Bar Graph: Concepts & Solved Examples
Master simple, grouped, stacked, and 100% subdivided bar charts, baseline distortion traps, cumulative inventory tracking, and constrained Diophantine optimization.
Core Conceptual Framework: Bar Graphs
A Bar Graph (or Bar Chart) is a foundational visual representation of categorical or discrete data using rectangular bars whose heights (in vertical column charts) or lengths (in horizontal bar charts) are strictly proportional to the numerical values they represent. In CAT and IPMAT Data Interpretation, bar graphs evaluate rapid visual estimation, ratio comparisons, percentage growth rates, multi-tier component decompositions, and constrained integer optimization.
1. Taxonomy of Bar Charts
A. Simple Bar Charts
-
Uses a single bar per category or time interval.
-
Orientation: Can be vertical (column chart) or horizontal.
-
Reading Metric: The height (or length) measured against the uniform scale represents the absolute metric directly.
-
Critical Check: Always verify whether the baseline starts at zero () or is truncated at an arbitrary non-zero value (). Truncated axes artificially amplify visual differences.
B. Grouped (Clustered) Bar Charts
-
Displays two or more bars side-by-side for each primary category to allow direct multi-metric comparisons across sub-categories (e.g., comparing Sales, Costs, and Profits across multiple fiscal quarters).
-
Legend Alignment: Each color or shading pattern corresponds to a specific entity across all clusters.
-
Analytical Utility: Ideal for evaluating ratios between entities within the same time period and tracking entity-specific trajectories over time.
C. Stacked (Component) Bar Charts
-
Each bar represents a cumulative total, subdivided into segments representing distinct constituent components stacked atop one another.
-
Cumulative vs. Component Values:
-
The bottom segment baseline is ; its height is read directly from the vertical axis.
-
Any interior (middle or upper) segment starts at a floating baseline. Its absolute value is:
where is the top boundary and is the bottom boundary of that segment.
-
The total bar height represents the grand sum of all constituent parts.
-
D. 100% Subdivided (Percentage) Bar Charts
-
All bars have an identical total height normalized to .
-
Each segment represents the relative percentage share of a component within that specific total.
-
The Base-Size Fallacy: A component occupying in Bar A may represent a smaller absolute quantity than a component occupying in Bar B if the total magnitude of Bar B is significantly larger than Bar A:
E. Bidirectional (Deviation) Bar Charts & Broken Scale Charts
-
Features bars extending both above and below a central zero-reference axis.
-
Used to depict positive and negative variances such as net profit/loss, trade surplus/deficit, or year-over-year percentage change.
-
Broken Scale Charts: Used when one or two outlier values dwarf the remaining data points; zig-zag break lines indicate non-linear jumps in scale.
2. Key Mathematical Operations in Bar Graphs
A. Segment Isolation in Stacked Charts
To determine the magnitude of component located between cumulative boundaries and :
Strategic Warning: Never read the top coordinate of an upper segment as its standalone value without subtracting the lower coordinate baseline.
B. Relative Share & Contribution Ratios
The percentage contribution of entity to the group total in period :
To determine which entity experienced the greatest change in share, evaluate the shift in across periods.
C. Multi-Period Net Inventory & Cumulative Flow Tracking
When bar charts depict rates of production (inflows) and sales/dispatches (outflows) over sequential intervals:
Over elapsed periods:
In extreme optimization problems (such as CAT solved examples), maximize or minimize net inventory by pairing maximum manufacturing with minimum sales, or vice-versa, subject to daily integer constraints.
D. Visual Ratio Comparison Shortcut
When comparing two ratios and from adjacent bars:
If both numerator and denominator increase from to :
-
If , the overall ratio increases.
-
If , the overall ratio decreases.
3. Strategic Approaches for CAT & IPMAT
-
Detect Truncated Axes Immediately: A bar spanning 110 to 120 looks twice as tall as one spanning 100 to 110 on a truncated axis starting at 100, yet the absolute difference is identical (+10). Never judge percentage growth by visual bar heights alone if the vertical axis does not start at 0.
-
Tabulate Complex Visuals First: For multi-entity grouped bar charts or stacked charts with multiple categories, spending 60 seconds building a clean matrix will prevent misreading values repeatedly across 4 questions.
-
Check Units and Scale Multipliers: Confirm whether the y-axis reads in thousands ('000s), lakhs, crores, millions, or percentages.
-
Leverage Integer Constraints: In puzzle-type DI sets (like multi-nominee elections, module installations, or discrete daily production), formulate Diophantine equations () to establish upper and lower bounds for candidate values.
