (240,45%)(240, 45\%).

  • For GPT-P and BERT-S, efficiencies must be a 2:1 pair from remaining {40%,80%}\{40\%, 80\%\}.

  • If GPT-P were 40%, BERT-S would be 80% and out-produce GPT-P, violating that GPT-P is highest.

  • Thus, GPT-P has efficiency 80% (with 160 hours, yielding 128B tokens), and BERT-S has efficiency 40% (with 300 hours, yielding 120B tokens).

Final Model Performance Matrix:

ModelGPU HoursEfficiencyToken Output (Billion)
GPT-P16080%128
BERT-S30040%120
Claude-G24045%108
Mistral-W15060%90
Gemini-M27030%81
LLaMA-V / Falcon-Y18033.3%60
Falcon-Y / LLaMA-V12050%60

Thus, the efficiency of BERT-S was 40%.

Question 50 · CAT Level

Directions for Questions 50 to 53

Answer the questions on the basis of the information given below.

Seven large language model architectures — GPT-P, BERT-S, LLaMA-V, Claude-G, Mistral-W, Gemini-M, and Falcon-Y — were evaluated in a standardized cloud GPU training benchmark.

The total GPU compute hours consumed by the seven models were 300, 240, 180, 160, 270, 120, and 150 hours (not necessarily in that order), and their token generation efficiencies (effective strike rates) were 33.3%, 60%, 30%, 40%, 80%, 45%, and 50% in some order.

Token Output (in billion tokens)=GPU Hours Faced×(Efficiency100)\text{Token Output (in billion tokens)} = \text{GPU Hours Faced} \times \left(\frac{\text{Efficiency}}{100}\right)

Benchmark Findings:

  1. No model produced 150 billion or more tokens, and the lowest token output produced by any model was 60 billion.

  2. GPT-P was the highest token producer overall, and either it consumed the maximum GPU hours (300) or had the highest efficiency (80%).

  3. The efficiency of one of GPT-P and BERT-S was twice that of the other.

  4. Total tokens produced by each model was distinct, except for LLaMA-V and Falcon-Y, who produced 60 billion tokens each.

  5. Gemini-M produced 81 billion tokens and Claude-G produced 108 billion tokens.

Which model consumed the least number of GPU hours?

  1. A

    LLaMA-V

  2. B

    Falcon-Y

  3. C

    Mistral-W

  4. D

    Either LLaMA-V or Falcon-Y

Show answer and explanation

Answer: (D) Either LLaMA-V or Falcon-Y

Question 51 Solution:

The least number of GPU hours in the set is 120 hours.

From the deduced matrix, the hours consumed by LLaMA-V and Falcon-Y are 120 and 180 in some order (both producing 60B tokens).

Therefore, the model that consumed the least GPU hours is Either LLaMA-V or Falcon-Y.

Question 51 · CAT Level

Directions for Questions 50 to 53

Answer the questions on the basis of the information given below.

Seven large language model architectures — GPT-P, BERT-S, LLaMA-V, Claude-G, Mistral-W, Gemini-M, and Falcon-Y — were evaluated in a standardized cloud GPU training benchmark.

The total GPU compute hours consumed by the seven models were 300, 240, 180, 160, 270, 120, and 150 hours (not necessarily in that order), and their token generation efficiencies (effective strike rates) were 33.3%, 60%, 30%, 40%, 80%, 45%, and 50% in some order.

Token Output (in billion tokens)=GPU Hours Faced×(Efficiency100)\text{Token Output (in billion tokens)} = \text{GPU Hours Faced} \times \left(\frac{\text{Efficiency}}{100}\right)

Benchmark Findings:

  1. No model produced 150 billion or more tokens, and the lowest token output produced by any model was 60 billion.

  2. GPT-P was the highest token producer overall, and either it consumed the maximum GPU hours (300) or had the highest efficiency (80%).

  3. The efficiency of one of GPT-P and BERT-S was twice that of the other.

  4. Total tokens produced by each model was distinct, except for LLaMA-V and Falcon-Y, who produced 60 billion tokens each.

  5. Gemini-M produced 81 billion tokens and Claude-G produced 108 billion tokens.

Who was the second highest token producer overall?

  1. A

    Claude-G

  2. B

    Mistral-W

  3. C

    BERT-S

  4. D

    Gemini-M

Show answer and explanation

Answer: (C) BERT-S

Question 52 Solution:

Ranking models by total token output:

  1. GPT-P: 128B

  2. BERT-S: 120B

  3. Claude-G: 108B

  4. Mistral-W: 90B

  5. Gemini-M: 81B

  6. LLaMA-V / Falcon-Y: 60B

BERT-S is the second highest producer with 120 billion tokens.

Therefore, the answer is BERT-S.

Question 52 · CAT Level

Directions for Questions 50 to 53

Answer the questions on the basis of the information given below.

Seven large language model architectures — GPT-P, BERT-S, LLaMA-V, Claude-G, Mistral-W, Gemini-M, and Falcon-Y — were evaluated in a standardized cloud GPU training benchmark.

The total GPU compute hours consumed by the seven models were 300, 240, 180, 160, 270, 120, and 150 hours (not necessarily in that order), and their token generation efficiencies (effective strike rates) were 33.3%, 60%, 30%, 40%, 80%, 45%, and 50% in some order.

Token Output (in billion tokens)=GPU Hours Faced×(Efficiency100)\text{Token Output (in billion tokens)} = \text{GPU Hours Faced} \times \left(\frac{\text{Efficiency}}{100}\right)

Benchmark Findings:

  1. No model produced 150 billion or more tokens, and the lowest token output produced by any model was 60 billion.

  2. GPT-P was the highest token producer overall, and either it consumed the maximum GPU hours (300) or had the highest efficiency (80%).

  3. The efficiency of one of GPT-P and BERT-S was twice that of the other.

  4. Total tokens produced by each model was distinct, except for LLaMA-V and Falcon-Y, who produced 60 billion tokens each.

