casino.com - internet casino

AN UPDATED TRIP GENERATION STUDY FOR LAS VEGAS HOTEL/CASINOS IS PRESENTED IN THIS FEATURE USING THE NUMBER OF GAMING POSITIONS AS AN INDEPENDENT VARIABLE. REGRESSION ANALYSIS USING GAMING POSITIONS AND PREVIOUS INDEPENDENT VARIABLES (EMPLOYEES, CASINO FLOOR AREA AND HOTEL ROOMS) WAS PERFORMED TO DETERMINE THE "BEST" SINGLE OR MULTIVARIABLE EQUATIONS.

INTRODUCTION

The hotel/casinos of Las Vegas, NV, USA, provide an attraction destination for visitors from locations worldwide. Many of the world's largest hotels providing more than 3,000 rooms at a single facility are located in Las Vegas. The number and proximity of these resorts provide Las Vegas with thousands of hotel rooms and gaming positions unlike anywhere else in the world. Understanding the trip generation characteristics of these facilities is key to determining associated traffic impacts and roadway mitigation measures.

This trip generation study was motivated by the work conducted by Kenneth W Ackeret and Robert C. Hosea III as presented in the May 1992 issue of ITE Journal in "Trip Generation Rates for Las Vegas Area Hotel-Casinos."1 Ackeret and Hosea calibrated the number of vehicle trips generated by a hotel/casino as a function of casino floor area, number of hotel rooms and employees. The results of their study are the basis for the current method of estimating the number of vehicle trips generated by Las Vegas hotel/casinos. However, this method is based on a relatively limited data set. The data used were collected in and prior to 1990 for 21 hotel/casino counts. Since then, numerous hotel/casinos were constructed that conducted facility driveway trip counts within the Las Vegas metropolitan area. This increased data consisting of 46 hotel/casino vehicle generation counts is expected to provide more statistically significant results in trip-making characteristics. Likewise, it is believed that the independent variable of the number of gaming positions would be a good variable for estimation of trip generation. In other areas of the country, gaming positions have been used as an estimator of hotel/casino trips. However, the numher of gaming positions as compared with expected trip generation has not been studied within the Las Vegas area. Therefore, one hypothesis tested in this feature is that the number of gaming positions is a good estimator of trips generated by hotel/casinos.

METHODOLOGY

The vehicle trip generation of a hotel/casino is dependent on various factors. Certain factors, such as size, location and type of hotel/casino all contribute to the trip generation of a facility. Facility size characteristics can be identified as the number of hotel rooms or the amount of casino floor area. Likewise, the number of gaming positions within a casino and/or the total number of employees can be factors affecting the number of vehicle trips generated by hotel/casinos.

The conducted study developed a set of regression equations for estimating the number of vehicle trips generated by a typical Las Vegas hotel/casino property, existing or proposed. The regression analysis was conducted using data obtained from various existing hotel/casinos that had vehicle generation counts conducted during the a.m. and p.m. peak hours to determine the number of vehicles entering and exiting individual casino properties. The following independent variables were used in the analysis:

* Number of gaming positions;

* Number of employees;

* Number of hotel rooms; and

* Casino floor area.

Data Collection

To determine equations that accurately estimate the number of peak-hour vehicle trips generated by a hotel/casino, an extensive amount of data was required. The data needed included the typical a.m. and p.m. adjacent-street peak-hour facility driveway count volumes, number of gaming positions, number of hotel rooms, number of employees and casino floor area of each hotel/casino. The data used by Ackeret and Hosea1 and the "Trip Generation Analysis Report,"2 conducted by the University of Nevada-Las Vegas, were supplemented with data collected more recently and used for this study.

Existing Count Data

The first step of the additional data-- collection effort involved obtaining count data conducted at facility access driveway locations for existing hotel/casinos. The count data obtained were from counts conducted at all facility driveways during 15-minute intervals on a weekday during the typical a.m. and p.m. peak-- hour periods. Therefore, the count volumes collected reflect the highest one-hour period between 7 a.m. to 9 a.m. and 4 p.m. to 6 p.m. on a weekday. It is recognized that the facilities' trip generation may be greater during other hours of the day or week. The typical weekday peak-hour time periods were studied based upon availability of data that was collected in accordance with local requirements for analysis during these time periods. Limited proprietary data exists, which indicates the peak hour of the casino generator generally occurs around 7 p.m. At this time, the trip generation of the hotel/casinos is generally 15 percent greater than during the typical commuter peak hours. This increase should be recognized when analyzing specific traffic demands at intersections comprising primarily hotel/casino traffic. The count data may be obtained by requesting it from the authors.

