Impact
Impact Summary
Working with Jakarta’s government disaster management agency and social media company Twitter, UOW researchers developed an open source platform, Petajakarta.org, to crowdsource and automate flood reporting. This technology reduced the time and cost of situational awareness, empowered citizens to report flood information through social media, and improved human adaptability to environmental disasters. The tool saved lives by enabling citizens and first responders to avoid and navigate safely through flood zones, and economic benefits were realised through cost savings from manual reporting and reductions in flood-related injury, property damage and disease. This model for emergency response is now seen as best practice by international agencies and has been further adopted by seven cities in four countries.
Related United Nations Sustainable Development Goals:
9. Industry, innovation and infrastructure
11. Sustainable cities and communities
15. Life on land
16. Peace, justice and strong institutions
Details of the Impact
PetaJakarta.org was the result of a world-first joint pilot study initiated by researchers at the SMART Infrastructure Facility, University of Wollongong (UOW), in collaboration with the Jakarta Emergency Management Agency (BPBD DKI) and social media giant, Twitter. Jakarta is regularly devastated by flooding during the annual monsoon. Its infrastructure lagged behind its growth rate, making it vulnerable to natural disasters, including severe, city-wide flooding. As the tenth largest city in the world, and with 80% of the population active Twitter users, there was an opportunity for Twitter to have significant impact on flood management there.
UOW researchers developed open source software, known as CogniCity, to detect, in real time, residents’ Twitter conversations about flooding via their location-enabled mobile devices. CogniCity sends an automated request to users for additional flood information and upon confirmation, displays the crowdsourced information on the web-based flood map, PetaJakarta.org, demonstrating the power of software-as-infrastructure for climate adaptation.
The design goal was to reduce computational time and create near real-time flood maps for citizens and emergency managers without compromising on useability or ease of use. Deployment in a real-world situation was critical, providing insight for stakeholders into broader issues, including user motivation, incentive mechanisms, participation patterns, and data accuracy.
The first pilot study was conducted in the 2014-15 monsoon season. Five major floods hit Jakarta, sparking more than 100,000 flood-related Twitter conversations and over 2.2 million Twitter impressions (tweets received by an account) in real time. PetaJakarta.org was able to accurately distil and map high-quality reports to 1,000 flood sites across the city, transforming Twitter into an emergency data-gathering and critical alert service ideal for disaster management. The resulting online map was viewed over 160,000 times in Jakarta during the monsoon season.
Prior to PetaJakarta.org, citizens frequently criticised BPBD DKI Jakarta for their slow response to natural disasters. The new platform, now fully integrated into the agency’s control room, helped rapidly identify and cross-verify hotspots of flooding inundation, reducing emergency map update times from 6 hours to 20 minutes. This resulted in a 94% efficiency gain in targeting emergency response to citizens, as well as improved decision-making in flood control operations.
PetaJakarta.org’s time-critical updates on flood conditions helped to save lives through improved response times and minimised fatalities due to ill-informed citizens navigating through flood zones. Data that compared the losses associated with similar floods before and during the intervention indicated that PetaJakarta contributed to lowering flood deaths by 34.2% and reducing damage costs by 30.2%. This is equivalent to an estimated saving of at least 13 human lives and AU$213 million in damage prevention over a two-year period.
The system’s impact on flood disaster victims in Jakarta was significant. Emergency service workers were able to better locate victims and minimise losses by using information victims had shared on PetaJakarta.org. The tool’s ease-of-use and clear, immediate benefit contributed to a high uptake, supporting citizens’ adaptability and resilience as environmental disasters increase due to climate change.
Benefits to Indonesian disaster management extended beyond the program. Facilitated by knowledge exchange and collaboration between research students at UOW and Universitas Indonesia, students at both universities gained professional experience during the pilot and worked closely with each other and BPBD DKI Jakarta, collecting data and surveying key infrastructure assets throughout the watershed.
Building on the Jakarta experience, lead researcher Dr Holderness emulated the PetaJakarta model at the Urban Risk Lab at Massachusetts Institute of Technology. With support from USAID, the project PetaBencana.id (Disaster Map Indonesia) – built on CogniCity – went live and is now used by both residents and emergency management agencies in the cities of Bandung, Surabaya and Semarang in Indonesia.
The benefits of PetaJakarta.org were highlighted as “best practice for automated crowdsourcing of reliable emergency response data” in the 2016 proposed US Federal Communications Commission ‘Rules Regarding the Emergency Alert System and Wireless Emergency Alerts’. The technology and methodology behind PetaJakarta.org has subsequently been adopted in various locations across the globe. It was implemented in Florida, US, during Hurricane Irma and Chennai, India, during the recent monsoonal season.
With support from the Australian National Data Service, data collected during the project were recorded in the Research Data Australia archive and included in the National Library of Australia. The open distribution of these data has been invaluable to scientists, engineers and researchers worldwide with interests in natural disaster management.
This research has enabled an innovative strategy to disaster management. Using the people-as-sensors paradigm, PetaJakarta.org effectively integrated Twitter with CogniCity, enabling effective citizen and government engagement and co-management of a megacity disaster. This earned PetaJakarka.org international recognition as best practice by the Red Cross and has informed their future disaster mitigation strategies.