Formulas
| Stacked Component Height Calculation | |
|---|---|
| Subdivided Percentage Breakdown | |
| Grouped Bar Proportional Share | |
| Net Inventory & Cumulative Flow Tracking | |
| Percentage Growth Rate Across True Baselines |
Traps and tips
- Trap: The Truncated Y-Axis Optical Illusion (Offset Baseline Distortion) — A bar spanning 110 to 120 appears twice as tall visually as one spanning 100 to 110 if the vertical axis is truncated at 100. However, the absolute change is identical (+10) and the actual growth is only 9.1%. Never evaluate percentage growth or ratios visually when the vertical axis does not originate at zero (). Always read numeric values off the axis.
- Trap: Stacked Component Baseline Shift Error — In a stacked (component) bar chart, the height of an upper segment is NOT the reading on the vertical axis. The vertical axis marks the cumulative total up to the top of that segment. You must subtract the coordinate of the bottom of the segment from the top coordinate to get its true value. If lower segments fluctuate, visual heights of upper segments will be visually displaced.
- Trap: Discrepant Base Size Fallacy in 100% Subdivided Charts — A component occupying 30% of Bar A (with a total base of 10,000 units = 3,000 units) is larger than a component occupying 50% of Bar B (with a total base of 4,000 units = 2,000 units). Never compare absolute segment sizes across 100% normalized bars without multiplying each by its respective total base magnitude.
- Tip: Cross-Multiplication for Visual Ratio Screening — When determining which clustered category has the greatest ratio , first visually eliminate all categories where . For the remaining candidates, verify using rapid mental cross-multiplication rather than long division.
- Tip: Formulating Integer Constraints in DI Puzzles — CAT bar graph sets often involve discrete units (e.g., consortium voting sheets, modular compute chassis, or daily production runs). Formulate integer linear equations ( and ) to deduce strict minimum and maximum permissible bounds ().
Solved examples
Solved Example Set 1: Questions 1 to 3
Directions for Examples 1 to 3
Four international autonomous robotics engineering teams – Team Apex, Team Falcon, Team Zephyr, and Team Orion – participated in a 4-stage autonomous quadruped robotic sprint relay over a closed 500-meter test circuit. Each team deployed four autonomous robotic runners (designated Stage 1, Stage 2, Stage 3, and Stage 4 robotic runners). In the relay, the first robot from each team starts running simultaneously from the origin line. As soon as the first robot from any team completes the 500-meter closed circuit, the second robot from that team starts running from the origin line, and this handoff process continues sequentially until the fourth robot has completed its lap. The same relay protocol applies to all teams. The winning team is the one whose robots take the least total composite time to complete the four laps combined.
The following bar graph provides information about the average speeds (in m/s) at which the four robotic runners from each team ran to complete their respective 500-meter laps:
| Team / Consortium | Stage 1 Speed (m/s) | Stage 2 Speed (m/s) | Stage 3 Speed (m/s) | Stage 4 Speed (m/s) |
|---|---|---|---|---|
| Team Apex | 10 | 20 | 12 | 25 |
| Team Falcon | 12 | 16 | 10 | 20 |
| Team Zephyr | 15 | 25 | 12 | 10 |
| Team Orion | 12 | 20 | 15 | 16 |
Question 1
Which team won the autonomous relay event?
(1) Team Orion
(2) Team Apex
(3) Team Zephyr
(4) Team Falcon
Question 2
What was the shortest distance between the starting origin line and the point on the track at which the second robotic runner from Team Apex and Team Falcon met?
(1) 133.33 metres
(2) 200.00 metres
(3) 333.33 metres
(4) The second robotic runners from Apex and Falcon did not meet
Question 3
Which of the following statements is definitely true?
(1) A robotic runner from Falcon was never behind the corresponding numbered runner from Apex.
(2) The third robotic runner from Zephyr and Falcon never met during the course of the event.
(3) The second robotic runner from Falcon and Zephyr definitely met once during the event.
(4) A robotic runner from Falcon was always behind the corresponding numbered runner from Zephyr.
Show solution
Step-by-Step Solution & Deductions
Each robotic runner completes 1 lap of . The time taken by a runner moving at uniform speed is:
The total composite time taken by a team across all 4 stages is:
Reading the speeds of the four runners from the bar graph:
-
Team Apex:
-
Team Falcon:
-
Team Zephyr:
-
Team Orion:
Question 1 Solution:
Calculate the composite time taken by each team:
-
Team Apex:
-
Team Falcon:
-
Team Zephyr:
-
Team Orion:
Since Team Orion takes the least composite time (), Team Orion won the event.
