Managing Systems Development Term Paper Leap integrated ticketing system
Post on 10-Dec-2014
365 Views
Preview:
DESCRIPTION
Transcript
Managing Systems Development Term Paper on LEAP Integrated
Ticketing System:
Course: iBusiness, Innovation Through ICT.
Module: Managing Systems Development (MIS 40670)
By
Darragh Leahy: Student Number: 07313586
David Malone: Student Number: 09188723
Ian Fleming: Student Number: 12250042
Word Count: 5,278
We (Darragh Leahy, David Malone and Ian Fleming) declare that the following document is
entirely our own work.
1
Table of Contents
I. Introduction
II. Requirements
Initial Problem Statement and Feature List
Desired User Goals
III. Research methods, data, evaluation.
Research Methods
Issues Discovered
IV. Solution: Proposed adapted System
Short-Term (zero-two years)
Medium-Term (two-five years)
Long-Term (five+ years)
Theoretical Future; End Game Goals:
V. Proposal to develop the system:
VI. Summary & Conclusions
VII. References
2
I. Introduction
The high-tech service which we examined was the LEAP Integrated Ticketing system which
public transportation in Dublin relies upon. The system has a complex and intertwined
relationship with its wider environment. LEAP began as a business requirement that came
from a government mandate, as such no general requirements analysis was performed.
There was a political imperative to bring in the system. In 2002, the Minister for Transport,
Mary O’ Rourke, decided she wanted to implement a smart ticketing system.
The Department of Public Enterprise awarded a procurement contract to the Railway
Procurement Agency (RPA) to implement a smart-card transport system in the Greater Dublin
Area. Several technology experts from the Department of Transport were involved in drafting
the mandate for a single smart-card scheme that would work on multiple forms of transport.
All public and transport operators were included in the process. When the requirements
were established the RPA issued contracts to build and operate the system on European
tenders websites. Unfortunately, there were no companies out there that could deliver the
requirements for the tender, so a review report was commissioned and the Government
decided to split the tender into two portions: one tender to build the system and the other
tender to operate the system. IBM won the contract to build the system and Hewlett-Packard
won the contract to operate the system and build the website for the system.
Leap was first launched in 2011 and there have been several versions of LEAP as
management at the NTA have attempted to deliver the majority of original requirements on
a phased basis.
As we couldn’t find figures publicly available, we assume that the original customer size and
market for LEAP was for all public transport users on the Island of Ireland (given the original
requirements for the system were for interoperability with Northern Ireland’s transport
system) however the goal of interoperability was never achieved. This target customer size
decreased over the course of the LEAP project, eventually targeting only users of public
transport in the GDA. The GDA encompasses the Dublin (represented by four local
3
authorities: Dublin city, Dun Laoghaire-Rathdown, Fingal and South Dublin) and Mid-East
regions (Kildare, Meath and Wicklow).
The most recent figure we could find in a 2011 report on the National Transport Agency’s
website was that 19% of the population of the GDA area used public transport. The
population of the GDA in 2006 was 1.6 million inhabitants, so 19% of this would equate to
304,000 users of public transport. In 2002, the population of the GDA was slightly less at
1.535 million inhabitants, so assuming rates of public transport use in the GDA were slightly
less four years previously (let’s say 18%); we believe the target market size for LEAP
services could be estimated at 276,300 people.
The system is currently used by users of buses in the GDA, including Dublin Bus and Bus
Eireann. LEAP is used by train users, DART and Iarnrod Eireann’s other services. It’s also
used by LUAS users and operates on all Veolia Transdev’s services.
II. Requirements
Initial Problem Statement and Feature List
Following the development of the governmental mandate, several technology experts from
the Department of Transport were involved in drafting the design for a single smart-card
scheme that would work on multiple forms of transport. All public and transport operators
were included in the process.
Pre-LEAP:
Users of public transport in the Republic of Ireland were:
· Unfairly forced to pay double/triple their fare prices for transport if they have to
change transport operators during their journey.
· Unfairly forced to pay for transport with no capping system in place for maximum
4
amounts spent per day.
· Subjected to inefficient waiting periods while scrambling for correct change for fare
payment.
· Delayed due to the time taken for other users to pay bus drivers.
· Unable to pay for transport in Northern Ireland without converting Euro into Pounds
Sterling (and vice versa) which has associated fees.
· Republic of Ireland needed an integrated public and private transport system.
When the procurement contract was given to the RPA instead of Dublin Bus, this created a
certain amount of animosity with Dublin Bus because they had been pushing for a scheme
like this for years and the RPA had no expertise in an area like this. The RPA acknowledged
their lack of experience so they hired an independent consultancy firm called Sequoia Smart
Solutions to come up with the original specifications/ requirements for the system in a
document (Mr. T Gaston 2013, pers.comm., 1 November).
Sequoia won the initial tender and a second tender also.
Original features of the LEAP system included:
● An e-purse with which users could make electronic payments for transport as well as
park-and-ride services.
● Multi-operating capping (e.g. you only spend a certain amount per week or per month
on transport before it’s capped).
● Rebate system if taking multiple journeys with different transport operators.
● Inter-operability with transport services in Northern Ireland.
● Ability to hold a variety of different ticket options for bus, rail and LUAS (e.g. monthly
tickets, rambler tickets etc. all on the one LEAP card).
● Would integrate with the Free Travel Scheme.
5
Current LEAP System:
● Has introduced the e-purse system.
● Still being charged double/triple fares for changing operators (although rebate
system to be introduced in March 2014 will change this).
● Still no capping system for maximum fares per day (will also be rolled-out in February
2014).
● No current system in place to accommodate tourists in the City. The price of buying a
LEAP card for €5 is quite expensive considering the tourist may only use the card for
a short period of time.
● Still no inter-operability with transport systems in Northern Ireland.
● Few private transport operators signed up to LEAP system – should include taxis and
a wide variety of private bus operators.
● Users of public transport outside the GDA have no access to the LEAP system and its
associated benefits.
● Users can’t top-up LEAP cards on buses which have no live system capabilities.
● Users still have to stop and converse with bus drivers regarding fares due to
stage-fare system creating delays and also confusion.
● Significant delays of 48 hours after topping-up online before LEAP card is credited.
● Often unaware that you can top-up online.
● Need to visit a LEAP retailer outlet if choosing to top-up online.
