For our final project for the course “Digital Analytics,” I was joined by Jill Haddon and Kaylee Holland to analyze the flight data from various airlines and discern what important information could be gleaned from it. We then created seven graphic visualizations of the dataset and further analyzed what the information could mean for airlines and their corporations.
This world visualization maps a random selection of flight destinations from various airlines. This map is color coded, with darker colors representing more flights to this destination. While this data is limited and would be improved by a larger database, it helps airlines to predict which destinations are the most popular, and which flights may need special attention/assistance to run smoothly. Combined with other datasets here like flight route vs. departure day, this can reveal which routes are used most frequently and on which days, which can further assist in allocating resources and employees to different flights. It is essential for airlines to track their most popular destinations and how these destinations specifically are interacted with by passengers, such as with extra flight accommodations, length of stay and day/hour of departure. This will help to optimize productivity for the airlines, and the experience of the passengers.
This grid of pie charts graphs the purchase of various flight accommodations (extra baggage, preferred seat and in flight meals) compared to the duration of the flight. This knowledge will allow airlines to properly allocate resources and support to their flights depending on the popularity of the flight accommodations. The flight duration was rounded to the nearest integer for simplicity of viewing. As detailed, the most popular length of flight for any accommodation is 9 hours, with the purchase rates being roughly the same from 5-7 hours long. 8 hours has the least amount of accommodation purchased. This shows that passengers will tend to give themselves a more luxurious and comfortable experience when the length of flight surpasses 8 hours. There appears to be a mental deterrence to purchasing flight accommodations for an 8 hour flight, which is worth noting when allocating resources. Lastly, extra baggage is by far the most popular flight accommodation, followed by in flight meals. Note: The rate of purchases for 10+ hour flights is very limited as the dataset did not have very many flights of this length, while comparatively had similar amounts of flights from 5-9 hours long.
This scatterplot shows a strong, negative, linear association between the length of stay and flight hours. There appear to be a few outliers in the data, but overall, it is a steady set. In general, length of stay refers to the duration of time a person stays in a particular location, such as a city or country, while flight hours typically refer to the amount of time spent in the air during a flight. When analyzing this data, it is important to consider the differences in traveling. Some may be traveling for business which would result in shorter stays, but often loner flight hours due to frequent traveling. Others may be traveling to multiple destinations in one trip which would result in shorter lengths of stay but longer flight hours overall. Those who must take connecting flights may also accumulate longer hours. Vacationers often take up fewer flight hours because they usually only need to take one or two flights to reach their vacation destination and return home after their trip. Analyzing the relationship between length of stay and flight hours can provide valuable insights for various stakeholders, including travelers, travel agencies, airlines, and tourism boards. Airlines can use this information to better optimize their flight schedules and routes. Travelers can also make more cost-efficient decisions by being able to see the data.
This stacked bar chart shows the hierarchy of routes based on the day and duration of the flight. Many of the routes and durations of flights are similar. However, as the chart goes on the length of the flight gradually decreases. Analyzing this data can help both passengers and airlines in making smart decisions on route planning and departure times. Airlines can enhance operational efficiency, improve customer satisfaction, and maintain a high status in the flight industry. Passengers can make more informed travel decisions by seeing possible delays.
A purchase lead typically refers to a potential customer who has expressed interest in booking a flight but has not yet completed the purchase. A purchase lead may have initiated the booking process by searching for flights, selecting travel dates, or even entering payment information but has not finalized the transaction by completing the purchase. In the data above, you can see that it is comparing purchase leads and length of stay at the destination they are purchasing. While before we were looking at what data point people were purchasing their flights from. The pink dots represent purchase leads on the x-axis and the y-axis looks at the time stayed in specific destinations. Looking at the specific data, as the purchase lead increases, the length of stay also increases.
The World (numeric shading) chart above shows where booking origins are on the map. The map above shows exactly where flights are being booked from. The top 5 countries where people are booking flights from with the most, Australia with 17,872, Malaysia with 7,174, South Korea with 4,459, Japan with 3,385 and China with 3,387 booked flights. By analyzing the data, it is very beneficial to our audience of marketing teams who would want to see where exactly people are booking their flights from could benefit most from the top 5places people are booking flights from. When comparing the World chart to the Flight Route vs. Departure Day, an airline marketing team could see where exactly people in the World are booking flights, where they plan to go, and the day which it is most popular to depart on.
The pie chart above is used to represent the average amount of passengers from the top 5 countries on flights. Bhutan (in purple) and Guatemala (in blue) both have the highest average with 3 people per flight. Then China (in orange) has an average of 2.07 passengers per flight. Russia (in yellow) has around 2.03 customers per flight and the fifth country is Ukraine with 2.2 passengers per flight in the pink. This data is important to analyze and compare with the flight accommodation map by following the trends on what specific accommodations people need from different countries and preparing for those accommodations in advance.
Leave a Reply