  5. Gemini-M produced 81 billion tokens and Claude-G produced 108 billion tokens.

How many billion tokens did Mistral-W produce?

  1. A

    90

  2. B

    96

  3. C

    120

  4. D

    None of these

Show answer and explanation

Answer: (A) 90

Question 53 Solution:

From the deduced matrix, Mistral-W consumed 150 GPU hours at an efficiency of 60%:

Token Output=150×0.60=90 billion tokens\text{Token Output} = 150 \times 0.60 = 90\text{ billion tokens}

Therefore, Mistral-W produced 90 billion tokens.

Question 53 · CAT Level

Directions for Questions 54 to 57

Answer the questions on the basis of the information given below.

A beta testing study of 120 mobile app users evaluated preferences for three UI features: Dark Mode (Feature A), Offline Sync (Feature B), and Voice Dictation (Feature C).

Findings:

  • The number of users who liked at most two of these features was three times the number of users who liked at least two of these features.

  • The number of users who liked only Offline Sync (aa) was strictly greater than those who liked only Dark Mode (bb), which, in turn, was strictly greater than those who liked only Voice Dictation (cc): a>b>ca > b > c.

If the number of users who did not like any of the three features was one-fifth of those who liked exactly two features, which of the following could be the number of users who liked exactly one of the three features?

  1. A

    80

  2. B

    81

  3. C

    82

  4. D

    85

Show answer and explanation

Answer: (B) 81

Question 54 Solution:

Let:

  • nn = users who like none

  • ss = users who like exactly one (s=a+b+cs = a + b + c)

  • dd = users who like exactly two

  • tt = users who like all three

Total users:

n+s+d+t=120— (1)n + s + d + t = 120 \quad \text{--- (1)}

From the given condition (at most 2 = 3×3 \times at least 2):

n+s+d=3(d+t)  ⟹  n+s=2d+3tn + s + d = 3(d + t) \implies n + s = 2d + 3t

Substituting into (1):

(2d+3t)+d+t=120  ⟹  3d+4t=120(2d + 3t) + d + t = 120 \implies 3d + 4t = 120

Since dd and tt are non-negative integers, dd must be a multiple of 4 (d∈{40,36,32,…,0}d \in \{40, 36, 32, \dots, 0\}).

Also:

n+s=2d+3(120−3d4)=90−d4n + s = 2d + 3\left(\frac{120 - 3d}{4}\right) = 90 - \frac{d}{4}

Given n=d/5n = d/5, dd must also be a multiple of 5. Combining conditions, dd must be a multiple of lcm⁡(4,5)=20\operatorname{lcm}(4, 5) = 20.

Thus, possible values of dd are 40, 20, or 0:

  • If d=40d = 40: n=40/5=8n = 40/5 = 8, n+s=90−10=80  ⟹  s=80−8=72n + s = 90 - 10 = 80 \implies s = 80 - 8 = 72.

  • If d=20d = 20: n=20/5=4n = 20/5 = 4, n+s=90−5=85  ⟹  s=85−4=81n + s = 90 - 5 = 85 \implies s = 85 - 4 = 81.

  • If d=0d = 0: n=0n = 0, n+s=90  ⟹  s=90n + s = 90 \implies s = 90.

Among the options, 81 is the only possible value.

Question 54 · CAT Level

Directions for Questions 54 to 57

Answer the questions on the basis of the information given below.

A beta testing study of 120 mobile app users evaluated preferences for three UI features: Dark Mode (Feature A), Offline Sync (Feature B), and Voice Dictation (Feature C).

Findings:

  • The number of users who liked at most two of these features was three times the number of users who liked at least two of these features.

  • The number of users who liked only Offline Sync (aa) was strictly greater than those who liked only Dark Mode (bb), which, in turn, was strictly greater than those who liked only Voice Dictation (cc): a>b>ca > b > c.

If the number of users who did not like any of the three features was strictly less than those who liked only Voice Dictation, the number of users who did not like any of the three features could not be more than:

  1. A

    19

  2. B

    20

  3. C

    21

  4. D

    22

Show answer and explanation

Answer: (C) 21

Question 55 Solution:

We have n<cn < c, with a>b>c>na > b > c > n.

  • s=a+b+c=(90−d/4)−ns = a + b + c = (90 - d/4) - n.

  • To maximize nn, we maximize n+s=90−d/4n + s = 90 - d/4, which occurs when d=0d = 0 (n+s=90n + s = 90).

Then a+b+c=90−na + b + c = 90 - n.

Since a>b>c>na > b > c > n, the minimum integer values for c,b,ac, b, a are:

c≥n+1,b≥n+2,a≥n+3c \ge n + 1, \quad b \ge n + 2, \quad a \ge n + 3

a+b+c≥3n+6a + b + c \ge 3n + 6

90−n≥3n+6  ⟹  4n≤84  ⟹  n≤2190 - n \ge 3n + 6 \implies 4n \le 84 \implies n \le 21

Therefore, nn cannot be more than 21.

Question 55 · CAT Level

Directions for Questions 54 to 57

Answer the questions on the basis of the information given below.

A beta testing study of 120 mobile app users evaluated preferences for three UI features: Dark Mode (Feature A), Offline Sync (Feature B), and Voice Dictation (Feature C).

Findings:

  • The number of users who liked at most two of these features was three times the number of users who liked at least two of these features.

  • The number of users who liked only Offline Sync (aa) was strictly greater than those who liked only Dark Mode (bb), which, in turn, was strictly greater than those who liked only Voice Dictation (cc): a>b>ca > b > c.

If each user liked at least one of the three features, the number of users who liked only Offline Sync was 36, and the total number of users who liked Dark Mode was equal to those who liked Offline Sync, then the total number of users who liked Dark Mode could not be more than:

  1. A

    84

  2. B

    75

  3. C

    74

  4. D

    85

Show answer and explanation

Answer: (B) 75

Question 56 Solution:

Each user liked at least one feature   ⟹  n=0\implies n = 0.