Correct Answer: Option (1)
Question 2 Solution:
Analyze the lap transition between the first and second runners of Team Apex and Team Falcon:
-
First Runner Stage:
-
Falcon Runner 1 finishes 500 m at: .
-
Apex Runner 1 finishes 500 m at: .
-
-
Second Runner Stage:
-
Falcon Runner 2 starts running at with speed .
-
Apex Runner 2 starts running at with speed .
-
-
When Apex Runner 2 starts at , Falcon Runner 2 has already been running for:
Distance covered by Falcon Runner 2 in this time (head start):
-
To overtake Falcon Runner 2, Apex Runner 2 has relative speed:
Time needed to close the gap:
-
Distance Apex Runner 2 would need to run from the starting line to catch Falcon Runner 2:
-
Since each runner runs only , Apex Runner 2 finishes his lap at before ever catching Falcon Runner 2.
Therefore, the second runners from Team Apex and Team Falcon did not meet.
Correct Answer: Option (4)
Question 3 Solution:
Evaluate Statement (2): Consider the Third Runners from Zephyr and Falcon:
-
Team Zephyr:
-
Runner 1 finishes at: .
-
Runner 2 runs at , taking .
-
Cumulative time when Zephyr Runner 3 starts: .
-
Zephyr Runner 3 runs at and finishes at: .
-
-
Team Falcon:
-
Runner 1 finishes at: .
-
Runner 2 runs at , taking .
-
Cumulative time when Falcon Runner 3 starts: .
-
-
Notice:
-
Zephyr Runner 3 starts at , whereas Falcon Runner 3 does not start until .
-
Moreover, Zephyr Runner 3 has speed whereas Falcon Runner 3 has speed (slower).
-
When Zephyr Runner 3 finishes at , Falcon Runner 3 has only covered .
-
Because Zephyr Runner 3 started earlier and ran faster throughout, Falcon Runner 3 was always behind. They never met.
-
Thus, Statement (2) is definitely true.
-
Correct Answer: Option (2)
Takeaway: In multi-stage speed relay problems, calculate cumulative handoff timestamps sequentially. Relative motion pursuit equations ( d = v rel t ) must always be constrained by the lap length boundary.
Solved Example Set 2: Questions 4 to 6
Directions for Examples 4 to 6
In an annual governance election for the Lead Architecture Steering Chair of an Open Cloud Infrastructure Consortium, 10 technology platform modules / delegates—namely P, Q, R, S, T, U, V, W, X, and Y—were nominated. The consortium comprises a total of 250 voting member enterprise organizations. Each enterprise voter submitted a priority ballot indicating their preferred nominees. Each voting enterprise was permitted to endorse a maximum of three different candidates on a priority basis: the first candidate as their first choice, the second candidate as their second choice, and the third candidate as their third choice. Every voter was mandated to endorse at least one candidate on the ballot.
The following bar graph represents the total number of times the name of each of the 10 nominated platform candidates appeared across all submitted voting ballots:
| Candidate | P | Q | R | S | T | U | V | W | X | Y | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Endorsements | 48 | 36 | 32 | 45 | 25 | 56 | 35 | 52 | 71 | 32 | 432 |
Question 4
How many of the following are not possible values of the number of enterprise voters who endorsed exactly one nominee on their ballot?
A. 64
B. 67
C. 162
D. 145
E. 165
F. 159
(1) 1
(2) 2
(3) 3
(4) 4
Question 5
What was the minimum possible number of enterprise voters who endorsed at least two nominees on their ballot?
(1) 90
(2) 91
(3) 92
(4) 93
Question 6
What could be the maximum value of the sum of the number of enterprise voters who endorsed exactly one nominee and that of those who endorsed exactly three nominees?
(1) 250
(2) 249
(3) 159
(4) 182
Show solution
Step-by-Step Solution & Deductions
1. Aggregate Data Extraction
From the bar graph, the number of endorsements received by each candidate is:
Total number of endorsements written across all sheets:
2. Algebraic Formulation
Let:
Since every voter endorsed at least 1 and at most 3 candidates:
Each ballot endorsing 1 candidate contributes 1 endorsement; 2 candidates contributes 2; 3 candidates contributes 3:
Subtracting Equation 1 from Equation 2:
Expressing in terms of :
3. Permissible Ranges
Since :
Because must be an integer:
Consequently:
-
For : , .
-
For : , .
-
Range of : (any integer from 68 to 159).