Beneficiaries
- Citizens of Jakarta, Bandung, Surabaya and Semarang in Indonesia
- Citizens of Chennai, India; Broward County, Florida USA; and Mekong Delta in Vietnam
- Emergency management agencies (e.g. BPBD in Jakarta)
- Jakarta Government
- Indonesian city governments in other affected cities
- World agencies with responsibility to identify and support victims of natural disasters, e.g., Red Cross
- Scientists, engineers and researchers in the fields of crisis informatics and disaster management
Countries
Impacted Countries
- India
- Indonesia
- United States
- Vietnam
Approach to Impact
Summary of the approaches to impact
UOW is committed to research focussed on improving society and creating public benefit. Our approach to impact translates innovations (often at a local level) into scalable, people-centric and sustainable solutions (with global reach). We do this through applied research facilities, impact-oriented seed funding, end-user driven research projects and a commitment to end-user training, collaborative implementation and proactive dissemination of research. Through this support we have developed a strong, collaborative culture that produces research that benefits government and citizens by enabling effective evidence-based planning, public policy and investment decision-making.
Approach to Impact
Our approach to impact is centred on developing innovations into sustainable solutions and is best summarised by the theme of our flagship interdisciplinary research program UOW Global Challenges: ‘Transforming Lives and Regions’. At the university level we enable this through the establishment of interdisciplinary, end-user focused applied research centres and facilities, and seed funding for impact-oriented and interdisciplinary projects. We encourage a focus on end-user driven, collaborative research projects. At the researcher level we maximise impact with responsive research methods, training and implementation support for research end-users, student-mediated knowledge exchange, and a commitment to open access and proactive dissemination.
Research centres and seed funding
Our commitment to translation is demonstrated by the research centres and facilities we target to provide research, knowledge and tools that inform evidence-based planning, public policy and investment decision-making. The SMART Infrastructure Facility was launched to generate, publish and disseminate ideas that support greater understanding of the value, interconnection and interdependencies of infrastructure both public and private. SMART researchers regularly contribute to policy discussions and consultations on infrastructure topics – for example they contributed submissions to Federal Government Standing Committees on SMART ICT, Value Capture and the recent Enquiry into Cities. The PetaJakarta project was led by SMART researchers with SMART providing administrative support to host international visits and workshops for the project.
Our UOW Global Challenges Program, a university-wide, challenge-led interdisciplinary research program with the theme “Transforming Lives and Regions”, provides support for impact-oriented research. PetaJakarta.org received a total of $100K in seed and project funding from UOW Global Challenges in 2014-15, which enabled the project team to maintain an on-the-ground presence in Jakarta to develop relationships with Indonesian government agencies and facilitate a student project to survey resilience and biodiversity in relation to flooding and risk exposure at local evacuation centres.
We also provide funding for end-user driven research projects through internal industry linkage grants. For example a long-term research collaboration to improve electronic documentation for aged care providers began with a $4,000 internal industry linkage grant from the university and end-user in 2003. Support from the grant enabled our researchers to produce a landmark report on aged care workers’ willingness and capacity to use IT, which was used to support the Government’s decision to provide $153 million in grant payments to aged care homes to enable transition to new record management systems. UOW also provided a matching scholarship and a tuition waiver scholarship to support two PhD candidates to continue the Aged Care research.
End-user focused research to maximise impact
To keep research focused on clear outcomes for end-users we involve key stakeholders throughout the research process. In 2014, as part of PetaJakarta, eight members of the BPBD DKI Jakarta visited UOW and conducted an intensive 10-day planning and design workshop. This allowed our team to understand and respond to their requirements accurately. The project design and implementation were collaborative and engaged beneficiaries such as community leaders, citizen-user groups, industry partners, the BPBD and other government agencies. The Jakarta Governor, Basuki Tjahaja Purnama, was directly engaged to promote and officially launch the project in 2014. In addition, during the 2015/16 monsoon season, the UOW project team held training sessions on PetaJakarta.org for over 1000 government employees across the metropolitan village and district levels of government to ensure they were skilled in using the technology.
We work closely with government and industry to develop applied research solutions and are responsive when opportunities for impact arise. For example, the importance of the PetaJakarta project was identified early and we assigned Dr Etienne Turpin and Dr Tomas Holderness to work full-time on the project, and recruited Dr Robert Ogie to join this highly skilled team. The success of the Jakarta implementation opened opportunities for expansion and further implementation in other geographies, such as the US, with researchers Holderness and Turpin moving to the Massachusetts Institute of Technology.
Student involvement in research also plays an important role in our approach to knowledge transfer and professional development, supporting useful research translation into desired outcomes. In the PetaJakarta project, two UOW Student Flood Support Teams were deployed to Jakarta in 2014-15. They benefited from working closely with the BPBD DKI Jakarta and students from the Universitas Indonesia. Engineering student Dan Simpson, who participated in the project and twice travelled to Jakarta, was awarded the Robert Hope Memorial Prize for his work in international sustainability.
