Bus travel time comparison before/after stop manipulation

[RaenonX]

Continue from #8. Blocked by #25.

• DO NOT group the stops when calculating this because traveling in a different direction produces different results.

[yayen-lin]

Type: "Moved to #27" for this single comment and remove all the contents below after you've checked that these contents are double-checked.

These detailed things should be inside a different issue.

I am also removing "references" label for this issue because this issue does not/should not contain any data reference. The label was meant to find the data.

---

Goal:

Premises:

• There are currently 47 available routes in the City of Madison.
• When we are to consolidate bus routes, we can look at the matrix (from dev-andy/report_andy/Oct_5th/Overlap_Percentage_cmap_Blues_t5.png) as well as the plot that we're designing in this issue.
  • (I personally consider consolidating bus routes would be a much deeper work).

Unresolved issues:

• We're still figuring out how to sort bus stops by its route.
  • sort by longitude won't work.
  • data/references haven't been found.
  • worst case:
  • might have to manually input those stop names
• Too many bus stops to be put on a graph.
  • proposal to solution 1:
  • For route 2, there are 122 stops in total, and divided by 2 directions we have 61 stops in each direction. It also splits the whole route by 9 published stops (iow, main stops I guess) into 8 smaller sections. Click here to view them in a map.
  • And plot those 9 published stops on to the x-axis.

Data Mining and Things To Consider:

• Line Plot Part:

  • Travel time from one stop to another.
  • This can be computed through this link; for example, route 2 seems to average 51 mins from one end to the other.
    • We can look for the table tag from the page source.
    • table 0: West to North, Weekday
    • table 1: West to North, Weekend
    • table 2: West to North, Holiday
    • table 3: North to West, Weekday
    • table 4: North to West, Weekend
    • table 5: North to West, Holiday
    • Split by 9 published stops, we can look for each section and calculate the travel time between each stop using distances obtained by the haversine formula. EDIT: (I'm thinking using euclidean distance because we're not looking at a very large region)
    • Apply element-wise operation on Lat and Lon columns to optimize runtime.
    • Then, we can look for:
    • stops that are too close to one another to improve on time-traveling of that route.
    • stops that overlapped with other routes and have a less number of boarding to reduce traffic.
    • May also consider:
    • Direction: West/East
    • Service Type: Weekday/Weekend/Holiday
    • Number of Boarding: see below

• Bar Plot Part:

  • We can calculate the number of passenger boarding using files from #1 (Daily Boardings Based on Routes and Bus Stops) and this link to calculate the boarding of each individual bus launch
  • May also consider:
  • traffic condition using the files from #1 (Pedestrian and Traffic Flow)
  • tbd