● Completely unable to avail of bus transport if they find themselves with no LEAP
credit and have no cash on them.
Not all of these features have been delivered, but the main reasons (according to our
survey) that the LEAP system is beneficial is for convenience reasons (efficiency – saves
time, ease of use, don’t need to carry change or cash) and cost-saving reasons i.e. the
system saves you money, it also stops overspending on transport across multiple providers.
An ongoing failing of the system is the time taken to deliver some of the promised features
(with inter-operability with Northern Ireland never being delivered). This can be seen as
being ‘the nature of the beast’ considering a phased approach to delivery and
6
implementation of certain features was adopted.
The e-purse feature of the project was first released in 2011, with additional features still
being released. There have been several versions of the project with additional features
being rolled out on a phased basis. After the e-purse they delivered a version that allowed
LEAP users to hold and top-up different flavours of tickets on their LEAP card, allowing for
the Rambler style tickets many users are familiar with. In March 2014, they are projected to
roll-out the rebate feature (allowing hop-on/hop-off cost accounted usage) for LEAP as well
as capping of fares at a maximum limit.
Desired User Goals
The desired user goals are:
1. Transport users can use multiple forms of public transport on a single journey without
having to pay the full fare traditionally associated with changing to a new transport provider
mid-journey.
2. Pay for transport without having to carry cash/exact change.
3. Pay a maximum amount for transport in a single day.
4. Save money and time by using a convenient smart ticketing card.
5. Make the user’s life simpler through having various monthly ticketing offers on one card.
III. Research methods, data, evaluation.
Research Methods
We integrated IDEO methods into all out research, using TRY, LOOK, LISTEN & LEARN IDEO
techniques.
7
Actual research took several forms:
● Field Research: Personal use of LEAP from card acquisition to travel.
● Field Research: Observation of LEAP use with general public.
● Field Research: Interviews with the general public observed using the system.
● Phone Interviews with experts: Tim Gaston (Director of Integrated Ticketing Scheme,
National Transport Agency) and Brendan Flynn (Technology Development Manager,
Dublin Bus).
● Quantitative Online Survey.
Our field research attempted to discern the ease of obtaining both a Leap card and Leap
credit. We wanted to test this multiple times to ensure we had covered a wide enough
survey base so as to provide a definitive answer. After visiting several shops around Dublin,
it quickly became apparent that the Leap distribution network is good and it proved very
easy to obtain both the card and credit.
We tested the actual operation of the system, to verify what ‘mistakes’ it took to force an
error. In general we found that the system worked well, not allowing us to tag-on twice in a
row and cause a duplicate charge. However, we did note that if we waited a few minutes, it
did allow us to tag-off at the same stop and charged us for the trip, even though we had not
changed stop. It should be noted that this was attempted on the Luas systems; when
attempted on the Dart, the Dart recognised the user had not travelled any distance and
refunded the full amount. During this time, we took the opportunity to conduct short
interviews with commuters regarding their use and perceptions of the Leap system as well
as shadowing their use of the Leap card machines. This turned up interesting issues
regarding the scan speed in a busy queue.
We finally investigated the top-up systems available, noting that both the Luas and Dart
provided transport side top-up facilities. Dublin Bus was a notable exception, as none of
their buses run a ‘live’ system. Rather, all transactions are recorded off-line and updated at
the end of the day.
Further, we found the online top-up system was subject to a prolonged delay and required a
visit to a local top-up point anyway to redeem the top-up. This is a redundant process.
8
Issues Discovered
Several issues became apparent when studying the LEAP system. The actual use of LEAP
seems to be relatively poor (not necessarily the number of Leap cards in circulation, but
based on our data and what we observed in use). In relation to costs, the public belief
(whether correctly or incorrectly) is that Leap does not save the customer money as well as
some frustration regarding the functionality of the tag-on/tag-off speed, especially during
busy periods. These problems have negatively affected public opinion of the system.
From this we can draw that marketing and brand image is an unaddressed issue for LEAP.
We believe this failing may be partly due to the government mandate rather than business
case that initiated the project. The problem with marketing is that it is expensive but to
alleviate this expense, social media exposure and usage could be enhanced. Facebook,
Twitter, Pinterest and Google plus are used widely by the public and are a valuable channel
to spread the word about LEAP.
Our quantitative survey produced some interesting results. 97 respondents filled in our
online survey, with 33 of these stating they do not use LEAP. The four main reasons for
non-uptake were, in order of importance: don’t use public transport or sparingly use it, lack
of awareness of the system, cost reasons (no perceived benefit) and setup issues.
Of those that did use LEAP, the main advantages in order of importance were: it’s
convenient, cashless, cheaper and user-friendly. Many disadvantages were noted but the
main ones were, lack of flexibility, tag-on/off issues, lack of top-up options, website issues
and poor balance information on readers. 83% of respondents were under 50 years of age.
55% of respondents were female and 45% male, giving a good base of data. Only 13% of
respondents stated they were happy with LEAP website and that it’s user-friendly. This
feature is something that could be addressed with minimal time and expense. Only 13% of
respondents stated they use the website to buy credit, this should be a focal point of activity
and this is letting the current system down.
One critical question we asked was: did users think LEAP was working? 51% of the 67
respondents stated YES, with NO at nearly 10% with the remainder stating Maybe. This was
a surprising finding given the problems that users mentioned in other questions asked, and
9
posed the question of whether these answers were truthful! We have to take them at face
value, but there’s no doubt the perception of users isn't what it should be, with only a little
over half thinking the system is working at present.
Further, from our discussions with Tim Gaston (Director of Integrated Ticketing, NTA) and
Brendan Flynn (Technology Development Manager, Dublin Bus), it became apparent that
stakeholder support from the various transport agencies was an issue until the NTA was
finally established with the power to regulate the agencies and ensure co-operation.
Based on our findings, even following the implementation of Leap, users of public transport
in the GDA are still affected by:
● No capping system for maximum fares per day.
● Being charged double/triple fares for changing operators.
● Few private transport operators signed up to LEAP system i.e. taxis.
● No nationwide deployment of Leap.
● Users can’t top up their LEAP cards on buses which have no live system capabilities.
● Issues with Dublin bus fare system.
● Delays of 48 hours after topping up online.
● Unable to avail of bus transport if they have no LEAP credit and have no coins on
them.
