La Quinta Write up - A project we had to do for this class. We had to decide where to locate a hotel PDF

Preview

Summary

A project we had to do for this class. We had to decide where to locate a hotel based on certain factors and a regression model. Did very well and got a solid A...

Description

Luca Mayer and Aleksander Christenson Professor Queenan MGSC 486 06.14.2019

La Quinta Case Study La Quinta Motor Inns is a mid-price, limited service segment in the lodging industry for guests who desire simple rooms and convenient locations over extra services like meeting rooms, inhouse restaurants, cocktail lounges or room service, and their target customers are frequent business travelers. They have hotels in Texas, Southeast, Southwest and Midwest, and employs over 5,800 people. Due to the rapid increase in population and real estate prices, Dallas is considered one of the best cities to do business in. Therefore, La Quinta Inns wants to expand its presence in the area. After realizing that their site selection process of hotels was both costly and inefficient, La Quinta decided to use a more quantitative approach for the site location of their next hotel in Dallas. The site selection committee narrowed the potential location to 6 places and then decided to get assistance from a quantitative system in order to gain more information in order to make a better decision. For this case study, we created a regression model to see which of the 6 locations in Dallas that would be the best for expansion. This allows us to compare each of the locations based on the chosen performance measure. We were given the option to measure the performance of the La Quinta Inn by occupancy ratio, profit, and operating margin. For the purpose of this project, to decide the next location for la Quinta Inns to build their new hotel in Dallas, we decided that the operating profit margin would be the beast measure. This is because operating margin accounts for profit, depreciation, interest, and revenue, giving us a picture on how well the operation of the hotel is doing and it is a more comprehensive measure than profit and occupancy rate. From the dataset we were given, 1983 proved to be a high-profit year for La Quinta Inns, while the result in 1986 was heavily affected by an economic downturn. The decline in results in 1986 seems to be more substantial and more of an outlier than the good results in 1983. We, therefore, chose to use the operating margin from 1983 as our dependent variable as we believe this result is a better representation of what future result might look like. After deciding the performance measure (dependent variable) and what year we wanted to use for our regressions, we had to select the independent variables we believe have the most explanatory power. We also tried to choose independent variables that have a low correlation among each other, so the variables do not capture the same information. We did a good job with this and we refer to the covariance matrix in the appendix. Knowing that proximity to local attractions affects a hotel performance and that 83% of La Quinta Inns customers are business travelers; we chose the variables that in our opinion capture these elements. The 10 variables we used are: *The equation can be found with the regression output which includes the P-values as well. ** To have an operational margin of 109% is not possible and the issue will be addressed in the following page.

 NEAREST - Distance to nearest La Quinta Inn, because the data coming out of several La Quinta Inns, prove that so-called cluster strategy increases the overall market share and operating margins of each location.  TRAFFIC – Traffic count (Traffic/hour). Much traffic could be good for the hotel because it means that there are lots of traveling people in that area, which improves the chance of the hotel to fill its rooms.  COLLEGE – College enrollment. Colleges host lots of events and students who are there all year round are likely to get visits from parents or friends. There can also be lots of businesspeople coming to the area due to the many businesses that usually surrounds colleges.  HOSP1 – Hospital beds within 1 mile, some customers are probably staying at the hotel only for the purpose of visiting the hospital close by. Also, if the hospital has long term patients and these patients has family or friends visiting regularly, a good and cheap hotel close by could be the best choice for them.  RMS1 – Hotel rooms within 1 mile. If there are lots of hotel rooms in the close area means usually that there is a high demand for hotel rooms.  OFC1 – Office space within 1 mile. Once again, it is all about convenience. Furthermore, as mentioned earlier business travelers are the main target of La Quinta Inn. Business travelers want to stay in a hotel close to their office space and the more office space within a mile, the more business travelers. We also think most of the business travelers prefer La Quinta Inn, due to having fewer amenities like meeting rooms or room service, which most of them don´t use due to the limited time spent in the hotel room. This allows La Quinta it decreases its price, without affecting the cleanness of the rooms and its good service.  PASSENGER – Airport passengers enplaned daily. The more passengers that enplane daily, the bigger is the probability for a higher demand for hotel rooms.  DISTCBD – Distance to downtown. Most people do not want to pay for an Uber, or other public transport to get downtown. Customers like the convenience of having the hotel nearby and being able to walk downtown.  MALLS – shopping mall square footage. The bigger the mall the higher the number of customers visiting the mall, and potentially more people needing a hotel room. The site selection committee has also mentioned this as an important factor when deciding on where to build their next hotel.  ACCESS – Accessibility (0 poor, 10 excellent). Similar to DISTCBD this is about convenience. If the hotel is hard to locate, customers are more likely to choose a more convenient hotel. With the chosen independent variables above, we used linear regression to create a model to predict the operational margin and got the following equation*.Operational Margin= 5.36 + (-.29*Nearest) + 0.25*Traffic + 0.0014(College) + 0.07*(Hosp1) + (-0.0008*RMS1) +(0.0003*OFC1) + (-0.0003*Passenger) + (-0.557*DISTCBD) + (-0.0014*Malls) + 2.15* Access. After inserting the values for each of the coefficients, the model tells us that location D, Dallas—Southern Methodist University is the best place for La Quinta Inns to build its next hotel in Dallas, with an OM of 109.4**. Something worth noting is that College and Passengers have the same value for all locations and will not help us to differentiate between the locations, but we found it necessary to include them for the purpose of creating a robust model. *The equation can be found with the regression output which includes the P-values as well. ** To have an operational margin of 109% is not possible and the issue will be addressed in the following page.

The regression equation has a R^2 of 0.975 and an adjusted R^2 of 0.97, which takes into account for having multiple independent variables. An adjusted R^2 of 0.97 tells us that 97% of the variation in the operating margin is explained by the independent variables selected. With a standard error of 5.18, the model predicts that the operating margin will on average be between 104.22 and 114.58. In other words, the predicted operation margin will on average deviate with about 5%. The independent variables that were found to be significant (P-value...