Given a=36a = 36 (only Offline Sync). Since a>b>ca > b > c, the maximum value for bb is 35.

Total liking Dark Mode = Total liking Offline Sync:

b+p+t+r=a+p+t+q=36+p+t+qb + p + t + r = a + p + t + q = 36 + p + t + q

b+r=36+qb + r = 36 + q

Since b≤35b \le 35, to minimize qq we take q=0  ⟹  b+r=36q = 0 \implies b + r = 36.

With b=35b = 35, we have r=1r = 1.

To maximize Dark Mode (120−(a+q+c)=120−(36+0+c)=84−c120 - (a + q + c) = 120 - (36 + 0 + c) = 84 - c):

We must minimize cc. Minimum s=a+b+c=80s = a + b + c = 80 (when d=40,t=0d = 40, t = 0):

36+35+c=80  ⟹  c=936 + 35 + c = 80 \implies c = 9

Max Dark Mode=120−(36+0+9)=75\text{Max Dark Mode} = 120 - (36 + 0 + 9) = 75

Therefore, the number of users who liked Dark Mode could not be more than 75.

Question 56 · CAT Level

Directions for Questions 54 to 57

Answer the questions on the basis of the information given below.

A beta testing study of 120 mobile app users evaluated preferences for three UI features: Dark Mode (Feature A), Offline Sync (Feature B), and Voice Dictation (Feature C).

Findings:

  • The number of users who liked at most two of these features was three times the number of users who liked at least two of these features.

  • The number of users who liked only Offline Sync (aa) was strictly greater than those who liked only Dark Mode (bb), which, in turn, was strictly greater than those who liked only Voice Dictation (cc): a>b>ca > b > c.

Under the conditions of Question 56, the number of users who liked both Dark Mode and Offline Sync but not Voice Dictation could not be more than:

  1. A

    40

  2. B

    39

  3. C

    37

  4. D

    Cannot be determined

Show answer and explanation

Answer: (B) 39

Question 57 Solution:

From Question 56, at the maximum condition:

  • s=80s = 80, so d=40d = 40 and t=0t = 0.

  • Total Dark Mode = b+p+t+r=75b + p + t + r = 75.

  • b=35b = 35, t=0t = 0, r=1r = 1.

35+p+0+1=75  ⟹  p=3935 + p + 0 + 1 = 75 \implies p = 39

Here, pp represents users who liked Dark Mode and Offline Sync but not Voice Dictation.

Therefore, this count cannot be more than 39.

Question 57 · CAT Level

Directions for Questions 58 to 61

Answer the questions on the basis of the information given below.

Seven autonomous navigation AI models — Algo-A, Algo-J, Algo-P, Algo-K, Algo-H, Algo-R, and Algo-S — competed in an engineering evaluation tournament.

Tournament Rules:

  • Round 1: If any model secures more than 50% of the total benchmark points awarded in Round 1, it is declared the outright winner; otherwise, the top 5 models advance to Round 2.

  • Round 2: If any model secures more than 50% of the total points awarded in Round 2, it is declared the outright winner; otherwise, the top 3 models advance to Round 3.

  • Round 3: The model securing the maximum points in Round 3 is declared the tournament winner.

  • In any round, no two models obtained the same number of points.

The table below shows the benchmark points awarded in Round 1:

Candidate ModelPoints in Round 1
Algo-A3047
Algo-J3112
Algo-P3084
Algo-K3140
Algo-H3008
Algo-R3149
Algo-S3160

(Total Round 1 points = 21,700. The top 5 models advancing to Round 2 were Algo-S, Algo-R, Algo-K, Algo-J, and Algo-P).

If 15,000 points were awarded in Round 2, then the sum of points obtained by the model that received the maximum points and the model that received the minimum points in that round could not be less than:

  1. A

    3455

  2. B

    3752

  3. C

    3680

  4. D

    3684

Show answer and explanation

Answer: (B) 3752

Question 58 Solution:

In Round 2, 5 models competed with 15,000 points in total. To minimize the sum of the highest and lowest scores:

  • The lowest candidate can receive 0 points.

  • To make the highest candidate's score as small as possible, the top 4 candidates must share the remaining 15,000 points as evenly as possible with distinct integer scores: 15,000/4=375015{,}000 / 4 = 3750.

  • The closest distinct integer scores for the top 4 candidates are 3752, 3751, 3749, and 3748 (3752+3751+3749+3748=15,0003752 + 3751 + 3749 + 3748 = 15{,}000).

Thus, the minimum possible score for the top candidate is 3752.

Sum=3752+0=3752\text{Sum} = 3752 + 0 = 3752

Therefore, the sum could not be less than 3752.

Question 58 · CAT Level

Directions for Questions 58 to 61

Answer the questions on the basis of the information given below.

Seven autonomous navigation AI models — Algo-A, Algo-J, Algo-P, Algo-K, Algo-H, Algo-R, and Algo-S — competed in an engineering evaluation tournament.

Tournament Rules:

  • Round 1: If any model secures more than 50% of the total benchmark points awarded in Round 1, it is declared the outright winner; otherwise, the top 5 models advance to Round 2.

  • Round 2: If any model secures more than 50% of the total points awarded in Round 2, it is declared the outright winner; otherwise, the top 3 models advance to Round 3.

  • Round 3: The model securing the maximum points in Round 3 is declared the tournament winner.

  • In any round, no two models obtained the same number of points.

The table below shows the benchmark points awarded in Round 1:

Candidate ModelPoints in Round 1
Algo-A3047
Algo-J3112
Algo-P3084
Algo-K3140
Algo-H3008
Algo-R3149
Algo-S3160

(Total Round 1 points = 21,700. The top 5 models advancing to Round 2 were Algo-S, Algo-R, Algo-K, Algo-J, and Algo-P).

If the tournament winner was decided after Round 3, and exactly 10,000 points were awarded in each of Round 2 and Round 3, then the total points received across all three rounds combined by a model that was NOT the winner could not be more than:

  1. A

    13,158

  2. B

    13,083

  3. C

    13,082

  4. D

    13,159

Show answer and explanation

Answer: (D) 13,159

Question 59 Solution:

To maximize the 3-round total points for a non-winning model:

  • In Round 1, the model should be the highest scorer: Algo-S with 3160 points.