Furthermore, there are a wide variety of different pricing structures in place with the LEAP
system across bus, rail and LUAS. We found the pricing structure for Dublin Bus, to be
complex and at times confusing for users. There is a summary of the different pricing
structures for the various LEAP transport operators in appendix four.
It’s also worth noting that the initial ballpark budget was €30 million in Capex for the LEAP
project. This figure has since increased to €55 million, of which nearly all is spent. Brendan
Flynn explained to us that a cost-benefit analysis was conducted and the cost of €30 million
appeared to justify the LEAP system. Tim Galston from the National Transport Agency
mentioned getting the IBM part of this project completed on budget was a notable, if one of
the only budgeting successes and that this was almost unique in the public sector. €12-13
10
million was the original estimate for the IBM work, and the final price after about 80
variations of the software amounted to €13.5m. While this does not constitute an issue with
the system operating as such, it certainly indicates the complexity as well as the failings in
management aspects of the project. Complexity is fine, as long as there is a very simple
high-level design, complexity should be a coding problem if a problem at all, not a
management issue. Both this complexity and the financial management of the project
(either in accurate forecasting, or improved delivery costs) will need to be addressed if
future objectives are to be met. This is made all the more pertinent as all costs, and cost
overruns are funded by the tax paying public through the government.
IV. Solution: Proposed adapted System
In developing a solution, we are acutely aware of a few stumbling blocks to any
redevelopment of the system. Primarily, the NTA is effectively broke. They simply cannot
afford the Capex to redesign some of the features and technology in the system.
Furthermore, as previously discussed in the ‘Issues’ section, the technical aspects of the
system and its development can be accurately described as, what Foote and Yoder (2000)
termed, a ‘Big-Ball-of-Mud’. This is most likely due to the compromises made in supporting
legacy hardware and the work required to integrate such different systems. Compromises for
legacy system operation also introduce an additional layer of technical complexity to ensure
backwards compatibility with the existing system. This further adds to a technical
redevelopment cost.
As the project was initiated before the formation of the NTA, there’s something of a
governance, or at least product owner, “mud-ball” in play as well. This can make initiating
any change to the system a significant challenge in itself.
We would recommend a review and centralisation of responsibility and a democratisation of
domain knowledge regarding the operation of the LEAP system at a technical and
management level.
In terms of a technical re-design, although as mentioned we are quite aware that due to
11
budgetary and technological constraints any proposals are unlikely to be considered, we
have devised the following redesign to the LEAP system, broken into short, medium and long
term phases.
Following is the breakdown of each phase and overview diagrams of how these may look:
Short-Term (Zero-Two years).
Initial improvements should take the form of implementing the initial project requirements in
the GDA as dictated by the mandate, particularly the fare-capping requirement. Further
steps should be taken to extend the LEAP system to more forms of public transport such as
the taxi services. This could potentially be done in co-operation with an existing transport
system like Hailo. https://hailocab.com/ireland
Medium-Term (Two-Five years).
In the medium-term only two direct solutions are to be proposed, this is to allow the LEAP
card to be linked to a credit/debit card via the website to allow automation of the top-up and
payment system, opening up LEAP to businesses as well as looking to use forms of public
transport (especially taxis where companies insist on credit lines). This would allow for a
single, universal transport card. Additionally, to allow LEAP terminals to take payment from
NFC, contactless bank cards.
The work involved in these solutions, however, is both significant in terms of scope and
expense and requires full support from the highest levels of management within the NTA to
succeed.
Long-Term (Five+ years).
Two revisions are proposed:
The deployment of the LEAP system nationwide (to include Northern Ireland) and by
extension to England through the Oyster system integration required in Northern Ireland.
The inclusion of mobile NFC payment communication standards (which have yet to be
12
officially IEEE or ISO standardised, budgeting here that a five year timeline will suffice for
maturity of such systems and standards) to allow payments with any device and turning the
LEAP system from a physical card into a truly digital card. This would, in turn, make LEAP a
very easy system to use for tourists as well as systems like Google Wallet, Amazon Pay or
other such payment systems could be easily interfaced with, to automatically convert from
their national currency to the Euro and deduct the correct fare as well as presenting
information regarding the fare in the detected language of the wallet, or currency of origin.
Theoretical Future; End Game Goals:
China, specifically Hong Kong, has an extremely well-developed and mature transport card
system called the Octopus Card http://www.octopus.com.hk/home/en/index.html which
offers not only integrated transport payment, but also NFC payment abilities for small
purchases in shops.
Ultimately, the system should view itself as an evolving service, with a view to becoming,
almost, a national petty-cash system. We also feel that the website for LEAP should be
redesigned and a mobile App created to encourage wider use of LEAP among businesses
and to allow consumers to check their balances easily.
13
Figure 1 (below): Short-Term: Conceptual Map for Proposed LEAP Improvements:
Figure 2 (below): Medium-Term: Conceptual Map for Proposed LEAP Improvements:
14
Figure 3 (below): Long-Term: Conceptual Map for Proposed LEAP Improvements:
Figure 4 (below): Proposed redesign of www.leapcard.ie website:
15
Figure 5 (below): Proposed design for LEAP mobile app:
V. Proposal to develop the system:
To call LEAP an integrated ticketing system is a contradiction in terms. Our survey revealed
many issues customers have and it would be foolish not to address these issues in the
short-term before jumping full steam ahead with a system redevelopment. Technology and
integration problems accounted for 69% of the issues that customers complained about. Our
proposed new system aims to address these obviously high-profile issues only after certain
system wide reviews have taken place. The ‘big-ball-of-mud’ issues outline the need for
proper planning and dialogue between all stakeholders.
The Irish market for public transport users is unique in that it is very small when compared
to other countries like China (Octopus) and England (Oyster) where successful versions of
LEAP have been introduced. With a small population, cost is important as it becomes hard to
16
firstly breakeven and secondly make profit from any transport system. Economies of scale
don’t work here! Major cultural issues also abound as the Irish people are notoriously slow
to embrace new technologies.
Our plan is to introduce new features and adaptations over time, what we like to call our
‘three-tier’ approach. Short-term (being defined as between zero to two years in length) will
see the system gradually upgraded to enhance features and become fully integrated. The
current single fare system is a priority obstacle to uptake as it erodes perceived benefits of
the system. A high majority of respondents to our survey stated that they use two or more
modes of transport in Dublin; this would signal to us that some commuters are using two
modes of transport on a daily basis. Due to the failure in the current system, the supposed
benefits are lost. For a system that promotes flexibility and cash-saving it is shocking to
think that over 30% of respondents in our survey stated that lack of cost benefits and
marketing were factors that are holding LEAP back from increasing its user base.
