Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Standing Committee on Infrastructure and Communications
21/08/2015
Smart information and communications technology in the design and planning of infrastructure

BROWN, Ms Tania, Chief Operating Officer, SMART Infrastructure Facility, University of Wollongong

DU CHEMIN HOLDERNESS, Dr Tomas, Research Fellow, SMART Infrastructure Facility, University of Wollongong

[12:22]

CHAIR: Welcome. Although the committee does not require you to give evidence under oath, I advise you that this hearing is a legal proceeding of the parliament and therefore has the same standing as proceedings of the House. As you would be aware, this is being broadcast. I invite you to make a brief opening statement before we proceed to discussion.

Ms Brown : Thank you for the opportunity to present here today. SMART Infrastructure Facility was funded by the federal government through the Education Investment Fund back in 2009. We are delighted to be here and to highlight some of our research, particularly in the disaster planning and remediation area, which we thought was relevant to the terms of reference of the committee. It has certainly been some research that Dr Holderness has been heavily involved in as co-principal investigator of the PetaJakarta Project.

Our submission highlighted some of our recommendations around open-source software and the importance of embracing social media and gathering crowd-sourcing data and how that can be turned into actionable information for decision makers, and then establishing some protocols in the disaster management space. To do this, we provided the case study on PetaJakarta. PetaJakarta means map Jakarta in Indonesian. We started that in 2013. We have received funding from the Department of Foreign Affairs and Trade via the Australia-Indonesia facility for Disaster Reduction for disaster reduction in Indonesia. So we have been working on the ground with the Jakarta disaster management agency, their equivalent of an SES, on how to implement this system and how to assist them in real time—map flooding by using our collaborative partner, Twitter. In Jakarta—not to steal all of Tomas's thunder—the equivalent of 25 million people are sending tweets like we send a text message. They tweet about where they are, where the flood is, what is happening to them in real time. Through the technology and the research we have created, we can now harness those tweets and map that in real time for that disaster agency so that they can start to be proactive, not just reactive.

I think it highlights the role social media can play in civic co-management, particularly during an emergency event. We get to use people as sensors, and they play an important role in leading to those decisions that can improve their lives. All of this is improved by an open-source platform whereby data is shared not just with the agency but with the citizens on the ground. I might leave it at that as an opening statement.

CHAIR: Tomas, did you want to comment now, or respond to questions?

Dr Du Chemin Holderness : No, I am fine.

Mr THISTLETHWAITE: How long has the Jakarta project been running?

Ms Brown : We started in 2013 and we have just concluded a year-long joint pilot study with the Jakarta disaster management agency and Twitter. We have now released a white paper, which we can forward to the committee, which Tomas and Dr Etienne Turpin wrote.

Mr THISTLETHWAITE: Did you say Twitter?

Ms Brown : Twitter is our collaborative partner, yes. They open the pipe to give us access to the tweets. I think us and the FIFA World Cup are the only people they have ever done it for. We are pretty proud of that.

Mr THISTLETHWAITE: I am thinking that this could have benefits for Australia in terms of our overseas development aid budget and cutting down on wastage, cutting down on infrastructure projects that perhaps may not be appropriate for certain climates or conditions. What is your view on that?

Dr Du Chemin Holderness : I think one of the biggest findings we have seen from the PetaJakarta Project has been the creation of an ecosystem for disaster risk management in Jakarta—stepping away from the tweets themselves but working towards a system of interoperability whereby different agencies can talk to each other but also that ecosystem can foster development, both in the public and the private sector as well as in academia. It is about opening that data under standards-compliant metadata systems and then sharing it with people on the ground. It is as much about building trust in your smart ICT, as it were, among the citizens so that they understand what is going on and they understand the results of their contributing information in real time and how that is going to help them, as much as anything.

In terms of thinking about Australia's position in the region, as a leader in ICT and in disaster risk management, which we have here in Australia, we are in a very strong position to contribute to our partner nations in South-East Asia and the Pacific region. In terms of seeding some of that development, the recently opened Department of Foreign Affairs and Trade innovationXchange hub is a very good example—of trying to promote Australian concepts and ideas and innovation that can then be applied in the region.

Ms ROWLAND: My question relates to how you can translate this into an Australian level as well. Having spoken to some insurers, I think with flood mitigation expenditure there are fights between local councils and state governments about who should be responsible. It is one of those things that money gets spent on and no-one notices it—except when you do not do it. You do not get any thanks for it. I know this is about tracking disaster, but it is about disaster planning as well. Can you see any potential utilisation of this at an Australian level?

Dr Du Chemin Holderness : Yes, very much so. One of the other things we have been doing in the Jakarta project is not just crowd-sourcing information about the disaster in real time but looking at how we can better equip Jakarta and Jakarta's infrastructure to cope with the deluge that comes in the monsoon season. I think the potential for those components of the work, where we use a spatial network model, using open data collected by citizens that tells us how the infrastructure is connected near them, could work very well in Australia so that we could better understand how our infrastructures will cope with future extreme weather events. That could be, as you said, of potential interest to the insurance industry.

