Analysis of Activity Data Set of Restaurant Guest Counts
Analysis of Activity Data Set of Restaurant Guest Counts
Introduction
The present paper discusses a business scenario under the background of business data collected to inform the managerial decisions for the purpose. As per the scenario, the decision maker being a manager of a small town café which enjoys robust guest inflow during the summer months and a small population from among the residents are visitors to the same during the off season is faced with staffing decision problems. The season typically starts around the Memorial Day each year. The café adjusts its staffing based on the guest count during the period and is based on the efficacy and capability of the staff as shown below.
One wait staff with one bus staff serves 75 guests.
At 75 guests the restaurant employs another wait staff employee making the staffing to two wait and one bus staff employee.
The paper analyses the data from the manager’s perspective and provides relevant insights for effective staffing planning for the Restaurant manager.
Discussion
Suitability of Data
As far as suitability of data is concerned, there is no data for off season to plan or predict the staffing according to guest inflow. This limits the use and applicability of the dataset for decision making. Moreover, the data for summer months is also not complete as the data reflects just about 15-20 days and not even a month of data. It is important to have a monthly at least for the peak season months which would mean at least two months data as well as for the off seasons which would entail at least 4-6 months data. This is required to get an optimum picture of the guest inflow during peak and off seasons and also to predict the staffing accordingly. Quite significantly there could be certain days during the off season as well that might witness more than 75 guests such as maybe the weekends or wedding seasons and thus might require better staffing accordingly.
The data is not current in the sense that it is from 2012 to 2014 and does not include last year’s data that is 2015. Also it would have made more sense to have five instead of three year data for better sufficiency and accuracy.
Factors Affecting Validity of the Dataset
The dataset when evaluated for Face validity does appear to be appropriate and relevant in terms of its overall method of collection as the data has been obtained from the management reports which employ objective, systematic and accurate means to collect and report data. The data intends to measure the workload as related to the requisite staffing and seems to be an appropriate measure of the same. The intended measure of the staffing level can be easily worked out using the data set and the guest count staffing relationship elaborated above. Thus the data set is relevant and appropriate to the business scenario. However, the validity can be affected by the span over which data is collected, what is considered as peak and off season and as to how current the data is.
Factors Affecting Reliability of Dataset
Graphical Representation of Data
The data is graphically represented through two chart representations. The mean data is first worked out as shown in Table 1 (Appendix 1.1) which shows the mean guest count during the peak period (May 20 to June 7) during a year including the Memorial Day for three years. The Chart 1 (Appendix 2.1) is the graphical representation of Table 1.
The Table 2 (Appendix 1.2) shows the Frequency Distribution of the data set and Chart 2 (Appendix 2.2) is the histogram for the distribution.
Reasons for Selection of Chart Type
Chart 1 (Appendix 2.1)
Chart 2 (Appendix 2.2)
Measures of Central Tendency and Variability
As shown in the Table 3, the mean, mode, median, standard error, skewness and kurtosis for the dataset are calculated and conclusions drawn thereby which are reported hereunder.
Conclusions
Data is negatively skewed which means the higher levels of guest count are unevenly concentrated in the last few days of the peak season and thus require enhanced staffing correspondingly.
Data has a negative kurtosis and thus as apparent from the Histogram, the data distribution has lighter tails meaning thereby that some extreme data is missing especially at the higher end this validates the initial observation that the data is inadequate to measure what it intends to measure and raises question over its reliability and accuracy.
A high standard deviation shows random variability for the three years data and hence supports the reliability of the data .
The positive difference between the median and mean values of the data shows that data is skewed in favor of greater values and is not normally distributed .
References
Mcclusky, A., & Lalkhen, A. G. (2007). Statistics II: Central tendency and spread of data. CEACCP, 127-130.
Pierce. (2007). Evaluating Information: Validity, Reliability, Accuracy, Triangulation . Retrieved June 18, 2016, from http://www.sagepub.com/: http://www.sagepub.com/sites/default/files/upm-binaries/17810_5052_Pierce_Ch07.pdf
SPC for Excel. (2016, February). Are the Skewness and Kurtosis Useful Statistics? Retrieved June 18, 2016, from https://www.spcforexcel.com: https://www.spcforexcel.com/knowledge/basic-statistics/are-skewness-and-kurtosis-useful-statistics
Appendices
Appendix 1 Tables
Appendix 1.1: Table1. Mean Guest count for three years from 2012 to 2014
Appendix 1.2: Table 2 Frequency Table of the daily Guest Counts during the period May 20 to June 7 for three consecutive years (2012-2014)
Appendix 1.3: Table 3. Measures of Central Tendency and Variability
Appendix 2 Charts
Appendix 2.1: Chart1. Bar Chart showing Mean Guest Counts (Lunch + Dinner) for three years from 2012-2014 between May 20 and June 7
Appendix 2.2: Chart2. Histogram showing the frequencies of the daily Guest Counts during the period May 20 to June 7 for three consecutive years (2012-2014)
