Case Drills: Five Operations Cases With Full Solutions
The biggest misconception about operations cases is that they are “math cases.” They are not. They are flow cases - a customer, material, order or task is stuck somewhere, and your job is to find where the system is losing time, money or reliability.
- Start with the flow: map input, process, output and constraint before calculating anything.
- Bottleneck rule: system capacity is limited by the slowest effective step, not the average step.
- Inventory rule: reorder point = demand during lead time + safety stock.
- Queue rule: high utilisation looks efficient, but near 100% utilisation destroys waiting time performance.
- Line balancing rule: takt/cycle time decides how much work each station can absorb.
- Warehouse rule: productivity improvement and headcount planning must be tested against peak demand, not average demand.
- Best case answer: diagnose the constraint, quantify the gap, test 2-3 levers, recommend with risk checks.
Big Picture: Every Operations Case Is a Flow Problem
Whether the case is a café queue, factory line, dark-store picker shortage or stockout issue, the logic is the same: trace the flow, find the constraint, quantify the gap, choose the lever, then check whether the recommendation creates a new problem elsewhere.
Core Explanation: The Case-Solving Toolkit
Use one operating sentence before you touch the numbers: “I will map the process, identify the bottleneck, calculate the service or cost gap, compare feasible levers and recommend the option with the best operational trade-off.”
That sentence prevents the most common trap: doing isolated arithmetic without showing business judgement.
The Four Case Shapes You Will See
Most operations cases fall into four patterns. Classifying the case quickly tells you which formulas and levers to reach for.
Metrics You Must Be Able to Use
In operations cases, metrics are not decoration. They tell you whether the system can meet demand, where it is fragile and whether the recommendation is practical.
Five Operations Case Drills With Full Solutions
Practise these as spoken cases. For each one, say the structure first, then calculate. That is exactly how you show both operations thinking and numerical control.
Case 1: Café Capacity Bottleneck
Prompt: A café receives 54 drink orders per hour during the morning rush. The process has three steps: order taking takes 1 minute per order, brewing takes 2.5 minutes per order with 2 baristas, and packing takes 0.8 minutes per order. What is the bottleneck and what should the café do?
Solution: The bottleneck is brewing because it can process only 48 orders per hour, while demand is 54. The capacity gap is 6 orders per hour.
If the café adds one more brewer, brewing capacity becomes 3 x 60 / 2.5 = 72 orders per hour. But the new system capacity becomes 60 orders per hour because order taking now becomes the constraint. So the recommendation is: add one brewer only if the café also monitors order-taking queues; otherwise the bottleneck simply moves upstream.
Answer line: “Brewing is the current bottleneck at 48 orders per hour versus demand of 54. Adding a third brewer solves the immediate gap, but order taking becomes the next constraint at 60 orders per hour.”
Case 2: Inventory Reorder Point
Prompt: A retailer sells 80 units of a SKU per day. Supplier lead time is 5 days. Management wants 120 units of safety stock. Order quantity is 800 units. Calculate the reorder point and average inventory.
Formula: Reorder point = demand during lead time + safety stock.
Solution: The retailer should place a replenishment order when inventory falls to 520 units. Average inventory is also 520 units because average cycle stock is 400 and safety stock is 120.
If the student wants to go deeper into policy design, this case naturally extends into setting inventory policy for a multi-product business, where different SKUs need different service levels and reorder rules.
Answer line: “The reorder point is 520 units. If we order 800 units each time, average inventory is 520 units before considering seasonality or demand variability.”
Case 3: Queue and Staffing at a Service Counter
Prompt: A service desk gets 40 customers per hour. One staff member can serve a customer in 3 minutes on average. Should the desk run with 2 staff members or 3?
Step 1 - Calculate service capacity per staff member: 60 / 3 = 20 customers per hour.
Solution: Two staff members look mathematically sufficient because capacity equals average demand. But in a real queue, arrivals and service times vary. At 100% utilisation, even small randomness creates long waiting lines. Three staff members create a capacity buffer and keep utilisation at 66.7%, which is operationally safer.
Answer line: “I would recommend 3 staff during the demand window. Two staff gives 100% utilisation, which is unstable for a variable-arrival service process.”
Case 4: Line Balancing for a Small Assembly Cell
Prompt: A product requires five tasks: A = 40 seconds, B = 50 seconds, C = 30 seconds, D = 60 seconds and E = 20 seconds. The target output is 48 units per hour. What is the minimum number of workstations and a feasible assignment?
