Overview
TowWarn is our proposed digital platform to send a friendly SMS reminder to drivers parked in a clearway during or just before operating hours that they need to move their vehicle before it gets towed. Unlike the current situation in which Transport Management Centre waits for government agencies and other service providers to phone in to report clearway violations, TowWarn digitally crowd-sources the task of detecting vehicles in clearways out to pedestrians and local businesses in a way that builds community spirit through friendly reminders rather than creating antagonism as a result of phoning to tow without further warning. For customers that chose not to opt-in to receiving tow notifications and have their vehicle towed, TowWarn allows them to check recent tows from an area by entering a customised signpost URL into a web-browser enabled smartphone without the need to install an app or turn on their GPS.
Submitted Proposal
Our idea proposal was submitted to the 2018 Clearways Innovation Challenge ran by Transport for NSW.
Mockups
Survey
To test whether our idea appealed to the target customers, we created an informal pilot survey using SurveyMonkey, and distributed it via a promoted Facebook post (targeted to users over 18 within a 50km radius of Sydney) and a promoted Twitter post (targeted to users from New South Wales). We incentivised participation through an informal title and a selection of giftcard choices.
If you left your 🚘 in a clearway and it was missing when you returned you'd... 🤔 (Giftcards available for awesome responses).
See survey results
In total we received 17 participants (14 via Faceboook, 3 via Twitter) who were surveyed on 5 multiple-choice questions about factors affecting design of the solution. Note that Q2 and Q5 allowed participants to select multiple answers. IP address geolocation reveals that 14 of the 17 participants were from NSW. The other 3 participants were using ISPs located in Victoria and Queensland.
Key findings- 87.5% of respondents would like SMS notifications to move their vehicle. Only 2 participants indicated that they would see this as an abuse of their contact info.
- 47% of participants indicated that they only turn their phone GPS on when they need it. This fed into our proposal by avoiding designs that would require participants to leave their GPS on permanently or wait for GPS signal.
We also received one response to the open question at the end of the survey:
An SMS or even a call if possible would be really good! Wherever my car is, I'm probably close by so would be able to move it if I accidentally left it in a clearway.
We acknowledge that there is a selection bias introduced through use of social media to promote the survey, thus part of our proposal is to run a larger formal survey and publish this in collaboration with user experience researchers.
FAQ
Q: Won't the sticker be hard to read for drivers overtaking a vehicle in a clearway?
Using phones while driving creates a safety risk, so we deliberately want to avoid anything that would encourage dangerous behaviour. By placing the sticker on the back of the sign, we are hoping to encourage reports from pedestrians, as well as encouraging friendly reminder reports from local businesses adjacent to a clearway (as vehicles parked near a business are potentially their own customers, friendly reminders that save their customers from being towed are expected to have a positive effect on the reputation of the business).
Q: Are the stickers all the same?
No. Each sticker encodes a unique identifier for the sign it is attached to so that the QR code / URL will take the user directly to the page for the section of clearway near the sign.
Q: Signs are mounted at least 2 metres above the curb, won't this make it difficult for people to read/scan the sticker?
We will experiment with the placement of the stickers, but they can potentially be placed anywhere and do not necessarily need to be on the sign itself. E.g. we could place a second (identical) sticker wrapped around the sign pole or on a nearby object to accommodate for human height differences.
Q: Why a Premium SMS based model?
The cost of sending SMS notifications in Australia is over 9 times higher than other countries such as the US (e.g. as of Nov 2018, Amazon Simple Notification Service charges only US$ 0.00645 per SMS in the US, but US$ 0.06097 per SMS in Australia). At 250 tows per week, and a conservative estimate of AU$ 0.09 per SMS, the cost of sending out SMS messages for all towed vehicles in NSW will be AU$ 22.50 per week. While this is still low compared to cost of development, these costs will grow once we factor in friendly reminders being sent out to vehicles that currently aren't being reported and the possibility of more notification types in future. Thus it is desirable to have a means to offset the cost of sending out SMS notifications to keep the system commercially viable.
A 19xx Premium SMS number provides a method to collect a small fee from customers who interact with the system without the need to inconvenience them with the need to provide payment details such as a credit card. Phone providers automatically charge this fee to the customer's phone bill when the customer interacts with the Premium SMS number, and is commonly used for scenarios such as SMS voting by TV shows. If our system were to use a Premium SMS number, we would receive a small payment whenever a driver texts back. For example, when they text back the code THANKS to indicate they will move their own vehicle. For customers on certain pre-paid plans that don't allow Premium SMS numbers, or have explicitly opted-out, we plan to offer equivalent functionality via a web browser, albeit less convenient.
A variant of this is using a Premium SMS number that people can text in order to conveniently report a vehicle by SMS (an additional benefit of this method is the small fee associated with reporting will deter people from spamming the system with inaccurate reports), or to SMS one's own vehicle location to the premium SMS number in return for a guaranteed SMS reminder to move it later when the clearway is about to come into operation.
Q: Do you have a commercial model beyond this?
The national economic cost of congestion in Australia is estimated to almost double from $16.5 billion in 2015 to $30 billion per year by 2030 (Bureau of Infrastructure, Transport and Regional Economics). We hope that by demonstrating the system in NSW as a community driven approach to manage clearway congestion, other states and territories will be keen to invest in funding future developments.
Commercially, this could be achieved by selling a data analysis tool targeted at traffic management centres and decision makers powered by the clearway violation reports collected through our system. Once a sufficiently large database of digital clearway violation reports are collected, these could be correlated with traffic flow data to learn traffic patterns indicative of a clearway violation, and to prioritise vehicles to tow based upon their estimated impact to traffic.
We also plan to introduce other types of notification types (when these are in the customer's interest, and the customer has consented to them) to help fill a gap in the market for geo-based reporting and monitoring. For example, shopping centres could set up QR stickers placed around their car parking area linked to our system to make it easier for the community to report issues in the car park. When someone reports an issue, such as a car blocking the truck delivery zone, the car owner will be directly notified to move (if they have opted-in to notifications). The shopping centre staff can also monitor for all messages associated with nearby parking areas so they can broadcast requests over their PA system in case the car owner doesn't notice the message. By charging the shopping centre a small recurring support fee for customised stickers linked to our system, we could ensure that use of the system remains free for the community to make reports and for vehicle owners to receive notifications.
Acknowledgements
A big thank-you to the open data forum team for calling out my (long) list of questions during the information session. I'd also like to thank Leonard Hoon from the Applied AI Institute for contributing his user experience expertise to explain issues associated with geolocation apps. Finally I'd like to thank everyone who took part in the survey—these responses helped me tailor the proposal towards building the solution NSW road users need rather than the idea I initially had in mind.