  • In Round 2 (10,000 points), since no model won in Round 2, the model could score at most 50% = 5000 points.

  • In Round 3 (10,000 points), since the model was not the winner, it could receive at most just one point less than the winner: e.g., the winner receives 5001 and this model receives 4999 points.

Maximum Total Points=3160+5000+4999=13,159\text{Maximum Total Points} = 3160 + 5000 + 4999 = 13{,}159

Therefore, the maximum points could not be more than 13,159.

Question 59 · CAT Level

Directions for Questions 58 to 61

Answer the questions on the basis of the information given below.

Seven autonomous navigation AI models — Algo-A, Algo-J, Algo-P, Algo-K, Algo-H, Algo-R, and Algo-S — competed in an engineering evaluation tournament.

Tournament Rules:

  • Round 1: If any model secures more than 50% of the total benchmark points awarded in Round 1, it is declared the outright winner; otherwise, the top 5 models advance to Round 2.

  • Round 2: If any model secures more than 50% of the total points awarded in Round 2, it is declared the outright winner; otherwise, the top 3 models advance to Round 3.

  • Round 3: The model securing the maximum points in Round 3 is declared the tournament winner.

  • In any round, no two models obtained the same number of points.

The table below shows the benchmark points awarded in Round 1:

Candidate ModelPoints in Round 1
Algo-A3047
Algo-J3112
Algo-P3084
Algo-K3140
Algo-H3008
Algo-R3149
Algo-S3160

(Total Round 1 points = 21,700. The top 5 models advancing to Round 2 were Algo-S, Algo-R, Algo-K, Algo-J, and Algo-P).

If 8,000 points were awarded in Round 2 and 12,000 points were awarded in Round 3, then the total points received across all three rounds combined by the winning model could not be less than:

  1. A

    7087

  2. B

    7009

  3. C

    7011

  4. D

    7085

Show answer and explanation

Answer: (A) 7087

Question 60 Solution:

To minimize the total points for the tournament winner:

  • The model should qualify from Round 1 with the minimum points: Algo-P with 3084 points.

  • In Round 2 (8,000 points, 5 models, top 3 advance), Algo-P needs only the minimum points to finish 3rd: candidates 4 and 5 get 0 and 1 point, so Algo-P needs only 2 points to advance.

  • In Round 3 (12,000 points, 3 models), to win, Algo-P needs the minimum plurality score: 12,000/3=400012{,}000 / 3 = 4000. The minimum winning score with distinct integer values is 4001 points (e.g., scores 4001, 4000, 3999).

Minimum Total Points=3084+2+4001=7087\text{Minimum Total Points} = 3084 + 2 + 4001 = 7087

Therefore, the total points could not be less than 7087.

Question 60 · CAT Level

Directions for Questions 58 to 61

Answer the questions on the basis of the information given below.

Seven autonomous navigation AI models — Algo-A, Algo-J, Algo-P, Algo-K, Algo-H, Algo-R, and Algo-S — competed in an engineering evaluation tournament.

Tournament Rules:

  • Round 1: If any model secures more than 50% of the total benchmark points awarded in Round 1, it is declared the outright winner; otherwise, the top 5 models advance to Round 2.

  • Round 2: If any model secures more than 50% of the total points awarded in Round 2, it is declared the outright winner; otherwise, the top 3 models advance to Round 3.

  • Round 3: The model securing the maximum points in Round 3 is declared the tournament winner.

  • In any round, no two models obtained the same number of points.

The table below shows the benchmark points awarded in Round 1:

Candidate ModelPoints in Round 1
Algo-A3047
Algo-J3112
Algo-P3084
Algo-K3140
Algo-H3008
Algo-R3149
Algo-S3160

(Total Round 1 points = 21,700. The top 5 models advancing to Round 2 were Algo-S, Algo-R, Algo-K, Algo-J, and Algo-P).

The number of points received across the first two rounds combined by a model that reached Round 3 could not be less than:

  1. A

    3010

  2. B

    3086

  3. C

    3087

  4. D

    Data insufficient

Show answer and explanation

Answer: (B) 3086

Question 61 Solution:

The candidate with the lowest score in Round 1 who advanced to Round 2 is Algo-P with 3084 points.

To advance from Round 2 to Round 3, a candidate needs only to finish in the top 3. With 5 candidates, the lowest score to achieve 3rd place occurs when candidates 4 and 5 receive 0 and 1 point respectively, so the 3rd place candidate needs only 2 points.

Minimum Combined Points=3084+2=3086\text{Minimum Combined Points} = 3084 + 2 = 3086

Therefore, the combined points could not be less than 3086.

Question 61 · CAT Level

Directions for Questions 62 to 65

Answer the questions on the basis of the information given below.

Six enterprise research labs — HyperScale (H), Radix (R), Synapse (S), Strata (St), Apex (A), and GridCore (G) — procured processor hours across nine specialized Quantum Processing Unit (QPU) tiers. The per-unit hourly rate ($) of each of the nine QPU tiers was a distinct integer chosen from:

{100,200,400,600,700,900,1100,1200,1400}\{100, 200, 400, 600, 700, 900, 1100, 1200, 1400\}

A total of 25 QPU instances were procured across the six labs. The number of instances procured by HyperScale was the minimum. Apex and GridCore procured the same number of instances, and Apex procured more instances than Synapse, who did not procure 3 instances. No two labs procured the same combination of QPU tiers.

Procurement Rules:

(i) Each lab procured at least 3 instances but not more than 5 instances. No lab procured more than one instance of any given tier.

(ii) The total amount spent by HyperScale was $2000, which was not less than the amount spent by GridCore.

(iii) HyperScale did not procure the costliest tier (1400).Apexprocuredboththecheapest(1400). Apex procured both the cheapest (100) and costliest ($1400) tiers.

