ANLY482 AY2016-17 T1 Group2: Project Findings

Home

Team

Project Overview

Project Findings

Project Management

Documentation

Mid-Term	Finals

Data Exploration

Throughout our Data Exploration phase, our team analyzed the variations in demands, which is defined as number of tickets sold, and this section will bring you through the interesting findings in relation to our project.

• Overall Demand Analysis

Initially, our team plotted the demand of all the events, and even though we realized that there are variations, however these variations are not really insightful.

(Picture HERE)

Thus, we drilled down further to look at the different classification of events. Our team has decided to classify the events as per the image below, as the differences in the event frequency will potentially affect the demands being analyzed.

The graph shows that the demands for the different types of events peak differently. For example, for the regular events, which are held on a annual basis, peak during April of every year and this is caused by the demands of Coachella at California. Similarly, we also found out that the demand for seasonal events peaked during June period for year 2010, 2012, 2014 and 2015. And this is due to the International Music Festival and Hardwell World Tour Concerts.

(Picture HERE)

• Control Chart Analysis

As from the previous sub-section, our team wanted to analyze the demands further for each specific event, and thus we selected a few events to analyze the demands using Control Chart.

The Control Chart is used to analyse the demands in a time-series manner, and it allows us to visualize the movements of the demands throughout a specified time-frame. The red-lines determines the upper and lower control limits, which signifies data points out of the normal-range (at 3 standard deviations away), and in this case we are more interested in looking into the upper control limits boundary. The green line signifies the average movements of the data points.

In this analysis, we will be looking at the specific events of different event types as below:

1. Seasonal Event - International Music Festival 2010 & 2014
2. Regular Event - Coachella at California 2010 to 2012
3. Regular Event - Bonnaroo at Tennessee 2010 to 2012

The main objective of the analysis is to find a pattern as to how the demand changes over time in a specific event. In addition, we are also motivated to find out at which point of time the demand is peaking at.

• Seasonal Event - International Music Festival

Seasonal Event - International Music Festival Control Chart Analysis
Analysis: Each data point within the Control Chart signifies a single performance within the International Music Festival, and it is arranged in a time-series manner. With reference to the Control Charts above, our team discovered that most seasonal events, such as the International Music Festival, have shown that the demands have an sudden spike as the the date gets closer to the closing stages of the International Music Festival. The extreme spike in demand towards the end of the of the International Music Festival may have caused a sales bottleneck for TixCo.

• Regular Event - Coachella at California 2010 - 2012

Regular Event - Coachella at California Control Chart Analysis
Analysis: Each data point within the Control Chart signifies a single performance within Coachella at California, and it is arranged in a time-series manner. The first observation in which the team observed is that there is no clear pattern in the increment in demands and demands seem to fluctuate as time progresses. Secondly, there seems to be alot of spikes or demands above the Upper Control Limits, and this is very different from what we have seen previously in the Seasonal Events. Brushing the data points above the Upper Control Limit (UCL), reveals that the more popular bands commanded a better demands.

• Regular Event - Bonnaroo at Tennessee 2010 - 2011

Regular Event - Bonnaroo at Tennessee Control Chart Analysis
Analysis: Each data point within the Control Chart signifies a single performance within the Bonnaroo at Tennessee, and it is arranged in a time-series manner. Similar to the Regular Event Coachella at California, there seems to be no clear pattern in the increment of demands, the demands seems to fluctuate very much as time progresses throughout the event. However, our team has noted that the overall demands is deemed to be much more stable, as there are lesser spikes (e.g., over the Upper Control Limit) in demands. Unlike Coachella in California, it seems that the spiked in demands seem to involve bands at random.

Summary of Data Exploration

As we per our team's analysis, we realized that there seems to be a multitude of factors, which will potentially affect the demands of each event. An example would be Coachella at California is showing high relation in terms of popularity of bands involved and tickets sold, however the Bonnaroo at Tennessee is showing another set of trends in which our team are not able to determine the correlated factors.