Independent Variables

The independent variables associated with each of the hotel/casinos were collected for the time period corresponding to the date of the reported field traffic counts. The specific traffic studies reports and ITE Journal feature used for the count data were also reviewed to extract information on the number of employees, hotel rooms and casino square footage. Data, other than the number of gaming positions, were also obtained from the Las Vegas Perspective,3 the Top Rank Nevada 1999, Statewide Book of Lists4 and the Book of Lists.5 Data that were still not available from these sources were obtained from telephone surveys with the hotel/casinos management. The gaming position data were obtained from the State of Nevada, Gaming Control Board for the current and historic count periods at each of the hotel/casino sites. The independent-variable data may be obtained by requesting it from the authors. The following are definitions of the key independent variables studied:

* Number of employees: The total number of people employed at a hotel/casino for all shifts;

* Casino floor area: The square footage that contained the gaming positions;

* Number of gaming positions: The number of gaming positions was defined to be the total number of slot machines within a casino plus the number of table games multiplied by seven positions per table. The average of seven gaming positions per table is consistent with the definition suggested by Finigan6 for typical gaming operations throughout the United States; and

* Number of hotel rooms: The total number of hotel rooms at the facility.

Location Classification

The hotel/casinos evaluated were divided into two groups based on their locations. A hotel/casino was considered to be a resort corridor facility if it was located within the Clark County Regional Transportation Commission's defined area bound by Valley View Boulevard on the west, Maryland Parkway on the east, Washington Avenue on the north and McCarran International Airport on the south (Robindale Road). These boundaries, which include the famous Las Vegas "Strip," provide the limits of the accepted resort corridor of Las Vegas. A hotel/casino not located within the resort corridor was classified as a local facility. This classification was necessary to accurarely distinguish the trip generation characteristics of these two distinct groups of hotel/casinos. Resort-corridor facilities experience significant nonvehicle trip sharing due to pedestrian walking trips as a result of the synergy created by the close proximity of the hotel/casinos. Local facilities primarily have vehicle trips and minimal pedestrian trips associated with them. This is due to the location of these resorts being outside of the resort corridor and away from other hotel/casinos.

Directional Distribution

A directional distribution of entering and exiting trips during the peak hours was calculated based upon the observed data. The average directional distributions were calculated separately for resort corridor and local hotel/casinos, as well as the a.m. and p.m. peak hours.

Average Trip Rates

Based upon the data collected, average trip rates were calculated for both groups of hotel/casinos to determine the average number of vehicle trips expected per independent variable. The average trip rates and standard deviation were calculated for each independent variable. An average trip rate was determined to be acceptable if the standard deviation was less than 10 percent of the average rate as described in the Institute of Transportation Engineers' (ITE) Trip Generation Handbook, An ITE Proposed

Recommended Practice.7

Regression Analysis

Regression analysis was conducted to develop regression equations for estimating the number of trips generated by the casinos as a function of several combinations of independent variables. The stepwise regression procedure was used as outlined in Basic Econometrics by Damodar Gujarati.8

In this study, regression analysis of the data was used to determine the "best fit" equation of a line that most accurately estimates the number of peak-hour trips for both resort corridor and local hotel/casinos. Both linear and nonlinear equations were evaluated as provided in ITE's Trip Generation.9 For the nonlinear equations, the natural logarithmic and squared polynomials were used to determine which type of equation would best estimate the trip generation of a hotel/casino.

After equations were obtained for the single-variable equations, the t-statistics were first evaluated to determine statistical significance of the coefficients of the variables. A confidence level of 95 percent was used. However, since regression equations that provide intercepts are viewed favorably by practitioners to differentiate between average rates and regression equations, a lower confidence level of 75 percent was utilized for the intercept. If the intercept did not prove to be significant by the t-test, the equation was recalculated with the y-intercept being forced through zero. The resulting equations were compared to determine which type of equation was the "best" model for estimation of the peak-hour trips. An analysis of variance was conducted for each model as provided by Gujarati.8 To compare the models, the R-squared and F statistics were evaluated. The R-squared value demonstrated how closely the equation fit the data points. In general, R-squared values closest to 1.0 are desirable. Also larger values of the F statistic are desirable. After determination of the best equation for each independent variable, the data, average rate and "best" regression equation were plotted. To determine which independent variable estimated the number of peak-hour trips best, the F statistic was again compared for each equation.