Ms Brown : The tool we worked on gathers tweets that mention the word 'banjir', which means flood in Indonesian. The tool can be translated for bushfires, typhoons or any key word that you want to gather information on. It is not limited to flooding. The issue is that it is a megacity tool. You need a volume of tweets to get reliable information back. We looked at the Sydney bushfires a few years ago, and I think there were about 200 tweets such as 'Oh, gee, there's a bushfire happening', whereas if something is happening where there are 25 million people—I think we had 100,000 tweets in 24 hours. So, there is a volume; it is a megacity tool. But I think it is growing as Twitter grows and as other social media platforms grow, and around how we as a society respond to issues. People in Jakarta were actually tweeting us and saying, 'Help! I'm stuck!' And we are not an emergency management responder, so we immediately forwarded that to the Jakarta agency. We do not want to get set up such that we are responding in real time to that. But the tool can help in planning for these events and how you manage them.

Ms MARINO: I was just looking at this from an Australian perspective, a bit like the other members here. Some evidence we had at a previous inquiry was about a mobile 3-D survey-grade laser scanning capability in Australia. If you look at that in the sense of being able to actually scope the front door of every building in Australia, when we talk about not just Australia but disaster management, knowing when and where the flood is going to reach and aligning that with the information that you could provide through Twitter and layer the two together, then the capability of managing disasters of a great variation would be very useful. That is why the issue around this type of technology, working across a number of fields, could be really useful. I wonder how you would see that type of mobile—survey grade—aligning with what you are doing and how that could apply within Australia for a number of applications.

Dr Du Chemin Holderness : I have two comments on that. The first is that it needs to be underpinned, as I previously mentioned, by an open data ecosystem. That laser scanning data is all well and good, but that information needs to be wrapped in an accessible, discoverable metadata standard that defines what that information is, and then it needs to be shared. I think we are moving towards a point where we obviously have leaders in-country in disaster risk management, quite rightly so—organisations like the SES. But to really harness the innovation of Australia's work in ICT, that data needs to be open, discoverable, so that other people can use it and so that you can do exactly what you just said to integrate with other layers of information. But that needs to happen in a way that the information is discoverable, accessible, and usable, and I am not sure that is currently happening. For example, in some of the other work we undertake at SMART, trying to understand the sustainability of different components of Sydney, we really struggle to access information like that, because it is all privately held.

Ms MARINO: That would have been my next question. Because the particular company has bought the technology themselves at significant cost and then the intellectual property that comes as a result of that, how do you see that being managed in this open data pursuit?

Dr Du Chemin Holderness : I think that is a tricky question to answer—

Ms MARINO: Yes, it is. It is one that I am particularly interested in.

Dr Du Chemin Holderness : I think you have to take a longer-term economic view of what innovation that data can yield by sharing it with many different partners in the private and public sector and academia. What are the returns on that over a longer time period than one company, for example, selling that data in small quantities or licensing that information in small quantities? If people do not even know that that company has that information, then they will not even start to be thinking about how they could potentially improve disaster risk management.

To come back to your previous point—what we want our smart cities to be—they should be all about people, increasing livability and also resilience for when these disasters come. But those processes need to happen across a whole wave of different sectors.

Ms MARINO: That is true; however, coming back to my question, there is this issue of an investment by a private entity. It is a cost that has come to their business. Who would pay? My question is: who do you see would pay for that investment and the result of that investment? We understand where it would be realised, across various sectors, but who do you see would pay for that particular type of information?

Dr Holderness : I think then we have to look at the question of whether that data is significantly beneficial for the public good. Obviously that company has made investment in that technology and acquiring that information, but here we are talking about different layers of data that exist within our cities and within the country. If we are going to move forward to develop smart ICT, those layers of information need to be accessible and be brought together. In short, I do not think I have an answer to who pays, but I think it would be interesting to ask and to maybe approach those companies that are doing that kind of work.

Ms MARINO: It is quite a critical part of this discussion.

Dr Du Chemin Holderness : There is a potential for something where you could have a non-commercial licence. For example, in Smart we have been quite successful in being able to collaborate with Twitter—a great example—on our PetaJakarta project. The data that we use is not openly available, but Twitter understands that we are using it in a real-time capacity to improve the response to flooding in Jakarta. So it is allowing us to send out all of these messages to people in Jakarta saying, 'Please confirm the situation on the ground,' which no or very few organisations have had access to. We have had access to both a private software and also a private dataset, but that company have released it to us free of charge, effectively, or as an in-kind cost to them because they can see the benefits and they can now see all of the innovative things and the hackathons that are being run and the innovation that is coming forward as a result of our work in the region. So it helps them build an ecosystem around their platform, and they have a licensing subscription service for that.

Ms MARINO: I can understand it in the sense of a platform like Twitter, but, if you get a small private entity that has literally gone out there and put its own money on the line, not a major entity like Twitter, there are some real issues for it because it needs a return on its investment.

Mr VAN MANEN: My question goes back to disaster management. A lot of our councils have done a lot of work around flood mapping. Up north, they would have mapping of various cyclone tracks, damage et cetera. How open and how available is that information to you in an Australian context to look at how what you are working on can be transferred to Australian situations?