Concerning Dublin Bus, asking the driver to validate tickets at the end of your journey is
frustrating and time-consuming, especially so during busy commuting periods where
demand is high. An intelligent tag-on system needs to be incorporated on this mode of
transport soon.
Stage two will be the medium-term plan between years two to five. This will see debit/credit
and retailer loyalty cards integrated into the system. A deal could be struck with major
retailers like Tesco, Mothercare and Odeon cinemas, to offer reward points that could
incorporate savings for the customer by having shopping cards linked to LEAP usage. For
example, offering discounts on certain brands in stores or free tickets for cinemas when a
certain amount is spent could generate more business for retailers in a competitive market.
This may also provide valuable and surprising purchaser information for use in data
analytics as a benefit to the subscribed retailers.
When implementing debit/credit card integration, an initial cost will be inevitable but having
the flexibility to upload money onto your leap card instantly via credit/debit card when your
balance is low or empty is not just convenient but a revenue generator for LEAP as it should
17
increase user numbers.
Stage three, will be titled ‘Full Integration’. This is the stage when the long-term goals of
LEAP come to fruition. It is sensible that once the numbers are such that the costs are
justifiable, that an App allowing NFC payments for LEAP is implemented. While obstacles
need to be overcome with iPhone manufacturers Apple, Samsung and HTC; an App could
save and generate income for all parties concerned as well as further increasing the
potential for revenue generating analytical data. The majority of the public have
smartphones, whether Android or iPhone so with all the major banks having developed Apps
linking the technology would make sense. Passbook and Google digital wallets could use
codes on your App to tag-on and off. Taxis could be brought on board as well, once it’s fully
automated and payments can be made instantly into taxi-operator accounts. Taxis would
need to be refitted with LEAP card scanners or be willing to take debit/credit card with LEAP
functionality built-in. This in theory should not be an issue as partnering with existing
provider Hailo may dictate successful expansion. Taxis being incorporated into the Octopus
system in Hong Kong proved unsuccessful as issues with privacy and a lack of automation
meant the system was unsustainable in the short-term. Utilising an existing and successful
system such as Hailo (https://hailocab.com/ireland) could help circumvent this issue.
For stages one, two and three to be successful a review of the current systems development
life cycle has to be initiated.
Planning – A plan with clearly defined project areas needs to put in place. All stakeholders
including relevant technology bodies and transport agencies need to be conscious of the
need for a coherent and viable plan for LEAP’s future. Provider terms of service, clearly
defined hardware compatibility guides and pricing strategy are a must-have. Our team, if
given a consultancy job of this nature, would firstly work to establish clear and effective
communication links in order to facilitate the planning process. A full internal and external
review will be performed of the system with market research and testing of the functionality
of the current system a priority. A CRM team needs to be put in place to assess customers
issues with current functionality and a forum or blog could possibly be set up to keep
customers up to date with future events regarding LEAP. A roadmap should be uploaded on
LEAP’s website for the public to view and suggest possible alterations. This should be done
18
over a six month timeframe to allow genuine feedback. Social media forums namely Twitter
and Facebook could be used to promote the system. The demographic of users of social
media is primarily 18–40 years of age but it is increasing. It’s an affordable channel to
advertise when budgets are limited.
Full testing on new upgrades and functions needs to be part of the planning phase and we
propose that for stage two and three, trial runs or limited introduction of debit/credit and
shopping card technologies be introduced. If it becomes apparent further revisions are
required, these will be factored into a RUP/Scrum based development cycle to ensure
prompt addressment of any particularly important issues.
Analysis – A complete domain analysis needs to be undertaken. Systems to measure against
would be Oyster in the U.K. and Octopus in Hong Kong. Ireland is late to the party when it
comes to integrated transport ticketing, however this should mean it benefits from the
knowledge of the mature systems rather than repeats their mistakes. Octopus is the
pace-setter and many of its achievements and mistakes can be learned from. In depth
research should be carried out on the perceived problems of the current system. Preliminary
research has been undertaken by our group and the feedback has already suggested many
possible upgrades to the system (Please see the appendix for survey analysis).
Design – The functionality and specification of the new system needs to be clearly laid out.
During stage one, when minor tweaks and upgrades are being made, a concurrent project
team should work closely with Leap staff and transport companies, with their goal being the
design of stage two and three. The problems regarding design failures from Oyster and
Octopus should be analysed to address possible duplication issues. The question is: in stage
one, how do we make the current system better integrated? The Dublin Bus scanning system
needs to be made compatible with the LEAP card to avoid unnecessary queues during peak
times. This major drawback for bus users is harming the future customer base.
Implementation – Stage one would go live following the successful completion of the User
Acceptance Testing phase. This expansion of functionality and flexibility for LEAP customers
should improve market penetration. Clever marketing may also improve uptake of the
19
system, for instance playing on the ‘green’ factor of the system as no receipts are issued, no
paper needed.
During the second phase the implementation of linked bank/credit cards is the primary
objective of the system. As mentioned previously the cost and financing of the deals
between LEAP and the banks and retailers involved may be complex but with greater
numbers come greater revenue. Furthermore, the cost may be reduced if consumer data is
collected correctly and supplied to the banks in exchange for a cost offset. That said, a
review should take place of what this data consists of and how data protection applies in
this situation.
Finally, the implementation of stage three with apps and digital wallet technology is brought
on board. NFC and mobile payment (m-commerce) technology is still in its infancy globally,
lacking standards and widespread adoption. Due to this, rather than design an intrinsically
flawed system now, we intend to leave the detail design and implementation strategy of this
phase until standards have been formalised and the technology has matured sufficiently to
design a safe and effective system.
Maintenance – Hardware issues related to the tag machines in train stations/Luas stops.
Without accurate maintenance plans to review, it is hard to judge. Ideally this could be
contracted out if deemed cost-effective, unifying all maintenance under one outsourced
company. This could have added cost benefits in the removal of parts warehousing costs. As
errors and fixes would be handled on an ad-hoc basis as they arise, this seems the best
route to take. Software system issues would remain with IBM/HP on a maintenance contract
basis in order to maximise the available expertise available to the system. This is
particularly important given the ‘always available’ nature of the system. It would be
extremely cost prohibitive to maintain the array of experts required to manage and upgrade
such a large, technologically disparate system inhouse.