-
Range of : (even integers only).
-
Range of : .
Question 4 Solution:
Check each given value against the permissible range for ():
-
A. 64: Outside range () Not possible.
-
B. 67: Outside range () Not possible.
-
C. 162: Outside range () Not possible.
-
D. 145: Inside range () Possible.
-
E. 165: Outside range () Not possible.
-
F. 159: Inside range () Possible.
The values that are not possible are 64, 67, 162, and 165, which gives a total of 4 values.
Correct Answer: Option (4)
Question 5 Solution:
Number of voters who endorsed at least two candidates .
From Equation 1:
To minimize , we must maximize .
Maximum value of .
(Alternatively: . To minimize this, maximize .)
Correct Answer: Option (2)
Question 6 Solution:
We need the maximum value of the sum .
From Equation 1:
To maximize , we must minimize .
The minimum possible value of is (occurring when ).
(Alternatively: maximum and maximum .)
Correct Answer: Option (1)
Takeaway: When dealing with multi-selection survey or voting data, setting up linear Diophantine equations allows you to parameterize all sub-groups in terms of a single variable and establish sharp minimum and maximum boundary values.
Solved Example Set 3: Questions 7 to 10
Directions for Examples 7 to 10
The following bar graph provides information about the number of certification examination vouchers (in thousands, '000s) purchased for four specialized enterprise cloud & AI credentials—PCA (Professional Cloud Architect), CKA (Certified Kubernetes Administrator), MLD (Machine Learning DevOps), and SEC (Cloud Security Specialist)—across 5 regional enterprise tech hubs in India (NCR, Pune, Hyderabad, Bengaluru, and Chennai) in August 2017.
The official voucher fee per candidate for PCA, CKA, MLD, and SEC was ₹1,500, ₹1,200, ₹800, and ₹1,400 respectively. Assume that these are the only five regions where vouchers for the aforementioned credentials were purchased, and no other certification vouchers except the four specified credentials were sold in the given regions.
| Regional Tech Hub | PCA (₹1,500) | CKA (₹1,200) | MLD (₹800) | SEC (₹1,400) | Hub Total |
|---|---|---|---|---|---|
| NCR | 8.0 | 5.0 | 9.0 | 5.0 | 27.0 |
| Pune | 2.0 | 2.0 | 10.0 | 6.0 | 20.0 |
| Hyderabad | 4.5 | 3.0 | 12.0 | 4.5 | 24.0 |
| Bengaluru | 5.5 | 2.5 | 8.0 | 2.5 | 18.5 |
| Chennai | 6.0 | 2.5 | 6.0 | 4.0 | 18.5 |
| Total Vouchers | 26.0 | 15.0 | 45.0 | 22.0 | 108.0 |
Question 7
In August 2017, for which credential was the voucher revenue generated through sales in Hyderabad the second lowest as a percentage of that credential's total nationwide voucher revenue?
(1) PCA
(2) CKA
(3) SEC
(4) MLD
Question 8
If the number of vouchers sold for CKA in August 2018 increased by 20% compared to that in August 2017 and the total revenue generated through the sales of vouchers for the four certifications put together in August 2018 increased by 10% compared to that in August 2017, then what was the approximate percentage change in the fee of a CKA voucher in August 2018 over that in August 2017? (Assume that the fee and the number of vouchers sold for certifications other than CKA remained unchanged in August 2018.)
(1) 41%
(2) 35%
(3) 48%
(4) 52%
Question 9
In August 2017, the percentage contribution of NCR to the total voucher revenue generated through the sales of vouchers of the four certifications put together was closest to:
(1) 22%
(2) 26%
(3) 30%
(4) 34%
Question 10
In August 2017, in which regional tech hub(s) was the number of MLD vouchers sold less than 30% of the total number of vouchers sold for the four certifications put together in that particular hub?
(1) Chennai and Bengaluru
(2) Only NCR
(3) NCR and Chennai
(4) None of these
Show solution
Step-by-Step Solution & Deductions
1. Data Matrix (Number of Vouchers in '000s)
Extracting the values from the clustered bar chart:
| Regional Tech Hub | PCA (₹1,500) | CKA (₹1,200) | MLD (₹800) | SEC (₹1,400) | Hub Total |
|---|---|---|---|---|---|
| NCR | 8.0 | 5.0 | 9.0 | 5.0 | 27.0 |
| Pune | 2.0 | 2.0 | 10.0 | 6.0 | 20.0 |
| Hyderabad | 4.5 | 3.0 | 12.0 | 4.5 | 24.0 |
| Bengaluru | 5.5 | 2.5 | 8.0 | 2.5 | 18.5 |
| Chennai | 6.0 | 2.5 | 6.0 | 4.0 | 18.5 |
| Total Vouchers | 26.0 | 15.0 | 45.0 | 22.0 | 108.0 |
2. All-India Revenue in August 2017
-
-
-
-
-
Total Revenue (August 2017):
Question 7 Solution:
The percentage contribution of Hyderabad to the revenue generated by the sale of each credential:
(Note: Because the fee is uniform across all regions, the revenue contribution percentage is simply the ratio of vouchers sold in Hyderabad to total vouchers sold for that credential nationwide.)