Dr Du Chemin Holderness : I am not sure I feel comfortable commenting to that directly, because the project for us has been focused in Indonesia, and in Jakarta specifically, so it is not something that we have investigated. We did do a very preliminary study with some of the most recent tropical cyclones, and there is data out there. For example, the Bureau of Meteorology provides the cyclone tracks. So I think the information is there. What needs to be investigated more, like you say, is the availability of that information. Is it all in one place? Where is that information, how discoverable is it and how easily accessible is it? Some of those datasets might not necessarily be as discoverable as we would like them to be to really be able to bring everything together, and often that needs to happen in real time or near real time in a disasters context.

Ms Brown : In some of our other research projects we have been very fortunate that, as an academic institution, agencies like Sydney Water and Endeavour Energy have given us access to consumption data and so on in the past because it is being used in an academic context. It comes back to the proprietary ownership issue. Trying to get data to use in a commercial environment makes it a lot more complex, but we have not really focused in on flooding in Australia.

CHAIR: Can I just refer to some of your recommendations where you say we need to adapt standard metadata formats and metadata standards. You are saying we do not have those now. Do you have examples of who does have them and what you consider best practice?

Dr Du Chemin Holderness : Those standards exist and are being used.

CHAIR: We do not have them here? Are you saying we need to adapt them as a recommendation?

Dr Du Chemin Holderness : Yes, we are recommending that they be adapted wider across Australia.

CHAIR: So they would be basically mandated?

Dr Du Chemin Holderness : Yes, to some extent. The work that we are describing here, as part of SMART's federal mandate, is to make research data more available to other researchers both in Australia and around the world. An example, a case study, that we present is our SMART metadata catalogue that builds on an Australia-New Zealand standard for geospatial information, ISO 19115. That allows you to describe the information that we have, and that includes describing whether it is available—so it may be private data or commercially-sensitive data that we cannot share, but at least we are sharing the fact that we have it and that these are the partners that we are working with—all the way through to open data that anyone can download and use for their own research.

Our recommendation here is that we found that very successful in fostering collaborations both in the private and the public sectors with SMART because people can see the information that we have. Tying the information and the data that we have to our research outputs is becoming increasingly important so that we, in this age of big data, can back up our findings. If we make a claim on something, we can refer you to the repository. Whilst maybe you cannot download all the data, you can at least see some of it or a subset or understand the type of information that it is. The recommendation is that that standard be adopted in more organisations.

CHAIR: So your SMART metadata catalogue identifies best practice standards that you think operate better than others, or should we ask you to respond to that?

Dr Du Chemin Holderness : It is an example of best practice where you have a large suite of geospatial data.

CHAIR: Thank you very much.

Mr VAN MANEN: We are here at NICTA. How much of the work that you are doing is relying on data that you can obtain from other open source avenues?

Dr Du Chemin Holderness : I would say it is a significant proportion. A really good example of that is the data that is in OpenStreetMap. We use the data in OpenStreetMap for projects both here in Australia and for projects in Indonesia and elsewhere. That is a really nice example of citizen crowd source data that is available on an open licence. It has almost untold benefits for that. We use that to look at things like transportation access. We are starting now to see how we can map the tweets on top of transportation so that we can improve the routing of the vehicles that have emergency supplies to those areas that are most affected. For us as researchers, because all of that information is in the public domain, that is brilliant. That is another example of a set of data that has been published under an open licence. But it also has significant metadata around it, so that we can understand what is in there.

Mr VAN MANEN: What lessons or outcomes have you learnt from the Jakarta work that can be extended out to other areas within Asia that are prone to similar conditions? We have seen various flooding events throughout Asia on a regular basis and even here in Australia.

Dr Du Chemin Holderness : The biggest finding for us is the application of an open source system that can handle both potentially private and sensitive data such as individual tweets, which are potentially very sensitive—people asking for help—but using a series of software that is open source so that anyone can see what is happening inside that software. Using that software in that region has been key for us to unlock partnerships with the Indonesian government and to establish trust with the government agencies, with our private partners and also citizens on the ground so that, whilst there are commercially sensitive things in there, people can at least see what is going on and have some trust in what we are doing with that data, because it is their data at the end of the day.

CHAIR: Thank you very much for attending our public hearing today. The secretariat will send you a draft transcript of proceedings so requests can be made to correct any errors of transcription. You did refer to a white paper in response to Ms Marino so, if we could have a copy of that, that would be great. I assume your SMART metadata catalogue is online in this day and age?

Dr Du Chemin Holderness : Yes.

CHAIR: Excellent! Could we have a link so that our committee can access that as well. If there is anything else that you want to add, please feel free to respond to other evidence that you hear. Thank you very much for participating today.

Ms Brown : Thank you for the opportunity. We would be delighted to share the link to the SMART metadata system and to share the white paper that Dr Holderness and Dr Turpin have put a lot of time and investment into. We are very proud of it.

CHAIR: Thank you.