20
VI. Summary & Conclusions
The Leap system appears best described as a ‘big-ball-of-mud’ from both the technical and
responsibility point of view. Having said this, the system does undoubtedly work! The main
issues presented are re-designs and developments of the system that are subject to
unneeded added complexity and cost.
The research found there are various double standards in place across the system, with the
Luas, Dart and Bus systems all functioning slightly differently and all systems except the
Dart seeming to having an intrinsic and relatively simple issue in the implementation (i.e.
Luas change - on no distance travel).
As a result of these two issues, our suggestions at a high level are to decompose the system
into a simple concept. From there the complexity can be evolved down through the layers
where needed, but only as appropriate. This would be in line with a RUP style approach to
system documentation. This should improve visibility of the system and hopefully prompt a
management review to clear some of the complexity from the systems. While we understand
the complexity was a natural by-product of many of the compromises made in the
implementation, in order to move ahead successfully in the future, ease of communication
and simplicity of design are paramount. Any costs absorbed now through such a review
would be offset by a reduction in cost of both new features and maintenance over the
lifetime of the Leap system.
As the system is constrained by budget, we found it difficult to propose any major
redevelopments that could be viably considered as upon conversation with our sources, it
was indicated the development costs would be astronomical. Hopefully over the coming
years, and assuming the review we have previously suggested takes place, some of our
suggestions will become viable additions to the Leap system.
21
VII. References
Foote, B. and Yoder, J. (2000) Big Ball of Mud. Pattern Languages of Program Design 4. (N.
Harrison, B. Foote and H. Rohnert, eds), Addisson-Wesley. Boston MA.
Boards.ie Leap Card Summary of Products Available and Functionality. 2013. Boards.ie Leap
Card Summary of Products Available and Functionality. [ONLINE] Available at:
https://www.boards.ie/vbulletin/showthread.php?t=2057066922
[Accessed 25 October 2013].
Dublinbus.ie Fare Information. 2013. Dublinbus.ie Fare Information. [ONLINE] Available at:
http://www.dublinbus.ie/Fares--Tickets/Fare-Information/Fares/
[Accessed 25 October 2013].
Leapcard.ie Homepage. 2013. Leapcard.ie Homepage. [ONLINE] Available at:
https://www.leapcard.ie/en/Home.aspx
[Accessed 15 November 2013].
Mezghani, M. (2008) Study on Electronic Ticketing in Public Transport: European
Metropolitan Transport Authorities. 2013.. [ONLINE] Available at:
http://www.emta.com/IMG/pdf/EMTA-Ticketing.pdf
[Accessed 21 November 2013].
22
National Transport Agency (2011) Greater Dublin Area Draft Transport Strategy: 2011-2030,
2030 Vision. 2013. . [ONLINE] Available at:
http://www.nationaltransport.ie/downloads/GDA_Draft_Transport_Strategy_2011-2030.pdf.
[Accessed 21 November 2013].
Transport For London Homepage. 2013. Transport For London Homepage. [ONLINE] Available
at:
http://www.tfl.gov.uk/corporate/projectsandschemes/19976.aspx
[Accessed 10 November 2013].
Octopus.com.hk Homepage. 2013. Octopus.com.hk Homepage. [ONLINE] Available at:
http://www.octopus.com.hk/home/en/index.html
[Accessed 15 November 2013].
23
VIII: APPENDICES:
APPENDIX 1: TELEPHONE INTERVIEW WITH TIM GASTON
Interview with Tim Gaston, National Transport Agency Director of Integrated
Ticketing Scheme. (01/11/2013). 1pm.
1. How were the requirements gathered (were users of public transport consulted
with surveys, etc.)?
Tim said there was no easy answer to this. They didn’t just gather customer/ user
expectations with surveys. The history of smart ticketing has been around for quite a few
years and the government was aware of these initiatives and wanted a similar system for
Dublin.
Tim said that LEAP was a business requirement that came from a government mandate.
There was a political imperative to bring in the LEAP system.
Tim stated that looking back it may not have been the right thing to do.
In some sense, the project could be compared to a benchmarking initiative with other
countries.
There had been a smart card committee at Dublin Bus since 1990s.
In 2002, the then Minister for Transport Mary O’ Rourke decide she wanted to bring in a
smart ticketing system. They developed the mandate. A couple of technology experts from
the Department of Transport were involved in drafting the mandate for a single smart card
scheme that would work on multiple forms of transport. All public and transport operators
were included in the process.
One single scheme would work at a national level.
An original requirement was interoperability with NI was built in.
Another requirement in the mandate was that it would integrate with the free travel scheme.
The LEAP system would work with the original legacy ticketing schemes in place in the state.
The contract was given to the Railway Procurement Agency instead of Dublin Bus – this was
a business decision that was made and it created quite an amount of animosity because
with Dublin Bus because they had been pushing for a scheme like this for years and the had
24
no expertise in an area like this. So the RPA recruited some international experts in the field.
The RPA brought in a project manager and a technical manager. Then the RPA advertised for
contractors to build the system. There were three separate contractors hired. The RPA used
public procurements for the LEAP system.
There was a project team established.
They hired an independent consultancy firm called Sequoia Smart Solutions to come up with
the original specifications/ requirements for the system in a document.
http://www.sequoiasmartsolutions.com/
Sequoia won the initial tender and then they won a second tender also.
2. Did the various transport authorities implement using a unified (or even the
same) design project and how was communication handled between the various
transport agencies?
There was an external body developed. There was requirement capture/ specification – they
had their own personal expertise, but they also ran several workshops with the major
transport providers to see what the requirements should be. They met with the Department
of Finance and also conducted market research.
Tim joined the team in 2003. There were the contractors and some of their own Irish
programmers involved. They then tendered for suppliers – the suppliers would bring their
own programmers. The original contract was to build and operate the LEAP system for 5
years.
Tenders were put on Official European tenders website.
The tenders went to market 2004 and tender responses came back in early 2005.
They then came up against a huge challenge – none of businesses that had gone for the
tender met the requirements they needed.
The government then did a review report – the experts said that their design of requirements
was spot on – it was the governance element that was wrong. This came back to the
decision to give the project to the RPA – because the RPA was not a regulator and was not a
non-statutory body they had no authority to tell Dublin Bus or any other firm to introduce
25
changes to fit the requirements for LEAP.