Development of the "best" regression equation was done using the stepwise regression analysis procedure. In this procedure, the effect of adding a variable to an existing equation is evaluated using the partial F statistic. This statistic evaluates whether the contribution of the additional variable to the accuracy of the model is statistically significant. If different variables are added to the same equation, then the variable with the highest partial F that is statistically significant is selected.

RESORT CORRIDOR HOTEL/CASINOS RESULTS

Directional Distribution

The existing trips generated by the resort corridor hotel/casinos had an average distribution of 59 percent entering and 41 percent exiting during the a.m. peak hour. Likewise, during the p.m. peak hour, the directional distribution was found to be 49 percent entering and 51 percent exiting.

Single-Variable Equations

Table 1 provides the results for average trip rate and regression equations for the a.m. and p.m. peak-hour trips generated by a resort corridor hotel/casino. As these single-variable equations illustrate, all four independent-variable equations are statistically significant during both peak hours based on the critical F value. The equation with the highest F value was considered to be the "best" statistically significant equation to estimate trip generation. For both a.m. and p.m. peak hours, the number of gaming positions is the best estimator of the trip generation for a resort corridor hotel/casino. For the a.m. peak hour, the number of employees is the next best trip generation estimator, followed by casino floor area and, lastly, the number of hotel rooms. For the p.m. peak hour, the casino floor area is the second-best estimator followed by number of hotel rooms and then employees.

Average Trip Rates

The average trip rates were found to be acceptable for estimation of the number of trips generated by a hotel/casino for all studied independent variables, because the standard deviation was less than the established limit of 110 percent of the average rate.

"Best" Multivariable Regression Equations

As discussed earlier, the forward step-- wise regression analysis approach was adopted to determine appropriate multivariable equations. Starting with the "best" single-variable equation, containing the variable for the number of gaming positions, partial F statistic analysis was conducted to determine an additional variable that would statistically best improve the accuracy of estimation. The best two variable equations for both the a.m. and p.m. peak hours was found to be combining the variables for the number of gaming positions and number of employees. Therefore, the "best" equations to estimate the number of vehicle trips during the peak hours were found to be as follows:

Determining the Most Appropriate Equation for Practical Use

In practice, there are occasions when certain independent variables are not available for forecasting of trip generation of a facility. Typically, prior to the construction of a hotel/casino, the number of employees and gaming positions is not known. Or if the number of employees is known, it may be an estimate. It is important that the practitioner use caution when using equations with the number of employees prior to the construction of a hotel/casino to ensure the number of employees are estimated most accurately. In such a case, other variables may have to be used for forecasting. Normally in such cases, the number of rooms and casino floor area would be known. Table 2 is provided as a guideline to determine which equations to use depending on which independent-variable data is available. The equations in each row are provided in descending order of their statistical significance. The equation shown in the top row was found to be the most statistically significant equation. Therefore, if these independent variables are known, this equation should be used to make the best trip generation estimation for the facility.

LOCAL HOTEL/CASINOS RESULTS Directional Distribution

The existing trips from count data at local hotel/casinos had an average distribution of 57 percent entering and 43 percent exiting during the a.m. peak hour. During the p.m. peak hour, the directional distribution was found to be 52 percent entering and 48 percent exiting.

Single-Variable Equations

Table 3 presents the results for single-- variable regression analysis for the a.m. and p.m. peak-hour trips generated by a local hotel/casino. From the single independent-- variable regression analysis, three of the four variables provide statistically significant models for estimating the expected number of trips generated by a local hotel/casino during the peak hours. The number of hotel rooms was determined not to be a good independent variable for trip estimation. This is likely due to the relatively small number of hotel rooms provided at a local hotel/casino. Likewise, the patrons of a local hotel/casino primarily live in the Las Vegas area, and the patrons of the hotel rooms are not a major proportion of the trips generated.

Average Trip Rates

The average trip rates were found to be acceptable for estimating the number of trips for all independent variables, because the standard deviation was less than the established limit of 110 percent of the average rate.