-
PCA:
-
CKA:
-
SEC:
-
MLD:
Arranging the credentials in ascending order of percentage contribution:
The credential with the second lowest contribution is CKA ().
Correct Answer: Option (2)
Question 8 Solution:
-
Total Revenue in August 2018:
-
Increase in Total Revenue:
-
Since sales volume and voucher fees of PCA, MLD, and SEC remained unchanged, the entire ₹12,380 thousand increase came from CKA:
-
Number of CKA vouchers sold in August 2018:
-
New Fee per CKA voucher in 2018:
-
Percentage Change in Fee:
Correct Answer: Option (1)
Question 9 Solution:
Calculate NCR's revenue contribution in August 2017:
Percentage of all-India revenue:
Correct Answer: Option (2)
Question 10 Solution:
Evaluate the percentage of MLD vouchers sold in each regional hub:
-
NCR:
-
Pune:
-
Hyderabad:
-
Bengaluru:
-
Chennai:
In every single regional hub, the proportion of MLD vouchers sold strictly exceeds . There is no state/hub where it was less than .
Correct Answer: Option (4) (None of these)
Takeaway: When unit prices are identical across geographical regions, the regional share of revenue for any single product is identical to its regional share of volume. For overall revenue changes driven by a single product, attribute the entire dollar increment directly to that product.
Solved Example Set 4: Questions 11 to 14
Directions for Examples 11 to 14
An advanced quantum-accelerated supercomputing laboratory operates ten modular server chassis, designated . Each chassis is housed in one of four specialized rack enclosure types identified by status indicator colors: Obsidian, Platinum, Amber, and Sapphire. The number of high-coherence cryogenic QPU (Quantum Processing Unit) core modules installed in each of the ten chassis is one of five discrete capacity values: 16, 20, 28, 36, or 44 QPU modules.
The following bar graphs provide information about the number of server chassis of each enclosure color and the cumulative frequency distribution of QPU module capacities across the facility. Crucially, no two server chassis of the same enclosure color contain an equal number of QPU core modules.
Question 11
What is the total number of QPU core modules installed across all the server chassis housed in Obsidian enclosures?
(1) 144
(2) 136
(3) 140
(4) 148
Question 12
The total number of QPU core modules across all the ten server chassis put together in the laboratory is at most:
(1) 288
(2) 296
(3) 300
(4) 308
Question 13
If the number of QPU core modules in the server chassis housed in the Sapphire enclosure is 44, which of the following can be the total number of QPU modules across all chassis housed in Platinum enclosures?
(1) 60
(2) 52
(3) 64
(4) Both (a) and (b)
Question 14
If the total number of QPU core modules in the server chassis housed in Amber enclosures is the maximum possible, and the total number of QPU modules across all ten chassis combined is the least possible, what is the number of QPU core modules in the server chassis housed in the Sapphire enclosure?
(1) 36
(2) 44
(3) 20
(4) 16
Show solution
Step-by-Step Solution & Deductions
1. Information from Chart 1 (Server Chassis by Enclosure Color)
-
Obsidian: 5 chassis
-
Platinum: 2 chassis
-
Amber: 2 chassis
-
Sapphire: 1 chassis
-
Total server chassis .
Rule: No two chassis of the same enclosure color contain an equal number of QPU modules.
Because there are 5 possible module capacities and exactly 5 Obsidian chassis, the 5 Obsidian chassis must contain each of the 5 capacity values exactly once:
2. Deductions from Chart 2 (Cumulative Capacity Distribution)
Let be the number of chassis containing QPU modules:
-
At least 16 modules:
-
At most 20 modules:
-
At least 28 modules: (consistent since )
-
At least 36 modules:
- From .
-
At most 36 modules:
-
Since total is 10, the remaining chassis have modules (i.e. exactly 44):
-
Then from :
-
From :
Since Obsidian already has one 16 and one 20, and .
