One late afternoon at the end of October, Jason Mah, assistant residence manager at Cambridge Hall, was sitting at the residence cafeteria waiting for his friend to join him for dinner. His friend, unfortunately, had arrived in the middle of the “rush” period at the cafeteria, and faced a potentially long wait in the cafeteria line. As a student would put it: “It is a common joke around here, if you sit to eat while your friend is entering the cafeteria to get his meal, you will have enough time to finish dinner before he returns.” Mah was aware of the situation and already had started to investigate in order to propose some improvements to the residence management. Mah hoped that some minor adjustments requiring no major investments or disbursements would be sufficient to resolve the situation.
CAFÉ D. POWND

The “Café D. Pownd” was a newly established cafeteria at the equally new Cambridge Hall residence that had opened in September, at the National Academy of Liberal Arts (NALA). The cafeteria served as a popular eating place for the 450 residents of Cambridge Hall, as well as for approximately 100 residents from Nottingham Hall, another residence located across the street. This cafeteria was open from 7 a.m. to
7 p.m. Monday through Friday and was closed weekends and holidays. It offered a wide variety of hot meals, beverages, salads, desserts, snacks and sandwiches. The majority of residents paid for their food purchases with their student cards through an electronic payment system established campus-wide.
Although the Café D. Pownd was newly established, there was overwhelming evidence suggesting major problems with the length of waiting times for service at the cafeteria. In short, the cafeteria layout appeared extremely inefficient during the peak period of service (between 5 p.m. and 6:30 p.m.) during which approximately two-thirds of the residents expected to be served. There was not a similar rush of people during breakfast or lunch owing to most students eating a light breakfast in their rooms and being on campus during lunch hours. Numerous complaints about the long waiting times had been brought to the attention of residence staff and residents were getting extremely restless. There was a strong possibility that customers might use other alternatives if waiting times were not decreased. The residents had other options for meals: suites in Cambridge Hall and Nottingham Hall were equipped with furnished kitchens,

.

Page 2

and residents could also order a wide variety of food for delivery to the residence from local restaurants using their residence meal plans.

MANAGEMENT ORGANIZATION

NALA was a sizeable institution with 30,000 students. With this size of student population to serve, the
Academy had a director solely dedicated to the housing service campus-wide (see Exhibit 1 for an organizational chart). The campus had six different residences, each one under the supervision of a fulltime manager. Since Cambridge Hall and Nottingham Hall were sharing some services there was one manager supervising both of them.
The housing service had two additional layers to its organization: the assistant residence manager (ARM) and the residence advisors (RA). Both were full-time senior students. The RA’s primary role was to serve as a guide and resource for the students under the supervision of the ARMs.
The ARMs and the manager met bi-weekly to discuss general operations and concerns surrounding daily residence life. Disciplinary problems, general administration matters, and operational issues such as the cafeteria service were raised during these meetings.
LAYOUT OF THE CAFETERIA

Exhibit 2 shows the layout of the cafeteria. There were three main waiting lines in the cafeteria. Line #1
(L1) was for people that were waiting for food that was already pre-cooked; all the residents had to do was to select items from the daily offerings (e.g., hamburger and french fries) for service from one of two servers. In line #2 (L2), people waited for service from an interactive cooking station. The interactive station allowed residents to choose the actual ingredients they wanted to have in their meal to be prepared by one of two chefs.
The third line (L3) was for payment of the food. There were two cash registers but only one was generally open, even during the peak period. All three lines were served on a “first come first served” basis. The arrival pattern for the three lines also varied greatly; sometimes people arrived alone or other times in groups. Looking at the design plans for the cafeteria, an observer would immediately see the potential for congestion in such a small area. The lengths of the lines on the layout were typical for the peak period between 5 p.m. and 6:30 p.m. In general, the Café D. Pownd had fought an uphill battle from the start.
The general layout of the serving area was not appropriate since a lot of things were forced into a small area. However, there was not much that could be done about this as the problem stemmed from a lack of space, not just poor layout design. There simply was not enough space to accommodate the large number of people during the peak period.
One serious problem, which arose from this lack of space, was the confusion building in the queues since people had no idea which line they were in. There had been times when people had been lining up in the interactive cooking line for several minutes when they actually wanted to be in the pre-cooked line.

Page 3

9B00D007

GENERAL ANALYSIS

Mah had decided to investigate the problem and to provide some insights to the residence management in order to improve the situation. Therefore, Mah collected some data during the peak period using a “stopwatch” approach (Exhibit 3). Through his observations, Mah estimated that 25 per cent of people entering the cafeteria were not opting for a conventional hot meal (whether from the pre-cooked or the interactive lines). They simply opted for self-service snacks or junk food. It would be a fair assessment to say that the pre-cooked line attracted more people than the interactive line by a ratio of 2 to 1.
Pre-Cooked Line

In addition to the resident dissatisfaction that arose from the length of waiting time, the long lines caused further problems. The line tended to create a wall of people in front of the soft drink machines, becoming burdensome even with as few as 20 people waiting for service. In order for people to get drinks, they had to barge through a line of people. Usually this confusion caused spills and was somewhat annoying both to the person trying to get the drink and for the people who had to get out of the way. If a resident was lucky enough to get a drink without too many hassles, by the time they had received their food from the precooked line, they would have already drank their beverage. In such cases the customer often went back to the machines for a refill at the expense of the cafeteria. Therefore, the long waiting times were a catalyst for other problems.
From observations, approximately 20 to 30 seconds of the service time was due to customers asking for information about the meals offered. This slowed down the servers considerably and was an obvious reason for the queue to build up.
The two servers for this line found it difficult to operate efficiently in such close quarters, and since they served simultaneously it was very important that they did not slow each other down.
Interactive Cooking Line

The lines for the pre-cooked and interactive cooking were right next to each other (Exhibit 2). Therefore the drink machine problem also affected the interactive cooking line. The maximum length of this line before it interfered with people trying to get drinks was only 10 people, since it was closer to the machines.
Payment Line

After residents had received their food, they had to once again line up to pay for it. Though the average waiting time here was much less that the other two lines, waiting still occurred. Most customers had specified that one minute or less, as a waiting time for payment, was desirable. With the previous two lines, the drink machine problem happened if the lines were too long. With this line, if the length was longer than five people, a barrier formed in front of the salad bar. There were two main reasons that kept this line from moving faster. First, if a customer had chosen either a salad or candy as one of their purchases, it had to be weighed at the cash register. Second, though all the prices were programmed into the computer, there were certain items not recorded in the database, such as daily specials. Some of the people working at the cash register often spent up to 30 seconds looking for the price of these items.