HomeMy WebLinkAbout36. Passenger Tracking System for Oshkosh Transit #11C41illo'D'e
OfHKOJH
City of Oshkosh - Transportation Department ON THE WATER
926 Dempsey Trail, Oshkosh, WI 54902 (920) 232-5342 (920)232-5343 fax
MEMORANDUM
TO: Mark Rohloff, City Manager
FROM: Christopher Strong, P.E., Director of Transportation'tc
DATE: May 4, 2012
RE: PASSENGER TRACKING SYSTEM FOR OSHKOSH TRANSIT
The first goal listed among five that the Common Council assigned to you at its February 14,
2012 meeting is as follows:
"1. Develop a tracking system for the Oshkosh Transit System.
Plan the implementation of a system to track and provide data reporting and analysis of
all bus users by location. Include the implementation plan and required resources for the
tracking system in the 2013 budget. "
The purpose of this memo is to discuss and compare potential approaches to meet this goal.
Staff Understanding of Background for This Goal
During last summer's discussions regarding the Transit Development Plan, some Council
members expressed concern as to the plan's recommendations for new routes. In some cases, it
was felt that the data used to develop the plan were lacking or incorrect. During the budget
workshops, at least one Council member suggested that there could be benefits in having better
data on understanding the trips that riders are making on the transit system.
Staff Understanding of This Goal
Staff assumes the following definitions from the goal:
• "system"—A technology-based approach is envisioned, which requires negligible staff
involvement to sustain on an ongoing basis.
• "track"—This implies that the system is intended to collect regular, if not continuous, data,
rather than isolated "snapshots" in time.
• "all bus users"—This implies that the system must cover all bus routes under our service, and
all users of those buses, regardless of time of day. This may also imply that data are expected
to be collected for individual trips, correlating a passenger's specific origin with a specific
destination.
• "by location"—The system is concerned with where trips are coming from and going to, but
not as much with demographic or market information, such as passenger age or trip purpose.
City of Oshkosh—Department of Transportation
Passenger Tracking System for Oshkosh Transit/May 4,2012
Technology Approaches to Meeting This Goal
Staff is not aware of any transit system in the United States that has implemented a technology
solution that would attain all of the aspects indicated above. There are some proven technological
approaches that could address parts of this goal.
1. Full Electronic Fare Payment System
Since every passenger must provide a fare (or show the driver that there is no need for a fare) to
board a bus, then every passenger can be theoretically intercepted through a fully implemented
electronic fare payment system. Under conventional electronic fare payment systems, a passenger
uses a fare payment media, such as a stored value card, to pay for fares. This card may be read
using contactless technology (akin to some bar code readers), or a swipe-type reader(akin to a
credit card reader). The reader either deducts a fare or, in some rail systems, calculates a
distance-based fare according to where the rail trip started and ended.
Bus systems that have implemented electronic fare payment typically use the fare reader only
when someone boards the bus. To meet staffs understanding of the Council's goal, the fare
payment system would need to be able to track both when passengers board the bus and when
they leave. This means that passengers would need to pay or show a fare on both ends of the trip,
instead of only as they board (current practice).
This technology aligns with a recently submitted grant application that would secure electronic
fareboxes for our buses. The cost estimate for these fareboxes is about $15,000 per bus
($250,000 for the fleet); however, this excludes the costs of extra farecard readers that would be
needed at the rear vestibules (for passengers leaving the bus), and any additional costs for
integrating other fare media, such as the Titan ID Card, that might be necessary. It is unknown
whether additional investment would be required to support detailed trip-level analysis of fares
that have been paid.
This technology offers several benefits. Using electronic fare payment for passenger tracking
provides the best opportunity to track where each user boards and exits the bus. In some cases,
this system can generate some demographic/market segment data, as some fare payment
categories are associated with specific user groups (for example, UWO, elderly and disabled).
From an operational perspective, the technology has benefits with auditing revenue collection
and loss control. In addition, the technology could provide the ability to implement a distance-
based fare structure, which might be a benefit in the future.
This approach, however, has several drawbacks where it would conflict with current operational
practice in ways that could potentially have adverse effects on ridership.
• Many passengers, such as children under 5 or UWO affiliates, do not need to pay a fare when
they board, and therefore would likely not have a fare card.
• UWO affiliates currently use a Titan ID, which would not be compatible with this system
without additional efforts at integration.
• Cash or tokens would not be allowed in this system.
City of Oshkosh—Department of Transportation 2
Passenger Tracking System for Oshkosh Transit/May 4, 2012
• Most importantly, this technology would require passengers to pay/show a fare when they
leave the bus, which will increase inconvenience for customers, create congestion as people
are leaving the bus, and make it more difficult to keep buses on schedule.
2. Automatic Passenger Counting
Automatic passenger counting (APC) can be understood as a"break-the-beam"technology which
counts bodies or objects as they leave or enter the vehicle. APC readers can be installed at both
the front and rear doors of the bus and can track both boarding and alighting. The technology is
frequently used in heavy rail transportation, but also has been successfully used on urban buses.
Using APC avoids some of the problems that were involved with the electronic fare payment
system. People can pay their fares as they currently do, and the APC system acts "in the
background" to collect the data. As an independent system, the APC would not require the transit
system to change any existing policies with respect to fare collection or rider expectations. APC
can eventually be used to simplify some Federal reporting requirements; however, this can only
occur after a lengthy (-12 month) calibration period.