Thus, and .
-
| QPU Module Capacity | Number of Chassis |
|---|---|
| 16 | 1 - 3 |
| 20 | 1 - 3 |
| 28 | 1 |
| 36 | 3 |
| 44 | 2 |
Question 11 Solution:
The 5 Obsidian chassis contain 16, 20, 28, 36, and 44 QPU modules:
Correct Answer: Option (1)
Question 12 Solution:
To maximize the total QPU modules across all 10 chassis, maximize the number of chassis having 20 modules rather than 16 modules:
Since and , the maximum value for is , leaving .
Correct Answer: Option (3)
Question 13 Solution:
-
Total chassis with 44 modules . One is Obsidian; if the Sapphire chassis has 44 modules, then neither of the 2 Platinum chassis can have 44 modules.
-
Total chassis with 28 modules , which is already Obsidian. Thus neither Platinum chassis can have 28 modules.
-
Therefore, the 2 Platinum chassis must take distinct module counts from the remaining available set: .
-
Possible combinations for the 2 Platinum chassis:
-
-
Comparing with the given options (60, 52, 64), only 52 is possible.
Correct Answer: Option (2)
Question 14 Solution:
-
Amber chassis maximized: Amber has 2 chassis with distinct capacities. To maximize their sum, Amber must take the two largest possible denominations: 44 and 36 (since with one Obsidian, one can be Amber; with one Obsidian, one can be Amber).
-
Total modules least: To minimize total modules, we minimize and maximize :
Total inventory across all 10 chassis: three 16s, one 20, one 28, three 36s, two 44s.
-
Accounting for Obsidian and Amber:
-
Obsidian has:
-
Amber has:
-
Remaining modules for the 1 Sapphire and 2 Platinum chassis: two 16s and one 36.
-
-
The 2 Platinum chassis must have distinct module counts, so they cannot both have 16. Hence Platinum must take .
-
This leaves the remaining 16 for the Sapphire chassis.
Therefore, the Sapphire chassis must contain 16 QPU modules.
Correct Answer: Option (4)
Takeaway: When category counts match the number of possible values under a uniqueness constraint, the entire set of values is assigned deterministically. Use cumulative frequency inequalities to isolate frequencies of intermediate values.
Solved Example Set 5: Questions 15 to 18
Directions for Examples 15 to 18
An advanced clean-energy technology company, AetherVolt Energy Systems, manufactures three specialized solid-state battery cell chemistries: Chemistry Alpha, Chemistry Beta, and Chemistry Gamma. On any given operational production day during the first three months of 2017, the production batch volume for each variety manufactured was either 3, 6, or 7 Megawatt-hour (MWh) modular container units. Similarly, the commercial dispatch volume to the regional power grid on any day for each variety was either 1, 3, or 4 MWh container units.
On any single day, the manufacturing volumes across the three battery chemistries were mutually distinct (taking values 3, 6, and 7 in some permutation), and the dispatch volumes were also mutually distinct (taking values 1, 3, and 4 in some permutation).
The following bar graph provides information about the cumulative count of undistributed (unsold) MWh battery units of Alpha, Beta, and Gamma stored at the central distribution yard at the end of each of the first three months of 2017. Assume each month has exactly 30 operational days and there were zero unsold units at the end of 2016.
| Month | Chemistry Alpha | Chemistry Beta | Chemistry Gamma | Total Undistributed Units |
|---|---|---|---|---|
| End of January | 62 | 88 | 90 | 240 |
| End of February | 152 | 168 | 160 | 480 |
| End of March | 225 | 245 | 250 | 720 |
Question 15
If the maximum possible number of MWh units of Chemistry Gamma were manufactured in January, then find the least possible number of days on which 6 MWh units of Chemistry Beta were manufactured in January.
(1) 6
(2) 8
(3) 10
(4) 12
Question 16
Find the maximum possible number of days on which 7 MWh units of Chemistry Gamma were manufactured over the period of three months.
(1) 80
(2) 82
(3) 84
(4) 85
Question 17
Find the minimum possible number of days on which 7 MWh units of Chemistry Alpha were manufactured over the period of three months.
(1) 0
(2) 15
(3) 12
(4) 10
Question 18
If the ratio of the number of days on which 3, 6, and 7 MWh units of Chemistry Beta were manufactured in February was 1 : 2 : 3, and the number of days on which 4 MWh units of Chemistry Beta were dispatched in February was the minimum possible, then find the number of days on which 3 MWh units of Chemistry Beta were dispatched in February.
