| Course | BUS 660 Quantitative Methods |
|---|---|
| Topic | Topic 7 |
| Paper type | Simulation and inventory analysis |
| Length | About 960 words, 2 pages |
| Format | APA 7 student paper |
| School | Grand Canyon University (GCU) |
| Program | MBA |
| Updated | October 2026 |
Free sample paper for BUS 660 Topic 7
Capacity Above Demand, Customers Still Waiting: Simulating a Bakery Counter's Saturday Rush and Sizing Its Flour Safety Stock
[Author Name]
Colangelo College of Business, Grand Canyon University
BUS-660: Quantitative Methods
Topic 7 Simulation and Inventory Analysis
[Instructor Name]
[Date]
The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.
Capacity Above Demand, Customers Still Waiting: Simulating a Bakery Counter's Saturday Rush and Sizing Its Flour Safety Stock
The Deterministic Answer
A composite bakery's retail counter serves customers on Saturday mornings from 8 to 11. On average, 50 customers arrive an hour, and one clerk serves a customer in an average of one minute, so the counter can serve 60 an hour. A deterministic calculation says capacity exceeds demand by 20 percent and no one should wait. Customers complain of long lines anyway. This analysis shows why the deterministic answer misleads, tests three staffing policies by simulation, and applies the same logic to the bakery's flour inventory. Figures are illustrative.
Why Averages Mislead
Arrivals do not come evenly at 72-second intervals, and service does not take exactly one minute. Customers arrive in clusters, and some orders take much longer than others. When several customers arrive close together, a queue forms; when arrivals are sparse, the clerk sits idle. Idle time in a quiet stretch cannot be saved and spent in a busy one, so variability creates waiting even when average capacity exceeds average demand (Render et al., 2018). The closer utilization comes to 100 percent, the longer the waits, rising sharply rather than gradually.
The Simulation
The simulation models the three-hour morning in enough detail to be repeated. Arrivals follow a Poisson process whose rate varies by hour, based on four weeks of register timestamps: 40 an hour from 8 to 9, 65 from 9 to 10 and 45 from 10 to 11, averaging 50. Service times are exponentially distributed with a mean of one minute, a reasonable fit to timed transactions. Each policy is run 1,000 times with the same random inputs for each run, so policies are compared on identical mornings. Law (2015) recommends both practices, many replications and common random numbers, for comparing alternatives reliably. Three policies are tested: one clerk all morning, a second clerk from 9 to 10 only, and two clerks all morning.
Results
With one clerk, customers wait an average of about 5.9 minutes, and in the worst tenth of mornings the average wait exceeds 11 minutes; the longest single wait averages about 17 minutes. Most of the waiting happens between 9 and 10, when arrivals exceed the clerk's capacity. Adding a second clerk for that hour alone cuts the average wait to about 1.4 minutes, about 2.3 in the worst tenth of mornings. Two clerks all morning cut the average to about 0.3 minutes.
Table 1
Simulated Saturday Mornings, 1,000 Runs per Policy
| Policy | Average wait, minutes | Average wait, worst 10% of mornings | Average longest wait | Added labor cost per Saturday |
|---|---|---|---|---|
| One clerk | About 5.9 | About 11.6 | About 16.7 | None |
| Second clerk 9 to 10 | About 1.4 | About 2.3 | About 8.4 | About 18 dollars |
| Two clerks all morning | About 0.3 | About 0.5 | About 2.9 | About 54 dollars |
Checking the Simulation
A simulation is only as good as its inputs and its logic, so two checks were made. First, the one-clerk model was run at a constant 50 arrivals an hour, a case with a known theoretical answer: a single-server queue with these rates should average about 4.2 customers waiting and about 5 minutes of wait. The simulation's result, about 3.9 minutes over three-hour mornings that start with an empty counter, is consistent with that, since the queue takes time to build. Second, the simulated waits were compared with four observed Saturdays, when staff timed waits during the rush; observed averages of 5 to 7 minutes fall within the simulated range for one clerk.
Limits of the Model
The model assumes customers wait however long the line is. In reality some leave when they see a long line, so the one-clerk results understate lost sales and overstate the waits of those who stay. It also assumes one queue for all clerks, which is how the counter is arranged. Both assumptions make the recommendation, if anything, conservative.
Recommendation for the Counter
The analysis recommends a second clerk from 9 to 10. For about 18 dollars a Saturday, it removes most of the waiting, including in bad mornings. Two clerks all morning cost three times as much and reduce waits further only from short to negligible. The recommendation reports both the average and how bad the poor mornings were, because customers remember the 17-minute wait, not the average.
The Same Logic for Flour
Inventory follows the same principle. The bakery uses on average 400 kilograms of flour a day, but daily use varies with orders, with a standard deviation of about 80 kilograms. Its mill delivers five days after an order. An order placed when stock falls to five days of average use, 2,000 kilograms, will run out about half the time before delivery arrives, because use over five days is as likely to exceed its average as to fall below it. Safety stock exists because of this variability, not because of the average (Silver et al., 2017).
Assuming daily use varies independently, the standard deviation of use over five days is 80 times the square root of five, about 179 kilograms. To run out in only about 5 percent of order cycles, the bakery should hold about 1.65 standard deviations of safety stock, about 295 kilograms, and reorder when stock falls to about 2,295 kilograms. The extra flour costs a little to hold; running out on a busy week would cost far more in lost orders.
