Maven Rail Challenge - An Exploratory Dashboard

METRICS

• Passengers is the number of tickets sold (demand), not the actual passenger travelling.
• Revenue is the revenue exclusive of the refund amount (measured in GBP).
• Trips is the number of trips. A trip has 4 unique information: departure/arrival stations, date of journey and departure time.
• Revenue per Trip is the revenue earned for each trip.
• On-times is the number of on-time trips. % of On-time is the percentage of on-time trips.
• Delays is the number of delayed trips. % of Delays is the percentage of delayed trips.
• Cancels is the number of cancelled trips. % of Cancels is the percentage of cancelled trips.

DASHBOARD STRUCTURE

The dashboard is divided into 3 main reports and 1 instruction page. Each main report has navigation buttons at the top right, which facilitates the move between reports.

1. Instructions:

This provides instructions on the navigation of the dashboards, the definitions of key metrics used and the meaning of symbols that appear in the reports.

2. Overview:The overview report provides overall information: 7 overall metrics are put on top including total number of routes, passengers, revenue, trips and the numbers of on-time, delayed and cancelled trips.

Below are 4 bar charts with their own purposes. First, the peak travel time is illustrated by calculating the total of passengers hourly, with the help of average line and columns colour to emphasise the idea. The second chart (below the first one) provides the number of delayed/cancelled trips by each reasons, which gives the users what is the most common reasons of those situations. The remaining two graphs come from the same idea of seeing how the number of passengers and revenue were distributed among each category of categorical variables. 5 features are split into two group, Ticket and Purchasing Behaviour. With the 4 graphs, the last 3 tasks are addressed, yet just the tip of an iceberg. Deeper explorations are presented in Route and Revenue Reports, along with two explaining ones.

3. Route Report:

The aim of this report is to provide more insights on each route (table) and station (bar chart), which address the first task, along with two comments on the abnormality then lead to the question at the bottom right, creating a storyline. The table gives details of each route, including the number of passengers, trips, revenue, revenue per trip, % on-time, % delays and % cancels. The table is sort in descending order by default to link to the first comment. The second comment can be taken by sort the % Delays column. More explorations can be done by sorting other columns as well. The bar chart visualises the number of passengers in each departure/arrival station. The users can choose type of station on the slicer above. The question at the bottom right leads the users to the answers where the second comment left off. By clicking to the question mark symbol, another report is shown, analysing detailly the delays of each route.

The average delayed minutes metric is provided, along with the % of those delays by 4 controllable reasons, Staffing, Staff Shortage, Signal Failure and Technical Issue. The reason of the above abnormality is shown, however, more explorations can as well be done by sorting columns, which might help recommend actions to decrease the delays of each route.

4. Revenue Report:

Overall revenue distribution between categories were shown in overview report. In the Revenue Report, abnormal event of revenue among time is illustrated and delved into instead, also with the help of the storyline followed by another explaining report.

INSIGHTS

1. The route from Manchester Piccadilly to Liverpool Lime Street and the return were among the most popular route, with over 3000 demands. 4 other routes starting from 4 stations in London were also on the list, leaving the rest having less than 1100 passengers in total.

2. Peak travel time were 6-8AM and 4-6PM, corresponding to times when people commute to and back from work. Furthermore, there was seasonality in number of passengers purchasing off-peak ticket, whose demands rose in weekends.

3. In terms of ticket class, most of the revenue and number of passengers were from standard class. Regarding ticket type, although the price for each type is in the order as shown below, the revenue is in the reverse, with Advance ticket generated most of the revenue. Interestingly, there were changes in revenue distribution among ticket type in February, resulting from the gradual shifts of purchasing behaviours from Anytime and Off-peak to Advance ticket. Further market insights is needed to determine the factors of this market shift, which might include promotions, policies, etc.

4. The most common culprit for trip cancellation was Signal Failure, which, along with Technical Issue and Weather, were also the common reasons of delays. However, looking closely to the route level, there were some route with high percentage of delays and 100% of delays by one factors.