(1) 22
(2) 20
(3) 18
(4) 16
Show solution
Step-by-Step Solution & Deductions
1. Core Mathematical Constraints
-
3 battery chemistries: .
-
Daily manufacturing: Each chemistry receives a mutually distinct batch volume from .
-
Daily dispatches: Each chemistry receives a mutually distinct dispatch volume from .
-
Daily net addition to undistributed inventory across all 3 chemistries:
-
Over a 30-day month, the cumulative net addition across all chemistries combined is:
-
End of January:
-
End of February: ()
-
End of March: ()
-
Question 15 Solution:
-
Undistributed Chemistry Gamma at end of January = 90:
To maximize , we must maximize .
Maximum units of Gamma that can be dispatched in 30 days .
Then , which means Gamma was manufactured 7 units and dispatched 4 units on all 30 days of January.
-
Impact on Chemistry Beta:
Because daily quantities manufactured are distinct across varieties, if Gamma had 7 units manufactured every day, Beta could never have 7 units manufactured in January. Thus Beta was manufactured either 6 or 3 units each day.
Similarly, Beta could never have 4 units dispatched in January. Thus Beta was dispatched either 1 or 3 units each day.
-
Let be the days on which 6 units of Beta were manufactured (so 3 units on days).
Let be the days on which 1 unit of Beta was dispatched (so 3 units on days).
-
To minimize , maximize :
-
If : (not an integer).
-
If : .
Therefore, the least possible number of days on which 6 units of Beta were manufactured is 10.
-
Correct Answer: Option (3)
Question 16 Solution:
Examine each month independently to maximize days Chemistry Gamma was manufactured 7 units:
-
January:
. As deduced in Q15, Gamma can be manufactured 7 units and dispatched 4 units on all 30 days.
-
February:
.
.
To maximize days of 7 units, maximize total manufacturing by maximizing dispatches:
Max dispatches in Feb .
Let and be days on which 7 and 6 units were manufactured respectively (so 3 units on days):
To maximize , minimize :
- .
Max days of 7 units in February .
-
March:
.
Identical to January: Gamma can be manufactured 7 units on all 30 days.
Total maximum days over the 3 months:
Correct Answer: Option (4)
Question 17 Solution:
-
Total undistributed units of Chemistry Alpha at end of March .
-
Total units manufactured in 90 days .
-
To minimize the days on which 7 units of Alpha were manufactured, minimize the total units manufactured, which requires minimizing total dispatches.
-
Minimum possible dispatches over 90 days .
-
Total manufactured .
-
Let and be the number of days on which 7 and 6 units were manufactured (so 3 units on days):
-
To minimize , set :
Since and , this is fully feasible.
-
Therefore, the minimum possible days on which 7 units of Chemistry Alpha were manufactured is 0.
Correct Answer: Option (1)
Question 18 Solution:
-
Total days in February .
Ratio of days 3, 6, 7 units of Chemistry Beta manufactured .
-
Days of 3 units
-
Days of 6 units
-
Days of 7 units
Total manufactured in February:
-
-
Undistributed Beta produced in February .
-
Let be the number of days on which 1, 3, 4 units were dispatched respectively:
-
To minimize (days of 4 units dispatched), minimize :
-
Then the number of days on which 3 units of Chemistry Beta were dispatched:
Correct Answer: Option (2)
Takeaway: When solving rate-based inventory tracking puzzles with discrete daily permutation constraints, exploit aggregate flow invariance across all categories to establish exact monthly sums, then optimize single components by extreme value matching.
Practice questions
Question 1 · Foundation
Directions for Question 1
A cloud infrastructure startup tracks its monthly active compute instances (in thousands) over a 5-month period from January to May. The reported server instances are:
-
January: 120 ('000)
-
February: 135 ('000)
-
March: 150 ('000)
-
April: 165 ('000)
-
May: 198 ('000)
A metrics dashboard prepared two different bar charts: Chart A used a true zero baseline (), while Chart B used a truncated baseline starting at ('000).
What was the actual percentage growth in active compute instances from January to May, and by what factor did the visual height of the May bar exceed the January bar in Chart B (truncated at 100)?
Show answer and explanation
Answer: (A) 65.0%; 4.90 times
Step-by-Step Solution:
-
Actual Percentage Growth:
-
Visual Heights in Truncated Chart B:
Because Chart B starts at baseline :
-
Visual height of January bar .
-
Visual height of May bar .
-
Factor of visual height .
-
(In the true Chart A, the factor is simply times. The truncated baseline artificially magnifies the visual appearance from 1.65x to 4.90x!)
