Schedules are a necessity for riding any form of public transit. They not only tell you when the next vehicle is coming, but they also govern the two major operating costs associated with transit: vehicle work assignments called blocks, and operator work assignments called a duty.
Public transit schedules started out and still are printed on paper. This means new schedules have to be printed for every change in arrival time or a route. But with the advent of trip planning apps, the humble paper public transit schedule is a new creation. It’s now a massive digital repository, which puts the world’s transit schedules at your fingertips. Despite the differences in scale, both types of schedules (printed and digital) are still made the same way: in a transit agency.
Step 1: Defining Service Characteristics
Traditionally, public transit schedules are made from a mix of travel times, desired route frequency, route maps, and other service characteristics. Planners work out service characteristics by first consulting with the public, operational staff, and decision makers. Planners usually use their judgement when setting input parameters, though some inputs may also be set by an agency’s service standards.
The following are the minimum service characteristics necessary for building a public transit schedule.
- Route alignment – where a route will go, and which stops will be serviced
- Running time – travel time from start point to end point. running times are usually defined between a pair of timepoints. Computer aided analysis use on board GPS and other equipment to help schedulers refine running times
- Layover – a combination of time for operator’s break, recovery time (a buffer to account for variability in running times), or increasingly electric vehicle charging time
- Cycle Time – running times in each direction plus layover
- Headway – frequency of a route expressed in minutes between trips (e.g., a route will come every 10 minutes)
- Span of Service – the time covered between the first trip of the day and the last trip of the day
Step 2: Timetable Development
Once these ingredients are collected, they’re given to a scheduler who assembles them into a timetable. Timetables define when each trip is scheduled to arrive at each stop (or station). Specialized scheduling software is used to help schedulers create timetables. However,
very simple schedules sometimes rely on Excel instead.
Schedulers must create a timetable for every planned variation in service characteristics. For example, there are different schedules for weekday and weekend service. Another common variation is called an exception, which is used for changes in a trip’s start time (e.g., due to an early school dismissal).
In theory, a separate timetable could be created for every day of the week to refine for small changes in travel time. However, a unique schedule for every little variation would be 1) too time consuming to produce, 2) too confusing for the public to use, and 3) too overwhelming for transit operators. Timetables are therefore rough approximations of typical day’s traffic delays, ridership, weather, etc.
The shelf life of a timetable is dependent on how often the scheduler can update it. In agency’s with low scheduling manpower, timetables remain unchanged for years.
Timetables are essentially what the public know as a public transit schedule. But timetables must still undergo a few more steps before they can be used by the transit agency.
Step 3: Vehicle Scheduling (or Blocking)
The timetable is then ready to be broken down and redistributed to a vehicle schedule in a process called “blocking”. Vehicle schedules are used to define when and where a vehicle will be throughout the day.
This was once an imprecise, labor-intensive, week-long process done entirely by hand, but now sophisticated software can accomplish this step is less than an hour for a mid-sized transit agency. Software also allows the agency to optimize the blocking process to minimize cost and the number of vehicles required.
But before the blocking can begin, the software needs to know:
- The total number of vehicles in each garage broken down by type (e.g., 40-foot, 60-foot articulated buses, etc.)
- Travel times from start points to end points for different periods throughout the day, which when organized in a table is called a deadhead matrix. Watch for a future blog post about using Google travel time estimates to fill in a deadhead matrix.
- Acceptable vehicles types for every trip. Passenger loads, and roadway constraints dictate what type of vehicle can be used.
- Other vehicle or route limitations like fuel tank size, number of vehicles that the layover location can fit, or a listing of routes that can be grouped (interlined) together
Some software packages can also shift a trip’s starting time slightly in order to reduce the number of vehicles required in the schedule.
Step 4: Crew Scheduling (Run Cutting and Rostering)
Now that the vehicles are accounted for in the public transit schedule, the transit agency must then assign work for their operators. This is done by run cutting. In the past, run cutting meant literally cutting blocks into pieces in order to create work assignments. Though this process is completely automated and optimized by software, transit agencies still use mostly the same terminology.
Just like in the blocking process, the software needs to know a few inputs before it can begin:
- Number of operators from each garage
- Relief points: locations where operators can switch off. Relief points can be either street reliefs (just the drivers change) or pull-reliefs (both the driver and a new vehicle change).
- Rules governing overtime, total work time, and other complex work restrictions such as spread time, split break time, guarantee time, platform time, sign-on time, and vehicle inspection time. These rules are usually defined by employee contracts, but can also be imposed by Local, State, or Federal laws. The glossary of TCRP Report 135 describes these terms in more detail .
Once the blocks have been cut into pieces, they’re assembled into daily runs. These runs undergo another optimization process called rostering, which groups them into weekly work assignments called duties. Operators sign-up for duties at a sign-up (also called a “bid,” “pick,” “line-up,” or “shake-up”).
Because sign-ups occur periodically throughout the year, it regulates how often a timetable can be updated. Most agencies have sign-ups two to four times a year. This means if you wanted to have a timetable for every month in the year, you’d need 12 sign-ups. This is possible, but not very practical for operators and the public.
You now understand the basic four-steps of the public transit scheduling process:
- Defining service characteristics – collecting the ingredients for the schedule
- Timetable development – creating the building blocks a public transit schedule
- Vehicle scheduling – assigning work to vehicles
- Crew scheduling – assigning work to operators
With this overview, you have a new lens from which to view advances in technology such as self-driving buses, big data, and electric vehicles. I’ll talk about these “disruptions” and how they will affect the future of the transit world in an upcoming blog.
If you want to dive deeper in the the public transit scheduling process, you should read the Transit Cooperative Research Program Report 135: Controlling System Costs: Basic and Advanced Scheduling Manuals and Contemporary Issues in Transit Scheduling. It’s free, and a great resource.