Step 1 - Calculate cycle time: Available time per hour / required units = 3,600 seconds / 48 = 75 seconds per unit.
Step 2 - Calculate total work content: 40 + 50 + 30 + 60 + 20 = 200 seconds.
Step 3 - Minimum theoretical workstations: 200 / 75 = 2.67, so at least 3 workstations are needed.
Efficiency: Total work / total available station time = 200 / (3 x 75) = 88.9%.
Solution: Three workstations are required. The proposed allocation meets the 75-second cycle time at every station. For a deeper version of this case, revise line balancing and workstation design.
Answer line: “The required cycle time is 75 seconds. Minimum workstations are 3, and the assignment A+C, B+E, D gives 88.9% line efficiency.”
Case 5: Warehouse Picking Productivity
Prompt: A warehouse must fulfil 2,400 orders during a 9-hour peak day. It currently has 15 pickers. Each order takes 4 minutes to pick. Management can either add pickers or improve picking time through batching. What should it do?
Current capacity: 15 pickers x 9 hours x 60 minutes / 4 minutes = 2,025 orders per day. The warehouse is short by 375 orders.
Solution: Adding 3 pickers barely clears the target and leaves almost no buffer. Batching alone improves productivity but does not meet demand. The best recommendation is to combine process improvement with limited staffing: 2 additional pickers plus batching creates enough capacity and a more realistic peak-day buffer.
This is the same operating logic behind pull-based replenishment and picker workload smoothing. If you want the replenishment side of the warehouse case, revise Kanban and pull-based replenishment.
Answer line: “The best option is not pure hiring or pure productivity improvement. Two extra pickers plus batching gives 2,623 orders of capacity and keeps utilisation below the danger zone.”
Definitions You Can Say in One Breath
- Operations case: a business problem where flow, capacity, inventory, cost or service reliability must be diagnosed and improved.
- Bottleneck: the step whose effective capacity limits the output of the whole system.
- Cycle time: the time required to complete one unit at a process step or workstation.
- Takt time: the maximum time available per unit to meet customer demand.
- Safety stock: extra inventory held to protect service levels against demand or supply variability.
Chai Point: Standardising a Variable Operations System
Chai Point is a useful Indian operations case because it tries to make a highly variable product - fresh tea - consistent across retail, office and delivery contexts.
Situation: Tea looks simple to the customer, but operationally it is full of variability: milk quality, brew strength, temperature, staff technique, rush-hour demand and last-mile delivery time can all affect the experience.
The move: The operating answer is not one magic lever. The primary driver is standardisation - recipes, preparation routines and equipment that reduce variation in the cup. Supporting drivers include demand planning by location, batching rules during peak hours, replenishment discipline for milk and tea ingredients, staff training, and technology visibility into orders and fulfilment.
Outcome or lesson: The lesson for case interviews is powerful: operations advantage often comes from making a messy, human process repeatable without making the customer experience feel mechanical. A good answer would not say “Chai Point wins because it has stores.” It would say the system works when standardised preparation is supported by demand planning, replenishment and frontline execution.
How AI Changes Operations Case Drills
AI does not remove the need for operations fundamentals. It makes weak fundamentals more visible because you can now generate more practice cases, test more scenarios and catch arithmetic mistakes faster.
- Scenario generation: Tools can create variants of the same case - different demand, lead time, staffing and service-level assumptions - so you stop memorising one template.
- Math checking: AI can review your capacity, utilisation, reorder point and line-balancing calculations, but you must still verify formulas and units.
- Operational simulation thinking: For cases with queues, peaks and uncertainty, AI can help you compare “average demand” versus “peak demand” recommendations and identify where buffers are needed.
Use ChatGPT or Claude like a case coach: paste one solved case, ask it to create three harder variants, solve them yourself first, then ask it to check only arithmetic, assumptions and missed bottlenecks. Do not ask for the answer before you attempt the case.
For the next layer, connect these drills to using AI for inventory optimisation and replenishment, where forecasting and reorder decisions become data-driven rather than purely spreadsheet-based.
Interview Relevance
“A delivery hub is missing its same-day dispatch target. Orders have increased, staff say they are busy all day, and customers are complaining about delays. How would you diagnose and fix the operation?”
Use the phrase “effective capacity,” not just “capacity.” It signals that you understand breaks, absenteeism, rework, machine downtime and variability.
Common Mistake
The biggest error is solving with averages only. Averages hide peaks, variability and queues, so candidates recommend systems that work on paper and fail in reality. One-line fix: always ask, “Does this capacity survive peak demand and variability?”