Implementing this would require retrofitting all of our buses to have APC sensors on both the
front and rear doors. The basic APC equipment is estimated to cost approximately $9,000 per
vehicle, or about $150,000 for the fleet. It is unknown whether this basic package has the GPS
capability that would be required to give location-specific information. Staff has not pursued
funding for APC previously, so this would require funding that would need to be pursued
through a combination of grants and future Capital Improvement Program requests.
This system would fall short of staff's understanding of the Council's goal, as it would not be
capable of tracking boardings and alightings for specific passengers. For example, it could
identify the number of people boarding at the downtown transit center, and the number of people
alighting at Washburn Street and 9th Avenue, but it could not identify how many people were
going from the downtown transit center to Washburn Street and 9th Avenue.
3. Automatic Vehicle Location w/Tablet Interface
Automatic vehicle location (AVL) using GPS is quite common in the transit industry, and is
currently in use on many City vehicles. By itself, this technology would merely provide a time-
stamped sequence of the locations of where a bus has been. Many vendors have developed
systems where the AVL is integrated with a fare collection system (as described under option 1).
One way to avoid the operational and customer service problems associated with option 1 is to
have the AVL integrated with a simpler interface. Currently, transit operators use a clicker to
count the number of fare payments by type as they make each circuit. Operators then manually
copy this information onto "trip sheets" when they return to the transit center at the end of a
circuit. The clicker is a manual device without any geographic reference. Replacing the clicker
with a touch-screen tablet interface would support the creation of a data set of passenger
boarding information, with time and location references, without any additional burden on
City of Oshkosh—Department of Transportation 3
Passenger Tracking System for Oshkosh Transit/May 4,2012
operators.
Commercial off-the-shelf AVL systems could cost $300,000 for the fleet, which would offer
additional functional capabilities that may be beyond what we need; however, these may not be
integrated with the tablet interface as described. The Council approved $100,000 toward AVL
implementation in the 2012 Capital Improvement Program. Based on conversations with a
couple of vendors, we believe that amount may be sufficient to develop a custom system which
might have fewer functional features than the more expensive AVL systems but would be better
suited to meet our system's operational and customer service needs.
An AVL-based platform offers several benefits. By generating real-time information on bus
location, customers can know how soon a bus will actually arrive at their location. Our customer
service staff can also reduce radio contact with operators, which reduces workload complexity
for the drivers. It also helps to evaluate on-time performance, address route timing issues, and
investigate customer complaints (such as speeding or route deviations). The data can also be
automatically fed into an enterprise database, which will eliminate the time required for manual
entry of trip sheets.
This technology would fall short of staff's understanding of the Council's goal. Like the APC,
the AVL is not set up to track individual passengers, and so it cannot generate a matrix of origin-
destination trips. In addition, this would not be able to track where passengers exit the bus
without significantly increasing the workload for operators, which will create issues similar to
those described under option 1.
Summary
Table 1 summarizes each of the technology options for the passenger tracking system. None of
the options meet all the criteria established by the Council.
Table 1: Summary of Technology Options for Passenger Tracking System
Option 1: Option 2: Option 3:
Electronic Automatic AVL/GPS
Fare Passenger with Tablet
System Criteria Payment Counting Interface
Automated Yes Yes Yes
Continuous Tracking Yes Yes Yes
Origin-Destination No No No
Number and Location of Boardings Yes Yes Yes
Number and Location of Alightings Optional Yes No
Boarding-Alighting Pair for
Possible No No
Individual User
Real-Time Optional Optional Yes
City of Oshkosh—Department of Transportation 4
Passenger Tracking System for Oshkosh Transit/May 4,2012
Origin-Destination
For transit planning purposes, it is often valuable to know the ultimate origin and destination of a
trip. Transit service that offers a shorter door-to-door travel time to the customer can be more
appealing and can, in some cases, attract new ridership.
None of these technology options provide a true origin-destination pair for each trip. In order to
do this, it would be necessary to know where a person started their trip before he or she arrived at
the bus, and where he or she went after alighting. To address this, some transit systems will
conduct occasional origin-destination surveys of their riders. These manual surveys can be
expensive to implement, especially depending on the level of statistical validity that is expected,
and can provide only a snapshot in time of origins and destinations.
Moreover, none of these technology options can answer questions of the potential ridership
effects caused by changes in service, such as higher frequency or closer service to a popular
destination. Potential ridership effects can be approximated through various efforts, including
travel models and targeted surveys. Getting this information requires detailed information
regarding the trips that people make, such as through trip diaries. Collecting and analyzing this
information at a statistically meaningful level would require an extraordinarily expensive survey
effort. Moreover, many trips simply cannot be accommodated effectively by transit in the near
future (e.g. second or third shift work trips, trips to jobs in other cities, trips with multiple stops,
trips involving large amounts of cargo, etc.).
Data gathered through these origin-destination approaches can supplement data collected through
one or more of the technology alternatives described earlier. However, they are not sufficient in
themselves to address staffs understanding of the Council's goal.
City of Oshkosh—Department of Transportation 5