Question 2 · Foundation
Directions for Question 2
A clean-energy analytics firm recorded the quarterly dispatches (in units) of three commercial EV charger models—Model Alpha, Model Beta, and Model Gamma—across three consecutive quarters:
| Quarter | Model Alpha | Model Beta | Model Gamma |
|---|---|---|---|
| Q1 | 400 | 600 | 500 |
| Q2 | 550 | 650 | 800 |
| Q3 | 700 | 750 | 950 |
In which quarter was the ratio of dispatches of Model Gamma to the total dispatches of all three models combined the highest, and what was that ratio (expressed in percentage to the nearest integer)?
Show answer and explanation
Answer: (B) Q2; 40%
Step-by-Step Solution:
-
Quarter 1:
-
Total dispatches
-
Gamma share
-
-
Quarter 2:
-
Total dispatches
-
Gamma share
-
-
Quarter 3:
-
Total dispatches
-
Gamma share
-
Comparing the three quarters: .
The ratio was highest in Q2 at 40%.
Question 3 · Foundation
Directions for Question 3
A stacked bar chart represents the quarterly expenditure (in ₹ lakh) of a SaaS enterprise across four quarters. Each bar is divided from bottom to top into three segments: Cloud Infrastructure (bottom), Marketing & Sales (middle), and AI R&D (top). The cumulative boundaries on the vertical axis are:
| Quarter | Top of Cloud Infra (Bottom Segment) | Top of Marketing (Middle Segment) | Total Bar Height (Top of AI R&D) |
|---|---|---|---|
| Q1 | 80 | 140 | 200 |
| Q2 | 90 | 165 | 240 |
| Q3 | 110 | 180 | 270 |
| Q4 | 120 | 210 | 320 |
What was the absolute increase in Marketing & Sales expenditure from Q1 to Q4, and what percentage of the total quarterly expenditure in Q4 did Marketing & Sales constitute?
Show answer and explanation
Answer: (A) ₹30 lakh; 28.125%
Step-by-Step Solution:
-
Marketing Expenditure in Q1:
-
Marketing Expenditure in Q4:
-
Absolute Increase:
(Trap note: Subtracting cumulative coordinates is wrong because the baseline shifted from 80 to 120).
-
Marketing Percentage of Total in Q4:
Question 4 · Foundation
Directions for Question 4
A 100% subdivided bar chart shows the percentage distribution of engineers across three technical specializations—Core Systems, Cloud Architecture, and Machine Learning—in two engineering divisions:
-
Division Alpha: Total engineering workforce . Distribution: Core Systems , Cloud Architecture , Machine Learning .
-
Division Beta: Total engineering workforce . Distribution: Core Systems , Cloud Architecture , Machine Learning .
What is the total number of Machine Learning engineers across both divisions combined, and what percentage of the total combined engineering workforce do Machine Learning engineers represent?
Show answer and explanation
Answer: (A) 690; 23.0%
Step-by-Step Solution:
-
Machine Learning Engineers in Division Alpha:
-
Machine Learning Engineers in Division Beta:
-
Total Machine Learning Engineers:
-
Total Combined Workforce:
-
Overall Machine Learning Percentage:
(Trap note: Taking the simple unweighted average of and yields , which is incorrect because Division Beta has more engineers than Division Alpha).
Question 5 · Foundation
Directions for Question 5
A clean-tech manufacturing facility tracks its monthly production and commercial dispatches of modular battery storage units (in '000 units) over four months. On January 1, the plant started with an opening inventory of 15 ('000 units).
| Month | Production ('000 units) | Dispatches ('000 units) |
|---|---|---|
| January | 50 | 42 |
| February | 65 | 58 |
| March | 55 | 62 |
| April | 70 | 64 |
At the end of which month was the closing inventory of undistributed battery storage units the highest, and what was that peak closing inventory (in '000 units)?
Show answer and explanation
Answer: (A) February; 30 thousand units
Step-by-Step Solution:
Using the inventory formula:
-
January:
-
Opening
-
Net change
-
Closing
-
-
February:
-
Opening
-
Net change
-
Closing
-
-
March:
-
Opening
-
Net change
-
Closing
-
-
April:
-
Opening
-
Net change
-
Closing
-
Comparing the closing inventory across all four months:
-
January: 23
-
February: 30 (Peak)
-
March: 23
-
April: 29
The peak inventory occurred at the end of February at 30 thousand units.
Try real CAT questions
Study this concept with interactive practice in PrepSoc.
Open the concept library