There was a creative compromise needed between the different transport companies. Some
of requirements they had were prescriptive and needed to be refined.
Minister for Transport Martin Cullen (2004-2007) created a non-statutory body, called the
Integrated Ticketing Board – they put the CEOs of the RPA, Dublin Bus and Irish Rail on this
board as well as the CEOs of private transport companies and also a Civil Servant and a
Former Secretary General as well. Tim now reported to this Board and it made life a lot
easier for him. Minister Cullen made it very clear that had to stop fighting with each other,
start working together and work it out. Tim was now making recommendations to the Board
and they were carrying these out.
If the National Transport Agency had existed back then they would not have had these
governance issues and everything would have been much simpler.
Minister Noel Dempsey eventually founded the NTA as the regulator for transport in Ireland.
Good governance was essential for success.
Unless you have other stakeholders who can influence the process and have a certain
ownership of the project then nothing will get done.
The Integrated Ticketing Board took the design of the system and went back to the market.
They separated the project into two tenders – one for the building of the system and the
other for the operation of the system.
IBM built the integrated ticketing system, they spent €13 million building the system’s back
office. HP won the tender to operate the system, build the website, etc.
They defined the system as everything involved in the system, from the communications
between devices, the back office, the call centre, the website, the cards, the retail element
of the selling of cards.
HP was given responsibility of providing parts that weren’t there as well as the parts that
were.
HP was responsible for running the back office, dealing with an exceptional circumstances,
downtime, etc.
HP responsible for the IT management of back office, making sure servers were working,
proper firewalls were working, database management.
26
A lot of the work was done offshore, in MANILA in the Phillipines.
HP built the website www.leapcard.ie, set up call centre, they needed to purchase leap
cards, and finally set up the top up network – how this would be done.
A problem was that they had to integrate with the old Dublin Bus ticketing system. The
problem was that DB system couldn’t top up people’s LEAP cards when they got on the bus.
They needed electronic retail outlets – HP brought in Payzone to handle this.
Payzone manage the devices in the shops around Ireland.
The store owners get between 2 and a half to 3% of the cost of all LEAP credit sold, and
Payzone get a small percentage of every purchase. Even though the store makes next to
nothing on each sale, the incentive is that they get people coming in to their stores to top up
their cards and these people probably purchase other items while they are there.
Tim explained that IBM probably used a waterfall method when they were programming the
system – they provided hardware and firewalls.
http://www-01.ibm.com/software/success/cssdb.nsf/CS/JSTS-7VKL75
http://www-05.ibm.com/tr/smartercities/exhibit/documents/pdf/5_Government_Singapore_
Land_Authority.pdf
IBM subcontracted a lot of work out to NSI in Singapore (couldn’t find this company on the
web) – they already had developed a smart card system for public transportation in
Singapore, and they just modified the software for LEAP, therefore using Waterfall methods
because NSI was an organisation that had widely practiced the waterfall method of
programming in their organisation. This was a very difficult period for stakeholders involved
with LEAP because for 9 months they didn’t see much in terms of deliverables and then
there were problems when they finally came back to them with the software.
Tim mentioned getting this project completed by IBM on budget as being a success and that
this was almost unique in the public sector because €12-13 million was the original
estimate, and the final price after about 80 variations of the software, the software was
delivered for €13.5m. So it was a huge success coming in on budget and on time.
Tim mentioned that IBM had negotiated a very, very hard deal. They wanted all of LEAP’s
requirements to be vastly simplified and Tim had to fight hard to keep the requirements as
they were. IBM were coming across that it would be fine and why were the Integrated
27
Ticketing Board being so unreasonable.
3. Did the requirements for the project evolve over its course or did the
requirements stay fixed from the start of the project?
They broadly stayed the same as at the start. They still haven’t delivered the interoperability
requirement with Northern Ireland but they couldn’t achieve this due to security
requirements that they couldn’t guarantee the authorities in N.I. would comply with. It would
require changes to the infrastructure in N.I. Tim hinted that this was around the area of
cards being tampered with.
Capping is being rolled out in Dublin Bus in December. You’ll be only able to pay a certain
amount for a fare. There will be capping across all different transport sectors.
4. What does the future hold for LEAP? Or what would you like to do with the
system in the future?
Tim said that one thing they were working on at the moment was looking to London and
their Oyster card scheme. Dublin has caught up to where London was about three years ago.
In the last 2 years Transport for London have been working on a new type of ticketing
system where people can use their bank cards to pay for transport: you hold your card to the
reader and it scans it using NFC – Near field communications technology. It’s a flat fare in
London. What they are developing and about to launch, you can use in it underground and
on overground trains. When you use your bank card you are good to travel for the rest of the
day. If you are doing a lot of travelling, the card is scanned and you are automatically
charged with the best value ticket choice for the customer. It’s very targeted and they have a
very powerful back office running these complicated algorithms to make it all work. They
wouldn’t tell Tim the exact figure but they have spent 10s of millions of euros developing
this. You don’t have to pre-register your bank card either, tourists can use it straight away.
If he could do one thing in the next 5 years it would be implementing a system like this in
Dublin. Even if a system like this comes in in Dublin, LEAP will still have its role to play; they
won’t replace it, as LEAP is still integrated with the public travel scheme. LEAP will still have
its use.
28
When I asked him about the idea of integrating a digital wallet or phone scanning with LEAP,
Tim explained that there are complications with using phones. One of the problems comes
down to different standards between telecommunications companies and transport
companies which need to be aligned as there is currently a gap. Android users need to group
together to let their phones be integrated with transport company readers. Tim explained
that if there is a reading time of any more than than half a second for readers it creates
huge delays and problems. As things stand, the mobile telecommunications industry is
creating one set of standards and the transport companies have another set of standards
and there is a stand-off going on. The transport companies are saying “our entire
infrastructure is already in place why can’t you just tweak your phones slightly?” There is
also the added complexity of the huge number of different providers involved. Security needs
to be managed, there are the owners of the payment gateways (VISA /Mastercard ,etc).
Also Apple have decided not to make i-phones NFC compatible which means half of all
phone users couldn’t use the system if it was in place. The jury is still out. Other types of
ticketing such as the bank card may be more achievable as the phone as payment system
will not be viable for the next 3-4 years.
