Skip to content

Statistical Data Tools for Monitoring Commuter Train Lateness

Notifications You must be signed in to change notification settings

Xophmeister/stupidTrains

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Stupid Trains

I hate commuting.

I am at the mercy of South West Trains and Network Rail, whose service is so often delayed, I decided it was time to monitor their lateness. This started off as a LibreOffice spreadsheet, but I'm a developer, dammit! We need an over-engineered solution.

If your hate of commuting is similarly offset by the desire to publicly humiliate those responsible with hard statistical data, this is the repo for you.

Database

The database is SQLite-based and is used to model train routes and journeys taken upon them. It contains a number of views that are used by the statistical analyses routines (in R). A full overview of the database schema can be found in the project wiki.

The database supplied with this repo contains the data that I have recorded. It will be updated every time I push from master, but not in perpetuity (contrary to popular belief, I have better things to do). If you want to collect your own statistics, you can use the nuke tool.

Update Tool

Usage: update [OPTION] RECORD

The record definitions are similar to regular expressions; I'm sure it's clear enough. Note that data must be entered in the correct format, otherwise the tool or the database constraints will complain.

-log

Record: DATE FROM SCHEDULE TO ARRIVE (EXCUSE CANCELLED?)?

Append a journey log entry:

  • DATE Journey date (YYYY-MM-DD)
  • FROM Origin station code
  • SCHEDULE Scheduled departure time (HH:MM)
  • TO Destination station code
  • ARRIVE Actual arrival time (HH:MM)
  • EXCUSE Guard's excuse for the service, if any
  • CANCELLED 0 = No (default), or 1 = Yes

If the ARRIVE time is after midnight, you must also specify a date component (YYYY-MM-DD HH:MM).

Note that, as the most common use-case, -log can be omitted.

-route

Record: ORIGIN DEPART (STATION ARRIVE)+

Create a new route from the ORIGIN station code, at the scheduled DEPART time, to each of the following STATION codes and advertised ARRIVE time tuples. There must be at least one destination tuple; arrival times must be greater than the departure time, but destinations needn't be in order. The route will be attached to the most recent origin; if one can't be found, you will be prompted to supply its validity dates and it will be added. Station codes must, however, exist.

e.g., WAT 08:00 WIM 08:10 KNG 08:30 is the 8am from Waterloo, arriving in Wimbledon at 8:10 and Kingston-upon-Thames at 8:30.

If the arrival time is after midnight, you must also specify a date component, the day after the Unix epoch. I'm too lazy to write logic that works this out automagically.

e.g., WAT 23:56 CLJ "1970-01-02 00:12"

-station

Record: CODE DESCRIPTION

Add a station to the database. Official codes needn't be used, providing you're consistent and stick to the three character format.

-origin

Record: ORIGIN DEPART START FINISH?

Adjust a route's ORIGIN station code and scheduled DEPART time's validity period: from midnight on START, to midnight on FINISH (which may be omitted if this information isn't available).

-json

Parameter: FILENAME+

Import data from a JSON file(s). See the project wiki for the schema.

Statistical Analyses

The database stores the scheduled arrival times of trains on a route. It must be noted that South West Trains doesn't publish arrival times for minor stations, so we must use the published departure times. As one generally departs after arriving, this skews the statistics in South West Trains' favour.

My protocol for recording data is as follows:

  • All times are as recorded by the station clocks, which are presumed to be correct and synchronised.
  • Seconds are rounded: >=30s is the next minute. If there is any ambiguity (e.g., if I can't immediately see a clock and it's very close), I will round down.
  • If I miss the train that I was intending to take, due to my lateness, my intended train becomes the next one that is appropriate.
  • The scheduled departure time is that of my intended train.
  • If a train is cancelled before departure, the scheduled departure time is that of my intended train (presuming I'm there in time to catch it), rather than its replacement.
  • If a train is cancelled during the journey, the scheduled departure time remains as that from the origin.
  • In the case of cancellations, I am to stick to the standard overground route, rather than attempt a combination of tube and bus routes: Such an alternative will not be recorded. An exception to this rule is in the event of a rail replacement bus, which will be recorded as normal.
  • The arrival time is when the train's carriage door is openable.

[TODO]

License

Copyright © 2013 Christopher Harrison

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the Software), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

The Software is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. In no event shall the authors or copyright holders be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the Software or the use or other dealings in the Software.

About

Statistical Data Tools for Monitoring Commuter Train Lateness

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published