Measuring the quality of public transit system

1
Measuring the quality of public transit system

• Tapani Särkkä
• Petri Blomqvist
• Ville Koskinen

2
Background

• Projects for Helsinki City Transport and in a
  DART-project
• More recently, a project for Ministry of
  Transportation and Communication
• To measure the planned quality of public transit
  system
• As a result, a quality measurement system has
  been developed

3
What is it and what it does?

• An EMME/2-model
• Basis on deterministic assignment
• Generates virtual passengers from each zone to
  each zone
• Analysis of two way trips possible

4
Example - Commuters

• Arrival to the job at 800-900 am, departure
  8.50.5 hours later
• Search for optimal paths for the virtual
  passengers by deterministic assignment
• The arrival time is varied 800, 801, ... , 900
• When a new one-way connection is found, search
  for an optimal path for return trip
• Different types of two-way trips can be analyzed

5
The quality indices of supply

• For each OD-pair the average of
• Total transit times and its components
• Transit speed in km/h or in relation to the auto
  speed
• Number of boardings
• Headway
• Number of transit connections between the OD-pairs

6
Indices of supply vs. demand

• For each O/D weighted average of the indices
• Weighting by total demand or transit demand
• Visualization of the OD-pairs that have low
  values of the index value but high demand
• ENIF for visualizations

7
Connections to a destination
8
Average speed by destinations
9
No connections (OD-pairs)
10
Lowest traveling speed
11
Averaging can be dangerous!

• Two example schedules
• The std or minimum of the index

Example 1 Example 1 Example 2 Example 2
0805 0805
0806 0820
0835 0835
0836 Ehdw15 min 0850 Ehdw15 min
Example 1 Example 2
std14 min max29 min std0 min max15 min
12
Tulosten visualisointi
13
Transit speeds (km/h)
14
Number of boardings
15
Walking times (min)
16
Uses of the measurement system

• Analysis of planned quality of transit system
• Monitoring it when the supply and the demand
  change
• ENIF can be configured as an interactive decision
  support system that is easy to use
• The demand matrix makes the measurement system
  very useful
• Cost/benefit analyses