Bushfire Social Intelligence

Sifting through vast amounts of social media data is time-consuming, labour intensive and costly. On top of this, there are over 250 siloed systems used in various fire fighting activities today, therefore introducing another digital solution potentially adds additional complexity.

Designing a workflow for finding social media related to specific fires required considering many touch points, user groups and jobs to be done. Our team embedded themselves onsite with the customer, observing pain points, opportunities and unconscious habits which all contributed to the solution design.

Using a Human Centred Design approach allowed us to contextualise the needs of all roles within a fire agency and illustrate the end-to-end experience; highlighting the people, processes, tools and physical space required to identify and use public information.

We conducted user interviews and contextual enquiries to understand the current service; prototyping a new process that simulates real-life emergency use cases. We worked in timeboxed design sprints, leaning on co-creation workshops with SMEs to truly understand if the value of the service was well defined, challenges were met, and business needs balanced.

We worked in timeboxed design sprints, leaning on co-creation workshops with SMEs to truly understand if the value of the service was well defined, challenges were met, and business needs balanced. We leveraged our existing design system to deliver a solution that is both simple and effective. Identifying four features addressing key opportunity areas;

1. A social media feed that can scale to add infinite crowd-sourced information platforms.
2. A tailored dashboard allowing users to customise their view, focus on their area of operation, and prioritise workloads.
3. A settings page allowing agencies to continuously update the Machine Learning ingestion rules; continuously responding to up-and-coming social media trends.
4. Implementation of technologies like Machine Learning for object recognition, topic classification, and location pattern detection.

Designing a workflow for finding social media related to specific fires required considering many touch points, user groups and jobs to be done. Leveraging our service prototype and UX patterns, we turned a 2-hour workflow that took 4 different systems, and 3 separate lines of communication; to a service that could be performed on a single platform in 5-15 minutes. We replaced a reactive process with a more intuitive solution.

Our team embedded ourselves onsite with the customer, observing pain points, opportunities and unconscious habits which all contributed to the solution design. Leveraging our established design system, we delivered a solution that is both simple and effective.

Significant effort has been placed to ensure the Dashboard is malleable and able to be viewed in multiple contexts. We achieved this by visualising social media spatially, on a global feed and tailored incident dashboards.

The feed aggregates posts from various social media platforms into a single view, using ML to identify relevant content through location inference, keyword matching, and image recognition. The feed can be customised to each user, allowing them to focus on specific incidents or areas they are responsible for.

All location-based posts are displayed geospatially, providing context with other data like surface temperature, fire spread predictions, people movement, and aerial imagery. Posts are then validated as intelligence and shared with the appropriate users. To ensure scalability and adaptability, we added a global filters setting page, allowing fire agencies to continuously update the parameters of the ML model within the platform.

The Dashboard is underpinned by image recognition, machine learning, and computer vision technology; where social media posts featuring fire or smoke are identified. The model also learns from a firefighters' interactions to become smarter over time; consequently, the more our users engage with the Dashboard, the smarter it becomes.

The Athena platform is a world-first, providing an end-to-end solution for generating bushfire intelligence. It is the single integration point for existing and siloed systems used by firefighting agencies today; automatically logging, structuring and building a layer of insights generated from Machine Learning (ML) and custom algorithms. The platform is built on a cloud-based architecture making it infinitely scalable, and as a Progressive Web App - globally accessible.

The Social Intelligence Dashboard bolsters an already industry-leading set of Bushfire Intelligence features. It aggregates posts from various social media platforms into a single view, using ML to identify relevant content through location inference, keyword matching, and image recognition. The feed can be customised to each user, allowing them to focus on specific incidents or areas they are responsible for.

All location-based posts are displayed geospatially, providing context with other data like surface temperature, fire spread predictions, people movement, and aerial imagery. Posts are then validated as intelligence by a human operator and automatically shared with the appropriate users involved in managing the related area.

With the addition of the Social Intelligence Dashboard to the Athena platform, users can add data from public information to better perform high-processing power tasks such as predicting bushfire spread or intensity, infrastructure impact analysis and social vulnerability indexing in seconds. Ultimately, revolutionising analysis turnaround time. When all features are combined to provide firefighters with a centralised and distilled source of decision-making intelligence, we reduce their cognitive load and enable them to make faster, more well-informed decisions.

The Black Summer bushfires of 2019-2020 killed 33 people, displaced or killed an estimated 3 billion animals, destroyed 3,000 homes and 16 million hectares of land, causing almost 10% of Australia's GDP in damages. On top of this, a significant concern facing fire agencies nationally is the impending loss of lived experience and expert knowledge. The majority of fire agencies in Australia have an ageing membership who to date, rely on previous lived fire experience to inform their decision making.

To combat this, Athena is able to ingest almost limitless amounts of time-space geography data and sort it into a centralised knowledge repository that can be leveraged by existing firefighters, up-and-coming firefighters, multiple emergency agencies and emerging parties looking to contribute to combating the global climate crisis.

By streamlining a 2-hour, multi-system workflow into a single, collaborative platform that now takes just 5-15 minutes, our service has already reduced the time it takes to detect fires and has provided invaluable intelligence for faster fire suppression and management. As fire agencies keep using these features, the platform is able to learn from those actions and capture the lived experience of ageing firefighters.

During the Social Intelligence Dashboard’s first fire season, we were able to detect 2 unreported grass fires and provided intelligence for a further 43 grass and bushfires. With a single bushfire costing the NSW state government upward of $140 million in damages, the intelligence produced by the platform has been able to significantly reduce operational and recovery costs.

This implementation of award winning design practices paired with emerging technologies showcases the innovative and forward thinking impact designers from the state of Victoria can offer towards solving problems like these for Australia and the world.