In France, there is a government body creating necessary standards to span the gap that
currently exists. The French market is heavily into NFC approach as they see the benefits
that accrue.
29
APPENDIX 2:TELEPHONE INTERVIEW WITH BRENDAN FLYNN
(Technology Development Manager, Dublin Bus).
25th October 2013.
1. The initial idea for LEAP and how they justified it.
Answer 1: The initial germ of the LEAP system goes back to about the year 2000. It started
with a government mandate from the Department of Public Enterprise to the Railway
Procurement Agency (RPA) to implement a smart-card transport system in the greater Dublin
area. Originally the scope of the scheme was much greater including financial and retail
services on the e-purse card such as paying for park-and-ride services, etc. The original
mandate included an idea for a rebate system (i.e. hop on a different bus you get some
money back). The original mandate for the scheme included the idea of nationwide use even
extending to cross-border use with Northern Ireland (idea that it would be integrated with
what would later become the Oyster system). The initial idea for the scheme was a
‘no-brainer’ because the Department of Public Enterprise and the different transport
companies realised that there were currently barriers to getting transportation for the public
and wanted to overcome these.
Dublin Bus and the rail companies already had monthly transport tickets but they wanted
these tickets to all be integrated on the one card. The realised that if you’re paying cash for
two buses that was two cash fares and that this was unfair to the public if they had to pay a
transfer penalty. So they started with the idea of a stored value e-purse and decided to try
and make it as fair as possible. The design of the LEAP system was undertaken by a
committee system with all the relevant partners (RPA, Department of Public Enterprise,
transport companies).
Dublin bus had had their own magnetic ticketing system for over 10 years at this time and
needed to see how they were going to upgrade this system. Dublin Bus also had a paper
ticketing system for cash and basically had the attitude of ‘Look, if you can give us the
requirements we will go about making this happen. LUAS were just getting up and running
and were saying that from the first day we will have smart card reader technology in place.
LUAS ended up starting off with their own LUAS smartcard as LEAP was not ready at the
30
time. Dublin bus brought in their own smartcards and smartcard readers in 2005.
Coming up with the specification for LEAP was quite a ‘torturous process’ as LEAP has a
‘torturous system’. Brendan described it as having ‘a big beast of a clearing system’.
In 2009 to 2010 the LEAP system was coming closer to fruition but there was a political
imperative to complete the project. They were forced to introduce the e-purse in 2011 but it
didn’t have all the features ready such as multi-operating capping (e.g. you can only spend a
certain amount per week or per month on transport before it’s capped).
Brendan explained that the rebate hasn’t been introduced yet but they will be launching it in
February or March 2014. E.g. you’ll get on a second bus and it will only charge you €1.50,
will refund you some money. This had been a major selling point of LEAP from the beginning
but they will only be able to roll-out the feature now.
What complicates matters even further and adds to the complexity is that all the different
transport companies are running their own in-house systems. The reason that they treated
the LEAP project in a phased approach was that it was more manageable this way. The
different transport companies directors get together and say ‘ok which feature can we all
realistically get implemented now’.
€30 million was the initial capital cost estimate for the LEAP project and now the cost has
increased to €55 million, of which this is nearly all spent. Brendan explained that a
Cost-Benefit analysis was conducted and the cost of €30 million appeared to justify the
LEAP system.
1. A statement of the requirements the LEAP system delivers?
Answer 2: Brendan believes the current system has delivered on the original requirements of
the stored value e-purse and that the stored rebate requirement (i.e. that they system can
deliver this, just has to be rolled out). If achieved the requirement of public transport users
to not have to carry change all the time. You can get all your different monthly tickets onto a
LEAP card now so this has benefits when you put all of these factors together. When I spoke
to him about a ‘flat-fare’ system and how this would not workable for Dublin Bus’s business
model currently due to the economics of it, Brendan agreed and made the point that bigger
cities like New York and London can introduce flat fares but with Dublin Bus if they were told
they had to introduce flat fares while receiving the same subsidies this simply would not be
possible at the moment and would affect the service. Brendan also made the point that it is
31
different now that there is a proper regulator in place like the National Transport Agency, a
serious regulator such as this could have the power to bring in something like flat fares
should they choose to do so. Brendan agreed that if they did bring in flat fares it would
simplify things for workers at Dublin Bus as well as public-transport using citizens.
1. How were the requirements gathered (were users of public transport consulted with
surveys, etc.)?
Brendan explained that there was market research carried out that there was consultation
with the public about the system but it would be better if I asked Tim from the NTA about
this as he knows more about it. Interestingly they had an idea of what they wanted for the
LEAP system and this broadly was in line with the results of the market research. The
requirements were decided by the public transport operators. Brendan admitted that when
they began they should have looked at things from a flat-fares perspective but the lack of a
transport regulator at the time prevented this. Instead what actually happened was they
took this very big complex, system and tried to somehow map or squeeze in multiple
features and tickets onto the LEAP system. Brendan explained that this was a concern for
him because though it will technically work with the LEAP system it can be confusing.
1. From where does each requirement originate?
Requirements came from the government mandate.
1. How the current system matches up to the original requirements?
Brendan believes that the current LEAP system does broadly match up with the original
requirements. The cross-border interoperability was never achieved but perhaps it will be
sometime in the future.
1. Why did management decide to go with RFID cards for LEAP over an alternative
technology?
Brendan stated that the decision to go with RFID was a no-brainer. In 2002 smart cards were
in their infancy; however they had looked at what was going on in Finland, Hong Kong and
how London had experimented with RFID technology. We spoke for a little bit about digital
wallets such as the Google wallet and I asked whether the LEAP system could be integrated
with a Google wallet. Brendan explained that with any system certain authentication is
32
required, he stated that it is probably a possibility but currently they are trying to get the last
few features rolled out for LEAP and may examine this possibility after this. Brendan
explained that ‘near-field’ communications have been around for 6 or 7 years but haven’t
really taken off to a massive extent yet. He gave the example of Wireless Application
Protocol (WAP) which was hailed as being the next big thing at the time but it never took off:
https://en.wikipedia.org/wiki/Wireless_Application_Protocol
1. Was the Oyster system in the U.K. or any other country’s integrated ticketing system
used as a blueprint for the LEAP system?
