Question Description

Read the following three Case Studies and answer the questions at the end of each in paragraph format.

Case Study 1: Drive-Thru Service Times @ McDonald’s
(McDonalds Drive-Thru Waiting Times spreadsheet)

When you’re on the go and looking for a quick meal, where do you go? If you’re like millions of people every day, you make a stop at McDonald’s. Known as “quick service restaurants” in the industry (not “fast food”), companies such as McDonald’s invest heavily to determine the most efficient and effective ways to provide fast, high-quality service in all phases of their business.

Drive-thru operations play a vital role. It’s not surprising that attention is focused on the drive-thru process. After all, over 60% of individual restaurant revenues in the United States come from the drive-thru experience. Yet, understanding the process is more complex than just counting cars. Marla King, professor at the company’s international training center Hamburger University, got her start 25 years ago working at a McDonald’s drive-thru. She now coaches new restaurant owners and managers. “Our stated drive thru service time is 90 seconds or less. We train every manager and team member to understand that a quality customer experience at the drive-thru depends on them,” says Marla. Some of the factors that affect customers’ ability to complete their purchases within 90 seconds include restaurant staffing, equipment layout in the restaurant, training, efficiency of the grill team, and frequency of customer arrivals, to name a few. Also, customer order patterns also play a role. Some customers will just order drinks, whereas others seem to need enough food to feed an entire soccer team. And then there are the special orders. Obviously, there is plenty of room for variability here.

Yet, that doesn’t stop the company from using statistical techniques to better understand the drive-thru action. In particular, McDonald’s uses graphical techniques to display data and to help transform the data into useful information. For restaurant man agers to achieve the goal in their own restaurants, they need training in proper restaurant and drive-thru operations. Hamburger University, McDonald’s training center located near Chicago, Illinois, satisfies that need. In the mock-up restaurant service lab, managers go through a “before and after” training scenario. In the “before” scenario, they run the restaurant for 30 minutes as if they were back in their home restaurants. Managers in the training class are assigned to be crew, customers, drive-thru cars, special needs guests (such as hearing impaired, indecisive, clumsy), or observers. Statistical data about the operations, revenues, and service times are collected and analyzed. Without the right training, the restaurant’s operations usually start breaking down after 1 0–1 5 minutes. After debriefing and analyzing the data collected, the managers make suggestions for adjustments and head back to the service lab to try again. This time, the results usually come in well within standards. “When presented with the quantitative results, managers are pretty quick to make the connections between better operations, higher revenues, and happier customers,” Marla states.

When managers return to their respective restaurants, the training results and techniques are shared with staff who are charged with implementing the ideas locally. The results of the training eventually are measured when McDonald’s conducts a restaurant operations improvement process study, or ROIP. The goal is simple: improved operations. When the ROIP review is completed, statistical analyses are performed and managers are given their results. Depending on the results, decisions might be made that require additional financial resources, building construction, staff training, or reconfiguring layouts. Yet one thing is clear: Statistics drive the decisions behind McDonald’s drive-through service operations.


  1. After returning from the training session at Hamburger University, a McDonald’s store owner selected a random sample of 362 drive-thru customers and carefully measured the time it took from when a customer entered the McDonald’s property until the customer received the order at the drive-thru window. These data are in the file called McDonald’s Drive-Thru Waiting Times. Note, the owner selected some customers during the breakfast period, others during lunch, and others during dinner. Construct any appropriate graphs and charts that will effectively display these drive-thru data. Prepare a short discussion indicating the conclusions that this store owner might reach after reviewing the graphs and charts you have prepared.
  2. Referring to question 1, suppose the manager comes away with the conclusion that his store is not meeting the 90-second customer service goal. As a result he plans to dig deeper into the problem by collecting more data from the drive-thru process. Discuss what other measures you would suggest the manager collect. Discuss how these data could be of potential value in helping the store owner understand his problem. 3. Visit a local McDonald’s that has a drive-thru facility. Randomly sample 20 drive-thru customers and collect the following data:
    • a. the total time from arrival on the property to departure from the drive-thru window
    • b. the time from when customers place the order until they receive their order and exit the drive-thru process
    • c. the number of cars in the line when the sampled vehicle enters the drive-thru process
    • d. Using the data that you have collected, construct appropriate graphs and charts to describe these data. Write a short report discussing the data

Case Study 2: SaveMor Pharmacies

A common practice now is for large retail pharmacies to buy the customer base from smaller, independent pharmacies. The way this works is that the buyer requests to see the customer list along with the buying history. The buyer then makes an offer based on its projection of how many of the seller’s customers will move their business to the buyer’s pharmacy and on how many dollars of new business will come to the buyer as a result of the purchase. Once the deal is made, the buyer and seller usually send out a joint letter to the seller’s customers explaining the transaction and informing them that their prescription files have been transferred to the purchasing company.