Conclusion
Average capacity above average demand still produces queues, because arrivals and service vary and idle time cannot be stored. Simulating 1,000 Saturday mornings shows that a second clerk in the busiest hour removes most of the waiting at modest cost. The same variability explains why the bakery needs about 295 kilograms of flour safety stock beyond its average needs during the mill's lead time.
References
Law, A. M. (2015). Simulation modeling and analysis (5th ed.). McGraw-Hill Education.
Render, B., Stair, R. M., Hanna, M. E., & Hale, T. S. (2018). Quantitative analysis for management (13th ed.). Pearson.
Silver, E. A., Pyke, D. F., & Thomas, D. J. (2017). Inventory and production management in supply chains (4th ed.). CRC Press.
The BUS-660 Topic 7 assignment instructions
BUS-660 Topic 7 asks MBA students to use simulation and inventory models to manage variability. Strong analyses show why deterministic answers based on averages mislead, describe the simulation in enough detail to repeat, and compare policies on the same random runs, reporting both averages and bad outcomes. Students should check the simulation against theory and observation, state its limits, and apply the same logic to safety stock, computing it from demand variability over the lead time. The model's limits, such as customers leaving long lines, should be stated. Students should also run enough replications to report variability, not a single simulated Saturday.
Inside the BUS 660 Topic 7 example
The finished example demonstrates why the deterministic answer misleads. A service point with average capacity above average demand still produces queues, because arrivals and service times vary and a busy stretch cannot be recovered from a quiet one, and the example shows that rather than asserting it. Inventory follows the same logic: safety stock exists because of variability, not because of the average. The simulation is described in enough detail to be repeated, with the distributions, the number of runs and the random inputs stated. Several policies are then tested against each other on the same runs, and the recommendation reports both the average outcome and how bad the poor runs were.
Reading the BUS 660 Topic 7 rubric
Faculty reward an analysis that shows variability causing waits. An Excellent paper explains why idle time cannot be stored and simulates 1,000 mornings with time-varying arrivals and common random numbers, following Law. Credit goes to a table of three policies with average and worst-decile waits and labor costs. The rubric values checking the model against the single-server formula and timed Saturdays. Strong papers recommend the peak-hour clerk on both average and bad mornings. Top marks go to computing about 295 kilograms of safety stock, with Silver, Pyke and Thomas. Papers that explain why customers wait even though average capacity exceeds average demand, using variability in arrivals and service, show the insight simulation exists to provide.
BUS 660 Topic 7 help before you submit
Solving a variability problem with averages is the error the topic exists to expose, and it produces the confident conclusion that no queue should form at a service point where queues obviously do. A second failure is a simulation reported without its parameters, which makes the result unverifiable and the method undemonstrated. Papers lose marks for a single run treated as an answer, since one replication of a random process is an anecdote. Comparing policies across different random draws confounds the policy with the luck of the sample. Reporting only mean outcomes hides the tail, and the tail is where stockouts, lost customers and service failures actually live. Simulations run without a warm up period report startup behavior as though it were steady state.
Get a BUS-660 Topic 7 example written to your instructions
Send the BUS-660 Topic 7 problems and the rubric your classroom posts, with the operational scenario and any data your section supplied. We write a custom example to those criteria, with the naive answer computed and refuted, the simulation fully specified, policies tested on identical draws and the bad runs reported, in 24 to 48 hours. The first is free.
More BUS 660 papers and MBA samples
- BUS 660 Topic 1: Payoff Table Analysis
- BUS 660 Topic 2: Decision Tree Analysis
- BUS 660 Topic 3: Forecasting Method Comparison
- BUS 660 Topic 4: Predictive Regression Model
- BUS 660 Topic 5: Linear Programming Formulation
- BUS 660 Topic 6: Network Optimization Problem
- BUS 660 Topic 8: Project Scheduling Recommendation
- BUS 655 Topic 7: Holdout Test Plan
- BUS 635 Topic 7: Property Revenue Strategy
- ACC 650 Topic 7: Standard Cost Variance Report
- HRM 635 Topic 7: Critical Talent Retention Analysis
BUS 660 Topic 7 questions, answered
Why do queues form when capacity exceeds demand?
Because both arrivals and service times vary. A server who can handle twelve customers an hour on average will still fall behind during a stretch when fifteen arrive, and the idle time during a quiet stretch afterward does not undo the wait those customers already experienced. Variability creates queues on its own, which is exactly why averaging the problem gives the wrong answer.
How many simulation runs do I need?
Enough that your reported result stops moving when you add more, which is usually more than students expect and easy to check. Run the simulation at increasing replication counts and watch the estimate settle. One run is an anecdote about one random sequence. Report the number you used and the variation across runs, because both tell the reader how much to trust the figure.
Why does safety stock exist?
To absorb variability in demand and in lead time, not to cover the average. If demand and replenishment were perfectly predictable you could hold almost nothing and never run out. The stock you carry beyond the expected requirement is buying protection against the bad draws, which is why the right level depends on how variable your demand is and how costly a stockout would be.
Where can I find a free BUS 660 Topic 7 sample paper?
The simulation and inventory analysis appears here in full: the deterministic answer, why averages mislead, the simulation, results, checks, limits, the counter recommendation, flour safety stock and the conclusion. Bring your service or inventory data and we will simulate it free.
How is the flour safety stock set in BUS 660 Topic 7?
Daily use varies by about 80 kilograms, so over a five-day lead time the standard deviation is about 179. Holding 1.65 of those, about 295 kilograms, limits stockouts to about 5 percent of cycles.