Yes it was. At the time Oyster wasn’t up and running but the Irish government transportation
officials were aware of its existence and they discussed the system with their counterparts
in the U.K. They also looked at similar systems in Finland, especially in the small town of
Tampere which was one of the first to roll-out RFID technology:
http://joukkoliikenne.tampere.fi/en/
They also had contacts in Hong Kong who they liaised with and had a reasonable amount of
dialogue with.
1. In your opinion, what are the advantages and disadvantages of the LEAP system?
In Brendan’s opinion the advantages of the LEAP system is its convenience. You can have all
your tickets under a single brand. You can purchase public transport in a simpler fashion at a
reduced cost.
1. Any compromises they were forced to make (and how they decided on these
compromises).
Interestingly Brendan said he didn’t think there were any disadvantages to the system from
the public’s point of view. However from the operators point of view (their view), one
disadvantage would be the complexity of the system as a work in development. There are
more constraints on operators as being part of an integrated system contrasting to if they
just had their own ticketing system. I asked them how they get the information off the ticket
machines on the buses. Brendan explained that this happens at night when the buses go
back to the Bus Depot. The information is downloaded onto their central system and there is
overnight processing done which is then sent onto LEAP where they are reimbursed for
fares, etc. Brendan pointed out that the Dublin Bus system is an older legacy system which
33
they purchased in 2003 which doesn’t have live, online capabilities like Irish Rail does. If
they could change things they would look for a live system.
In terms of compromises, Brendan did not think they had to make too many. One
compromise would be having the freedom to introduce your own tickets (e.g. monthly,
weekly paper tickets, these have to be on LEAP now). Brendan asked why couldn’t there be a
family of cards available. Brendan explained that this was a stipulation of the government
mandate that all other prepaid tickets would have to be removed as these could be seen as
competing with LEAP. However Brendan never bought into this, surely if you could do this it
would give companies the freedom to introduce low cost tickets. For example you could have
special low cost tickets for tourists as opposed to them having to pay for the high cost of a
LEAP card and the expensive cost of the deposit that goes with it. So Brendan feels there
should have been a family of cards (did he mean plastic cards similar to LEAP or paper
tickets???). He pointed out that there still is no solution to the tourist predicament with
LEAP.
1. Anything they would have done differently/would like to do in the future.
If he could have done things differently he would have started with a blank sheet and said
‘OK, What does the city want in terms of a system?’. He would have looked at a cheaper,
simpler pricing structure/ policy and built the structure of LEAP around that. Instead they did
the reverse which was to have a sprawling mix of things. They are actually now saying ‘OK,
How can we pare things back now?’. Brendan believes that capping fares and offering
rebates will improve the system a lot.
If he had a wish list if would be to further simplify the ticketing range. Can we edge towards
fewer fares on buses as the current system is complicated for drivers and for customers.
With an aim to improve they may look at mobile phone integration – so LEAP may evolve in
this direction.
1. In your opinion, do you believe the LEAP system has been a success so far?
Brendan would say the LEAP system has been a success so far. There is a lot of benefit to
the system considering the range of features they had to squeeze into it. If public transport
users have a daily spend that is high currently, the capping feature will make a difference to
them immediately.
34
APPENDIX 3: LEAP SURVEY RESULTS:
35
36
37
38
APPENDIX 4:
EXAMPLE OF HOW DIFFERENT LEAP TAG-ON REQUIREMENTS AND DUBLIN
BUS FARE PRICING STRUCTURES MAY CAUSE USER CONFUSION:
LEAP CARD – Summary of Ticket Products Available:
PAY-AS-YOU GO E-PURSE:
· Dublin Bus (All services) – Tag on only - May require to interact with driver
· Irish Rail (Short Hop Zone only) – Daily and weekly (Monday-Sunday) capping applies
– Tag on and off
· Bus Eireann (limited to Greater Dublin Area –excepting 100X and 101) – Tag on only
via driver
· LUAS (All services) – Daily and weekly (Monday-Sunday) capping applies – Tag on and
off
· Matthews Coaches (Bettystown-Laytown-Dublin service) – Tag on only via driver
· Wexford Bus (All services) – Tag on only via driver
· Swords Express (All services) – Tag on only via driver
PERIOD PASSES:
Dublin Bus:
Rambler 5 Day Adult - Tag on only via Validator on righthand side
Rambler 30 Day Adult - Tag on only via Validator on righthand side
Rambler 5 Day Student – requires Student LEAP Card - Tag on only via Validator on
righthand side
Rambler 30 Day Student – requires Student LEAP Card - Tag on only via Validator on
righthand side
Rambler 1 Day Family - Tag on only via Validator on righthand side
39
Annual Bus Only Travelwide Ticket (requires personalised LEAP card) - Tag on only via
Validator on righthand side
Bus Eireann:
Daily (24 hours) Greater Dublin Area zonal unlimited travel tickets - Tag on only via
driver
Weekly (7 day) Greater Dublin Area zonal unlimited travel tickets - Tag on only via
driver
Wexford Bus:
Weekly 10 Journey Ticket Adult - Tag on only via driver
Weekly 10 Journey Ticket Student - Tag on only via driver
Swords Express:
10 Journey Ticket - Tag on only via driver
Dublin Bus LEAP FARES:
Adult Fares Cash Fare Leap Card Fare
Stages 1 to 3 €1.65 €1.40
Stages 4 to 7 €2.15 €1.90
Stages 8 to 13 €2.40 €2.10
Over 13 Stages €2.80 €2.45
Outer Suburban 1 €2.80 €2.45
Outer Suburban 2 €4.40 €4.00
Xpresso in 1 zone €3.25 €2.90
40
Xpresso in 2 zones €4.40 €4.00
Route 90 & 91 €1.80 €1.60
Route 236 €1.65 €1.40
City Centre fare €0.65 €0.55
Child Fares Cash Fare Leap Card Fare
Schoolchild * €0.80 €0.70
Stages 1 to 7 €1.00 €0.90
Over 7 Stages €1.25 €1.10
Outer Suburban 1 €1.25 €1.10
Outer Suburban 2 €2.10 €1.90
Xpresso in 1 zone €2.15 €1.90
Xpresso in 2 zones €2.50 €2.20
Route 90 & 91 €0.95 €0.90
Route 221 & 236 €0.95 €0.90
City Centre fare €0.65 €0.55
41
Airlink Fares
Adult €6.00 €6.00
Child €3.00 €3.00
Nitelink Fare
Adult €5.70 €5.00
42
top related