We are committed to proactive dissemination and open access to our research to increase the reach of our impact. The PetaJakarta project adopted an “open data, open source approach”, allowing the CogniCity software to be freely available for other cities to download and customise for their use, e.g. different languages and disaster contexts. In collaboration with the Australian National Data Service, research outputs were archived and given open access across the globe for the advancement of knowledge and research. Similarly, visionillawarra.org.au, a dynamic regional planning tool developed by the SMART Infrastructure Facility to provide a more integrated approach to local and regional planning, was created to enable evidence-based planning and integrated development across the Illawarra.
Associated Research
As a predominantly end-user driven research project, much of the relevant research and impact occurred simultaneously, with development of PetaJakarta.org resulting from foundational research knowledge led by Drs Turpin and Holderness of the SMART Infrastructure Facility (UOW), with implementation of the tool informing further research. Activities occurred from 2014-16, and data were collected during the monsoon months of November to March.
Initial research produced an innovative tool that took advantage of existing social media flood reports and used it to crowdsource, filter, and harness reports in real-time to improve situational awareness and decision-making in data-starved regions. This required integrating ethnographic user and software design research. Ethnographic research on Jakarta provided understanding of the information needed by key stakeholders and the socio-cultural factors to include in the proposed solution (CogniCity).
The research led to collaboration with Jeremy Pitt (Imperial College) and Frank Dignum (TU Delft). Further research, led by Robert Ogie (UOW), focused on methodologies to improve flood management by integrating social media flood data with a digitised network representation of the city-wide drainage and flood control infrastructure. With DFAT funding, first generation smart water-level sensors were developed and deployed in Jakarta’s waterways to complement data from citizens.
References
1. Holderness, T., and Turpin, E., (2015) White Paper – PetaJakarta.org: Assessing the roles of social media for civic co-management during monsoon flooding in Jakarta, Indonesia. SMART Infrastructure Facility, UOW. ISBN: 978-1-74128-249-8. Documentation of the project’s inception, technical solution design, implementation and rationale.
2. Perez, P., du Chemin, T.H., Turpin, E. and Clarke, R., 2015. Citizen-driven flood mapping in Jakarta: a self-organising socio-technical system. In Self-Adaptive and Self-Organizing Systems Workshops (SASOW), 2015 IEEE International Conference (pp. 174-178). The PetaJakarta.org system is portrayed as a self-organising socio-technical system.
3. Ogie R.I., Forehead H., Clarke R., Perez P. (2017) Participation patterns and reliability of human sensing in crowd-sourced disaster management. Information Systems Frontiers 1-16. Social media data collected by PetaJakarta.org was grouped into four types of human sensors based on reliability to guide social media use for disaster management.
4. Ogie, R.I., 2016. Adopting incentive mechanisms for large-scale participation in mobile crowdsensing: from literature review to a conceptual framework. Human-centric Computing and Information Sciences, 6(24), pp. 1-31. The case study PetaJakarta.org system demonstrates the usefulness of a conceptual framework for incentive adoption.
5. Ogie, R., Holderness, T., Dunbar, M. and Turpin, E., 2017. Spatio-topological network analysis of hydrological infrastructure as a decision support tool for flood mitigation in coastal mega-cities. Environment and Planning B: Urban Analytics and City Science, 44(4), pp.718-739. Using PetaJakarta.org, Jakarta’s drainage infrastructure is mapped.
6. Ali A.U., Ogie R. (2017) Social media and disasters: Highlighting some wicked problems [Leading Edge]. IEEE Technology and Society Magazine, 36(4): 41-43. Two wicked problems were raised in using social media data for disaster management: how to effectively judge the objectivity of social media data while preserving its communicative nature.
7. Red Cross (2015) World Disasters Report. This report included PetaJakarta.org as a case study, appraising the system and how it holds potential for other coastal mega-cities with high exposure to flood disasters. https://ifrc-media.org/interactive/wp-content/uploads/2015/09/1293600-World-Disasters-Report-2015_en.pdf
8. Holderness, T., and Turpin, E., (25 January 2016) How tweeting about floods became a civic duty in Jakarta. The Guardian. This discusses PetaJakarta.org’s implementation and impact on managing disasters for an international audience. https://www.theguardian.com/public-leaders-network/2016/jan/25/floods-jakarta-indonesia-twitter-petajakarta-org
9. Federal Communications Commission, USA (24 March, 2016) Rules regarding the emergency alert system and wireless emergency alerts. Rule 81 introduced the PetaJakarta.org project, flagging it as “best practice for automated crowdsourcing of reliable emergency response data…” https://apps.fcc.gov/edocs_public/attachmatch/FCC-16-127A1.pdf
10. Open Data Institute (2016) Peta Jakarta: real-time flood mapping in Jakarta. This blog provides a comprehensive review of PetaJakarta.org project, and its continuous extension at the Urban Risk Lab at MIT on the new platform PetaBencana.id. https://theodi.org/odi-showcase-peta-jakarta-real-time-flood-mapping-jakarta