(iv) The amount spent by Apex was twice the amount spent by GridCore. The amount spent by Strata was the average of the amounts spent by Apex and GridCore.

(v) An instance procured by Strata was priced $1100. The amount spent by each of Radix and Synapse was not more than the amount spent by Strata.

(vi) Two instances were procured of each of the tiers priced 100,100, 400, and $600.

(vii) Not more than two labs procured 5 instances each.

The price ($) of the QPU tier that was procured by the maximum number of research labs was:

  1. A

    200

  2. B

    700

  3. C

    900

  4. D

    1100

Show answer and explanation

Answer: (A) 200

Question 62 Solution:

Step 1: Instances per Lab

Total instances = 25. Bounds per lab: 3≤ni≤53 \le n_i \le 5. At most two labs procured 5 instances.

  • Apex and GridCore procured the same number, so they must be the two labs with 5 instances each (nA=nG=5n_A = n_G = 5).

  • HyperScale procured the minimum, so nH=3n_H = 3.

  • Synapse procured fewer than Apex and not 3, so nS=4n_S = 4.

  • Remaining instances = 25−(5+5+3+4)=825 - (5 + 5 + 3 + 4) = 8 for Radix and Strata, so nR=4n_R = 4 and nSt=4n_{St} = 4.

Step 2: Spend Analysis

  • GridCore bought 5 items costing ≤$2000\le \text{\textdollar}2000. The cheapest 5 items are {100,200,400,600,700}\{100, 200, 400, 600, 700\}, summing to 100+200+400+600+700=$2000100 + 200 + 400 + 600 + 700 = \text{\textdollar}2000. Thus, GridCore spent exactly $2000\text{\textdollar}2000.

  • From (iv), Apex spent 2×2000=$40002 \times 2000 = \text{\textdollar}4000.

  • Strata spent (4000+2000)/2=$3000(4000 + 2000) / 2 = \text{\textdollar}3000.

Lab Allocation Summary:

Research LabInstancesSpend ($)Tiers Procured ($)
GridCore52000100, 200, 400, 600, 700
Strata43000400, 600, 900, 1100
Apex54000100, 200, 1100, 1200, 1400
HyperScale32000200, 700, 1100
Radix42900 / 3000200, 700, 900, 1100/1200
Synapse43000 / 2900200, 700, 900, 1200/1100

Counting the frequency of each tier:

  • Tier $200 is procured by GridCore, Apex, HyperScale, Radix, and Synapse (5 labs).

Thus, the tier procured by the maximum number of labs was $200.

Question 62 · CAT Level

Directions for Questions 62 to 65

Answer the questions on the basis of the information given below.

Six enterprise research labs — HyperScale (H), Radix (R), Synapse (S), Strata (St), Apex (A), and GridCore (G) — procured processor hours across nine specialized Quantum Processing Unit (QPU) tiers. The per-unit hourly rate ($) of each of the nine QPU tiers was a distinct integer chosen from:

{100,200,400,600,700,900,1100,1200,1400}\{100, 200, 400, 600, 700, 900, 1100, 1200, 1400\}

A total of 25 QPU instances were procured across the six labs. The number of instances procured by HyperScale was the minimum. Apex and GridCore procured the same number of instances, and Apex procured more instances than Synapse, who did not procure 3 instances. No two labs procured the same combination of QPU tiers.

Procurement Rules:

(i) Each lab procured at least 3 instances but not more than 5 instances. No lab procured more than one instance of any given tier.

(ii) The total amount spent by HyperScale was $2000, which was not less than the amount spent by GridCore.

(iii) HyperScale did not procure the costliest tier (1400).Apexprocuredboththecheapest(1400). Apex procured both the cheapest (100) and costliest ($1400) tiers.

(iv) The amount spent by Apex was twice the amount spent by GridCore. The amount spent by Strata was the average of the amounts spent by Apex and GridCore.

(v) An instance procured by Strata was priced $1100. The amount spent by each of Radix and Synapse was not more than the amount spent by Strata.

(vi) Two instances were procured of each of the tiers priced 100,100, 400, and $600.

(vii) Not more than two labs procured 5 instances each.

The total amount spent ($) by Synapse was:

  1. A

    2900

  2. B

    3000

  3. C

    Either 2900 or 3000

  4. D

    None of these

Show answer and explanation

Answer: (C) Either 2900 or 3000

Question 63 Solution:

From the deduced allocation table, Radix and Synapse both procured 4 items, and their spend cannot exceed Strata ($3000).

Their possible combinations of items are {200,700,900,1100}\{200, 700, 900, 1100\} (summing to 2900)and2900) and {200, 700, 900, 1200}(summingto(summing to3000).

Since no additional clue distinguishes between Radix and Synapse, Synapse could have spent Either 2900 or 3000.

Question 63 · CAT Level

Directions for Questions 62 to 65

Answer the questions on the basis of the information given below.

Six enterprise research labs — HyperScale (H), Radix (R), Synapse (S), Strata (St), Apex (A), and GridCore (G) — procured processor hours across nine specialized Quantum Processing Unit (QPU) tiers. The per-unit hourly rate ($) of each of the nine QPU tiers was a distinct integer chosen from:

{100,200,400,600,700,900,1100,1200,1400}\{100, 200, 400, 600, 700, 900, 1100, 1200, 1400\}

A total of 25 QPU instances were procured across the six labs. The number of instances procured by HyperScale was the minimum. Apex and GridCore procured the same number of instances, and Apex procured more instances than Synapse, who did not procure 3 instances. No two labs procured the same combination of QPU tiers.

Procurement Rules:

(i) Each lab procured at least 3 instances but not more than 5 instances. No lab procured more than one instance of any given tier.

(ii) The total amount spent by HyperScale was $2000, which was not less than the amount spent by GridCore.

(iii) HyperScale did not procure the costliest tier (1400).Apexprocuredboththecheapest(1400). Apex procured both the cheapest (100) and costliest ($1400) tiers.