The problem is that there is no guarantee regarding what proportion of the existing customers will make the switch to the buying company. That is the issue facing Heidi Fendenand, acquisitions manager for SaveMor Pharmacies. SaveMor has the opportunity to purchase the 6,780-person customer base from Hubbard Pharmacy in San Jose, California. Based on previous acquisitions, Heidi believes that if 70% or more of the customers will make the switch, then the deal is favorable to SaveMor. However, if 60% or less make the move to SaveMor, then the deal will be a bad one and she would recommend against it.

Quincy Kregthorpe, a research analyst who works for Heidi, has suggested that SaveMor take a new approach to this acquisition decision. He has suggested that SaveMor contact a random sample of 20 Hubbard customers telling them of the proposed sale and asking them if they will be willing to switch their business to SaveMor. Quincy has suggested that if 15 or more of the 20 customers indicate that they would make the switch, then SaveMor should go ahead with the purchase. Otherwise, it should decline the deal or negotiate a lower purchase price.

Heidi liked this idea and contacted Cal Hubbard, Hubbard’s owner, to discuss the idea of surveying 20 randomly selected customers. Cal was agreeable as long as only these 20 customers would be told about the potential sale.

Before taking the next step, Heidi met with Quincy to discuss the plan one more time. She was concerned that the proposed sampling plan might have too high a probability of rejecting the purchase deal even if it was a positive one from SaveMor’s viewpoint.

On the other hand, she was concerned that the plan might also have a high probability of accepting the purchase deal when in fact it would be unfavorable to SaveMor. After discussing these concerns for over an hour, Quincy finally offered to perform an evaluation of the sampling plan.


  1. Compute the probability that the sampling plan will provide a result that suggests that SaveMor should reject the deal even if the true proportion of all customers who would switch is actually 0.70.
  2. Compute the probability that the sampling plan will provide a result that suggests that SaveMor should accept the deal even if the true proportion of all customers who would switch is actually only 0.60.
  3. Write a short report to Heidi outlining the sampling plan, the assumptions on which the evaluation of the sampling plan has been based, and the conclusions regarding the potential effectiveness of the sampling plan. The report should make a recommendation about whether Heidi should go through with the idea of using the sampling plan.

Case Study 3: Credit Data, Inc.

Credit Data, Inc., has been monitoring the amount of time its bill collectors spend on calls that produce contacts with consumers. Management is interested in the distribution of time a collector spends on each call in which they initiate contact, inform a consumer about an outstanding debt, discuss a payment plan, and receive payments by phone. Credit Data is mostly interested in how quickly a collector can initiate and end a conversation to move on to the next call. For employees of Credit Data, time is money in the sense that one account may require one call and 2 minutes to collect, whereas another account may take five calls and 1 0 minutes per call to collect. The company has discovered that the time collectors spend talking to consumers about accounts is approximated by a normal distribution with a mean of 8 minutes and a standard deviation of 2.5 minutes. The managers believe that the mean is too high and should be reduced by more efficient phone call methods. Specifically, they wish to have no more than 1 0% of all calls require more than 1 0.5 minutes.


  1. Assuming that training can affect the average time but not the standard deviation, the managers are interested in knowing to what level the mean call time needs to be reduced in order to meet the 1 0% requirement.
  2. Assuming that the standard deviation can be affected by training but the mean time will remain at 8 minutes, to what level must the standard deviation be reduced in order to meet the 1 0% requirement?
  3. If nothing is done, what percent of all calls can be expected to require more than 1 0.5 minutes?

Grading Criteria:

  • Submit Completeness, showing work where applicable
  • Assignment Correctness


Mathematics Add-In for Microsoft Word
*will allow you to use Mathematic notation within a Microsoft Word document

Assignment Submission Instructions:

  1. Prepare you assignment offline in a word processing software such as Microsoft Office