"Best" Multivariable Regression Equations

Similar to the resort corridor hotel/ casino analysis, the forward stepwise regression analysis approach was adopted to determine appropriate multivariable equations. For both the a.m. and p.m. peak hours, it was determined that no multivariable equations are more significant than the single-variable equations. Therefore, the "best" equations to estimate the number of vehicle trips during the peak hours are as follows:

a.m. peak hour

Ln(T) = 0.710 Ln(X) + 1.506

R^sup 2^ = 0.77

where

T = number of a.m. peak-hour trips; and

X = number of employees.

p.m. peak hour

Ln(T) = 0.794 Ln(X) + 1.278

R^sup 2^ = 0.83

where

T = number of p.m. peak-hour trips; and

X = number of gaming positions.

Figures 1 and 2 illustrate these regression equations as well as the average trip rates, for the a.m. and p.m. peak-hour "best" equations, respectively. Likewise, Figure 3 is shown to provide estimation of trip generation for gaming positions during the am. peak hour. Other figures may be obtained by requesting them from the authors.

Determining Most Appropriate Equation

Similar to the resort casinos, for practi

cal use, there are occasions where certain independent variables are not available for estimating the trip generation of a hotel/casino facility. Typically prior to the construction of a hotel/casino, the number of employees and gaming positions is not known. Likewise, the number of employees may be an estimate that may not be accurate to the actual number of employees after completion of the project. Therefore, the equation relating the casino floor area to peak-hour trips is believed to be acceptable for estimating trip generation. However, if certain variables are known, the following list provides the order to use the independent-variable regression equations to determine the best trip generation estimation for a local hotel/casino during each peak hour:

CONCLUSIONS

From this analysis, it was found that the addition of gaming positions as an independent variable was a significant improvement over previous developed equations. It is important to recognize that these trip generation equations were developed using data from Las Vegas hotel/casinos. As previously described, the trip-making characteristics of the resort corridor Las Vegas hotel/casinos are unique based on facility size and proximity to each other. Caution should be used when applying these equations to hotel/casinos outside of Las Vegas. However, the local hotel/casino equations developed may have applications to other gaming establishments throughout the United States. It is recommended that further study be conducted comparing these equations to other regional gaming facilities for the various independent variables.

References

1. Ackeret, K.W, and R.C. Hosea III. "Trip Generation Rates for Las Vegas Area Hotel-- Casinos." ITE Journal (May 1992): 33-37.

2. Souleyrette, RR., E.M. Parentela and S.K. Sathisan. "Trip Generation Analysis Report: Hotels-Casinos Within the Las Vegas Urbanized Area." Research Report, Transportation Research Center, University of Nevada-Las Vegas, May 1991.

3. Las Vegas Perspective. Las Vegas Review-- Journal, 1990.

4. Top Rank Nevada 1999, Statewide Book of Lists. Las Vegas, NV, USA: Nevada Business Journal, 1999.

5. Book of Lists. Las Vegas Business Press, Volume 6, 1996.

6. Finigan, DJ. "Transportation Planning for Casino Land Uses." Institute of Transportation Engineers 65th Annual Meeting, 1995 Compendium of Technical Papers, pp. 412-416.

7. Trip Generation Handbook, An ITE Proposed Recommended Practice. Washington, DC, USA; ITE, October 1998.

8. Gujarati, D. Basic Econometrics. McGraw-Hill Book Co., 1978.

9. Trip Generation, 6th Ed. Washington, DC, USA: ITE, 1997.

CURTIS D. ROWE, PE., PTOE, is an Associate with Kimley-Horn & Associates in their Denver, CO, USA, office. He has a bachelor's degree in Civil Engineering from the

University of Nebraska-Lincoln and a masters degree in Civil and Environmental Engineering from the University of Nevada-Las Vegas. He can be reached at curtis. rowe@kimley-horn.com. Rowe is a Member of ITE.

MOHAMED S. KASEKO, Ph.D., is an Associate Professor of Civil Engineering at the University of Nevada-Las Vegas. He has a master's degree from Cornell

University and a doctorate degree from the University of California-Irvine. His current research interests include transportation planning, operations and intelligent transportation systems. Kaseko is an Associate Member of ITE.

KENNETH W. ACKERET, PE., PTOE, is a Wce President with Kimley-- Horn &Associates in their La, Las Vegas, NV USA, office. He has a bachelor's degree in

Civil Engineering from the University of the Pacific in Stockton, CA, USA, and master's and doctorate degrees in Civil and Environmental Engineering from the University of Nevada-Las Vegas. He is currently Chair of ITE technical com mittee TPC-98-04, regarding transportation plan ning of hotel/casinos. Ackeret is a Fellow of ITE.

casino.com - internet casino

Menu