(iv) The amount spent by Apex was twice the amount spent by GridCore. The amount spent by Strata was the average of the amounts spent by Apex and GridCore.

(v) An instance procured by Strata was priced $1100. The amount spent by each of Radix and Synapse was not more than the amount spent by Strata.

(vi) Two instances were procured of each of the tiers priced 100,100, 400, and $600.

(vii) Not more than two labs procured 5 instances each.

Which of the following statements is/are definitely true?

(i) The amount spent by Radix was strictly more than that by Synapse.

(ii) The QPU tier priced $1100 was procured by four research labs.

  1. A

    Only (i)

  2. B

    Both (i) and (ii)

  3. C

    Only (ii)

  4. D

    Neither (i) nor (ii)

Show answer and explanation

Answer: (C) Only (ii)

Question 64 Solution:

  • Statement (i): Radix and Synapse spent $2900\text{\textdollar}2900 and $3000\text{\textdollar}3000 in either order. Radix is not necessarily more than Synapse. (False)

  • Statement (ii): The $1100\text{\textdollar}1100 tier is definitely procured by Strata, Apex, HyperScale, and exactly one of Radix and Synapse. That is 1+1+1+1=41 + 1 + 1 + 1 = 4 labs. (Definitely True)

Therefore, Only (ii) is definitely true.

Question 64 · CAT Level

Directions for Questions 62 to 65

Answer the questions on the basis of the information given below.

Six enterprise research labs — HyperScale (H), Radix (R), Synapse (S), Strata (St), Apex (A), and GridCore (G) — procured processor hours across nine specialized Quantum Processing Unit (QPU) tiers. The per-unit hourly rate ($) of each of the nine QPU tiers was a distinct integer chosen from:

{100,200,400,600,700,900,1100,1200,1400}\{100, 200, 400, 600, 700, 900, 1100, 1200, 1400\}

A total of 25 QPU instances were procured across the six labs. The number of instances procured by HyperScale was the minimum. Apex and GridCore procured the same number of instances, and Apex procured more instances than Synapse, who did not procure 3 instances. No two labs procured the same combination of QPU tiers.

Procurement Rules:

(i) Each lab procured at least 3 instances but not more than 5 instances. No lab procured more than one instance of any given tier.

(ii) The total amount spent by HyperScale was $2000, which was not less than the amount spent by GridCore.

(iii) HyperScale did not procure the costliest tier (1400).Apexprocuredboththecheapest(1400). Apex procured both the cheapest (100) and costliest ($1400) tiers.

(iv) The amount spent by Apex was twice the amount spent by GridCore. The amount spent by Strata was the average of the amounts spent by Apex and GridCore.

(v) An instance procured by Strata was priced $1100. The amount spent by each of Radix and Synapse was not more than the amount spent by Strata.

(vi) Two instances were procured of each of the tiers priced 100,100, 400, and $600.

(vii) Not more than two labs procured 5 instances each.

How many types of QPU tiers were procured by both HyperScale and Radix, but NOT by GridCore?

  1. A

    1

  2. B

    2

  3. C

    3

  4. D

    Cannot be determined

Show answer and explanation

Answer: (D) Cannot be determined

Question 65 Solution:

  • Tiers procured by HyperScale = {200,700,1100}\{200, 700, 1100\}

  • Tiers procured by GridCore = {100,200,400,600,700}\{100, 200, 400, 600, 700\}

  • Tiers in HyperScale but NOT GridCore: strictly {1100}\{1100\}.

Now check whether Radix procured {1100}\{1100\}:

  • If Radix procured {200,700,900,1100}\{200, 700, 900, 1100\}, then {1100}\{1100\} was procured by both HyperScale and Radix but not GridCore (Count = 1).

  • If Radix procured {200,700,900,1200}\{200, 700, 900, 1200\}, then the count is 0.

Since Radix's allocation is not uniquely determined, the answer Cannot be determined.

Question 65 · CAT Level

Directions for Questions 66 to 69

Answer the questions on the basis of the information given below.

Six enterprise IT executives — Dr. Gupta, Dr. Patel, Dr. Sharma, Dr. Chen, Dr. Hayes, and Dr. Hudson — attended an enterprise technology asset liquidation auction. Each executive purchased exactly one hardware asset.

The six hardware assets purchased (not necessarily in that order) were:

Display Panel,  Workstation,  GPU Server,  Switch Console,  Firewall Appliance,  Server Rack\text{Display Panel}, \; \text{Workstation}, \; \text{GPU Server}, \; \text{Switch Console}, \; \text{Firewall Appliance}, \; \text{Server Rack}

The original list prices of these items (not necessarily in that order) were:

$200,  $1500,  $300,  $2000,  $900,  $1200\text{\textdollar}200, \; \text{\textdollar}1500, \; \text{\textdollar}300, \; \text{\textdollar}2000, \; \text{\textdollar}900, \; \text{\textdollar}1200

The prices paid at auction (after various institutional discounts) were:

$800,  $600,  $1200,  $1000,  $75,  $50\text{\textdollar}800, \; \text{\textdollar}600, \; \text{\textdollar}1200, \; \text{\textdollar}1000, \; \text{\textdollar}75, \; \text{\textdollar}50

Auction Records:

  1. Dr. Gupta purchased an article of office workspace furnishing (Display Panel or Workstation).

  2. The Server Rack was purchased at a 50% discount, and its buyer’s surname begins with the letter 'H'.

  3. Dr. Chen purchased the item with a list price of $1500 for four-fifths (4/5) of its list price.

  4. The Firewall Appliance was sold for $100 less than its list price.

  5. The asset that was sold for $50 was an article of office workspace furnishing.

  6. Dr. Patel spent $400 less than Dr. Chen.

  7. Dr. Hayes paid for the Display Panel with a 100bill(valueequalto50100 bill (value equal to 50% of its list price) and received 25 in change.

  8. Dr. Gupta spent less than Dr. Sharma, who spent less than Dr. Patel.

  9. The asset with the highest original list price was not sold for the highest auction price, and the lowest list price asset was not sold for the lowest auction price.

  10. The price paid for the Switch Console was strictly greater than that for the GPU Server.

Who purchased the GPU Server?

  1. A

    Dr. Gupta

  2. B

    Dr. Patel

  3. C

    Dr. Sharma

  4. D

    Dr. Hudson

Show answer and explanation

Answer: (C) Dr. Sharma

Question 66 Solution:

Step 1: Identifying Purchases Step by Step

  • From (3): Dr. Chen paid 45×1500=$1200\frac{4}{5} \times 1500 = \text{\textdollar}1200 for an item with list price $1500\text{\textdollar}1500.

  • From (6): Dr. Patel spent $1200−$400=$800\text{\textdollar}1200 - \text{\textdollar}400 = \text{\textdollar}800.

  • From (7): Dr. Hayes paid for the Display Panel with a $100\text{\textdollar}100 bill and received $25\text{\textdollar}25 change   ⟹  \implies Price paid = $75\text{\textdollar}75. List price = 2×100=$2002 \times 100 = \text{\textdollar}200.

  • From (8): Price(Gupta) < Price(Sharma) < Price(Patel = $800\text{\textdollar}800). The prices below $800\text{\textdollar}800 are {600,75,50}\{600, 75, 50\}. Since Hayes paid $75\text{\textdollar}75, Dr. Sharma must have paid $600\text{\textdollar}600 and Dr. Gupta must have paid $50\text{\textdollar}50.

  • From (5): The item sold for $50\text{\textdollar}50 was a furnishing item (Display Panel or Workstation). Since Hayes bought the Display Panel, Dr. Gupta bought the Workstation for $50\text{\textdollar}50.

  • From (2): The Server Rack was bought at 50% off by an executive whose name begins with 'H' (Hayes or Hudson). Since Hayes bought the Display Panel, Dr. Hudson bought the Server Rack.

  • The only remaining unpaid auction price is $1000\text{\textdollar}1000, so Dr. Hudson paid $1000\text{\textdollar}1000. Since it was 50% off, the list price of the Server Rack was 2×1000=$20002 \times 1000 = \text{\textdollar}2000.

  • From (4): The Firewall Appliance sold for $100\text{\textdollar}100 less than its list price. With list prices {300,900,1200}\{300, 900, 1200\} and remaining paid prices {800,600,1200}\{800, 600, 1200\}, the only pair with difference 100 is List = $900\text{\textdollar}900 and Paid = $800\text{\textdollar}800. Thus, Dr. Patel bought the Firewall Appliance (List $900\text{\textdollar}900, Paid $800\text{\textdollar}800).

  • From (10): Price(Switch Console) > Price(GPU Server). The remaining paid prices are $1200\text{\textdollar}1200 (Dr. Chen) and $600\text{\textdollar}600 (Dr. Sharma). Thus, Dr. Chen bought the Switch Console (Paid $1200\text{\textdollar}1200), and Dr. Sharma bought the GPU Server (Paid $600\text{\textdollar}600).

Auction Summary:

ExecutiveHardware AssetList Price ($)Price Paid ($)Discount ($)
Dr. HayesDisplay Panel20075125
Dr. GuptaWorkstation30050250
Dr. SharmaGPU Server1200600600
Dr. PatelFirewall Appliance900800100
Dr. ChenSwitch Console15001200300
Dr. HudsonServer Rack200010001000

Therefore, Dr. Sharma purchased the GPU Server.

Question 66 · CAT Level

Directions for Questions 66 to 69

Answer the questions on the basis of the information given below.

Six enterprise IT executives — Dr. Gupta, Dr. Patel, Dr. Sharma, Dr. Chen, Dr. Hayes, and Dr. Hudson — attended an enterprise technology asset liquidation auction. Each executive purchased exactly one hardware asset.

The six hardware assets purchased (not necessarily in that order) were:

Display Panel,  Workstation,  GPU Server,  Switch Console,  Firewall Appliance,  Server Rack\text{Display Panel}, \; \text{Workstation}, \; \text{GPU Server}, \; \text{Switch Console}, \; \text{Firewall Appliance}, \; \text{Server Rack}

The original list prices of these items (not necessarily in that order) were:

$200,  $1500,  $300,  $2000,  $900,  $1200\text{\textdollar}200, \; \text{\textdollar}1500, \; \text{\textdollar}300, \; \text{\textdollar}2000, \; \text{\textdollar}900, \; \text{\textdollar}1200

The prices paid at auction (after various institutional discounts) were:

$800,  $600,  $1200,  $1000,  $75,  $50\text{\textdollar}800, \; \text{\textdollar}600, \; \text{\textdollar}1200, \; \text{\textdollar}1000, \; \text{\textdollar}75, \; \text{\textdollar}50

Auction Records:

  1. Dr. Gupta purchased an article of office workspace furnishing (Display Panel or Workstation).

  2. The Server Rack was purchased at a 50% discount, and its buyer’s surname begins with the letter 'H'.

  3. Dr. Chen purchased the item with a list price of $1500 for four-fifths (4/5) of its list price.

  4. The Firewall Appliance was sold for $100 less than its list price.

  5. The asset that was sold for $50 was an article of office workspace furnishing.

  6. Dr. Patel spent $400 less than Dr. Chen.

  7. Dr. Hayes paid for the Display Panel with a 100bill(valueequalto50100 bill (value equal to 50% of its list price) and received 25 in change.

  8. Dr. Gupta spent less than Dr. Sharma, who spent less than Dr. Patel.

  9. The asset with the highest original list price was not sold for the highest auction price, and the lowest list price asset was not sold for the lowest auction price.

  10. The price paid for the Switch Console was strictly greater than that for the GPU Server.

Who paid the highest auction price?

  1. A

    Dr. Sharma

  2. B

    Dr. Hudson

  3. C

    Dr. Hayes

  4. D

    Dr. Chen

Show answer and explanation

Answer: (D) Dr. Chen

Question 67 Solution:

From the deduced summary table, the auction prices paid were:

  • Dr. Hayes: $75

  • Dr. Gupta: $50

  • Dr. Sharma: $600

  • Dr. Patel: $800

  • Dr. Hudson: $1000

  • Dr. Chen: $1200

Dr. Chen paid the highest amount of $1200.

Question 67 · CAT Level

Directions for Questions 66 to 69

Answer the questions on the basis of the information given below.

Six enterprise IT executives — Dr. Gupta, Dr. Patel, Dr. Sharma, Dr. Chen, Dr. Hayes, and Dr. Hudson — attended an enterprise technology asset liquidation auction. Each executive purchased exactly one hardware asset.

The six hardware assets purchased (not necessarily in that order) were:

Display Panel,  Workstation,  GPU Server,  Switch Console,  Firewall Appliance,  Server Rack\text{Display Panel}, \; \text{Workstation}, \; \text{GPU Server}, \; \text{Switch Console}, \; \text{Firewall Appliance}, \; \text{Server Rack}

The original list prices of these items (not necessarily in that order) were:

$200,  $1500,  $300,  $2000,  $900,  $1200\text{\textdollar}200, \; \text{\textdollar}1500, \; \text{\textdollar}300, \; \text{\textdollar}2000, \; \text{\textdollar}900, \; \text{\textdollar}1200

The prices paid at auction (after various institutional discounts) were:

$800,  $600,  $1200,  $1000,  $75,  $50\text{\textdollar}800, \; \text{\textdollar}600, \; \text{\textdollar}1200, \; \text{\textdollar}1000, \; \text{\textdollar}75, \; \text{\textdollar}50

Auction Records:

  1. Dr. Gupta purchased an article of office workspace furnishing (Display Panel or Workstation).

  2. The Server Rack was purchased at a 50% discount, and its buyer’s surname begins with the letter 'H'.

  3. Dr. Chen purchased the item with a list price of $1500 for four-fifths (4/5) of its list price.

  4. The Firewall Appliance was sold for $100 less than its list price.

  5. The asset that was sold for $50 was an article of office workspace furnishing.

  6. Dr. Patel spent $400 less than Dr. Chen.

  7. Dr. Hayes paid for the Display Panel with a 100bill(valueequalto50100 bill (value equal to 50% of its list price) and received 25 in change.

  8. Dr. Gupta spent less than Dr. Sharma, who spent less than Dr. Patel.

  9. The asset with the highest original list price was not sold for the highest auction price, and the lowest list price asset was not sold for the lowest auction price.

  10. The price paid for the Switch Console was strictly greater than that for the GPU Server.

What hardware asset did Dr. Hudson buy?

  1. A

    The GPU Server

  2. B

    The Workstation

  3. C

    The Firewall Appliance

  4. D

    The Server Rack

Show answer and explanation

Answer: (D) The Server Rack

Question 68 Solution:

From Constraint 2, the buyer of the Server Rack has a surname starting with 'H' (Hayes or Hudson).

Since Dr. Hayes bought the Display Panel (Constraint 7), Dr. Hudson must have bought The Server Rack.

Question 68 · CAT Level

Directions for Questions 66 to 69

Answer the questions on the basis of the information given below.

Six enterprise IT executives — Dr. Gupta, Dr. Patel, Dr. Sharma, Dr. Chen, Dr. Hayes, and Dr. Hudson — attended an enterprise technology asset liquidation auction. Each executive purchased exactly one hardware asset.

The six hardware assets purchased (not necessarily in that order) were:

Display Panel,  Workstation,  GPU Server,  Switch Console,  Firewall Appliance,  Server Rack\text{Display Panel}, \; \text{Workstation}, \; \text{GPU Server}, \; \text{Switch Console}, \; \text{Firewall Appliance}, \; \text{Server Rack}

The original list prices of these items (not necessarily in that order) were:

$200,  $1500,  $300,  $2000,  $900,  $1200\text{\textdollar}200, \; \text{\textdollar}1500, \; \text{\textdollar}300, \; \text{\textdollar}2000, \; \text{\textdollar}900, \; \text{\textdollar}1200

The prices paid at auction (after various institutional discounts) were:

$800,  $600,  $1200,  $1000,  $75,  $50\text{\textdollar}800, \; \text{\textdollar}600, \; \text{\textdollar}1200, \; \text{\textdollar}1000, \; \text{\textdollar}75, \; \text{\textdollar}50

Auction Records:

  1. Dr. Gupta purchased an article of office workspace furnishing (Display Panel or Workstation).

  2. The Server Rack was purchased at a 50% discount, and its buyer’s surname begins with the letter 'H'.

  3. Dr. Chen purchased the item with a list price of $1500 for four-fifths (4/5) of its list price.

  4. The Firewall Appliance was sold for $100 less than its list price.

  5. The asset that was sold for $50 was an article of office workspace furnishing.

  6. Dr. Patel spent $400 less than Dr. Chen.

  7. Dr. Hayes paid for the Display Panel with a 100bill(valueequalto50100 bill (value equal to 50% of its list price) and received 25 in change.

  8. Dr. Gupta spent less than Dr. Sharma, who spent less than Dr. Patel.

  9. The asset with the highest original list price was not sold for the highest auction price, and the lowest list price asset was not sold for the lowest auction price.

  10. The price paid for the Switch Console was strictly greater than that for the GPU Server.

What was the percentage discount on the asset with the highest original list price?

  1. A

    70%

  2. B

    50%

  3. C

    25%

  4. D

    75%

Show answer and explanation

Answer: (B) 50%

Question 69 Solution:

The asset with the highest original list price was the Server Rack, with a list price of $2000\text{\textdollar}2000.

It was purchased by Dr. Hudson for $1000\text{\textdollar}1000:

Discount %=2000−10002000×100=50%\text{Discount } \% = \frac{2000 - 1000}{2000} \times 100 = 50\%

Therefore, the percentage discount was 50%.

Study this concept with interactive practice in PrepSoc.

Open the concept library