Client story City of Whittlesea floods data platform with IoT information, optimising efficiency and building value
By Insight Editor / 26 Nov 2020 / Topics: Intelligent edge
By Insight Editor / 26 Nov 2020 / Topics: Intelligent edge
Australia’s local councils manage a broad array of services for ratepayers and residents – ranging from road and park maintenance to childhood immunisation.
Councils aren’t flush with cash – they constantly seek out value and efficiency – but they are also expected to meet rising ratepayer expectations about quality of service or responsiveness. It’s why many are digitally transforming, recognising data as the ally of efficiency.
That’s certainly the case at the City of Whittlesea (CoW), one of Australia’s fastest growing municipalities with around 220,000 residents spread across 490 square kilometres, stretching from Melbourne’s northern suburbs out to its more rural fringe.
CoW is undertaking an Internet of things (IoT) driven digital transformation that promises to radically change the sorts of services that the council can provide, and the value it delivers for residents.
Hans Wolf, chief information officer for CoW, says that what he wanted to foster was a data collection and analytical capability that would help improve decision making across the council – both simple process related decision making and longer-term strategic planning.
At the simplest level this could be about scheduling park waste bin collections – so that instead of crews emptying bins even if they were less than a quarter full just because that was what the roster called for that day, they would be alerted only when a bin needed emptying. At a more strategic level, Wolf saw the opportunity to collect environmental data from the Council’s suburban districts and compare that with data collected on the rural fringe, that could be used to optimise urban planning based on a finer grained understanding of the environment.
Wolf explains that previously; “We didn’t have any data on the use of some of the assets that councils provide. We didn’t really know…if you build a park, how do you know how many people are using that park, whether there’s a playground in it, whether that makes a difference to how many people use that park? “Council would make decisions based on a lot of experience and a lot of intuition,” he says.
There is also the opportunity to be more transparent with residents thanks to the asset tracking sensors says Wolf. “We were keen to know where do our street sweepers go? People were forever saying, “Oh my street hasn’t been swept for ages,” but we’ve got data now that says, actually we did sweep your street. So it’s nice to be able to provide that back to the community.”
After securing a Smart Cities grant from the Federal Government, the City of Whittlesea worked with RMIT, La Trobe University and Minnovation Technologies to develop an solution that has also been rolled out across Moreland City Council, Banyule City Council, Mitchell Shire Council and Nillumbik Shire Council which share in both its cost and benefit.
The Northern Melbourne Smart Cities Network developed for the Councils centres on a LoRaWAN network that allows data, initially from five different types of sensors, to be collected. These first five sensor types are for people counting, air quality and environmental factors monitoring, water level monitoring, waste management collection and asset tracking.
Working with Microsoft partner Insight, CoW then designed a system that collects and stores that data in an Azure based data platform where it can be explored by Council staff using Power BI.
George Ibrahim, head of business insights and analytics at CoW has led the initiative to introduce self-service analytics and set up the data platform.
The IoT data is stored in the Azure data platform which has two zones – one for raw data, and then the curated section where data is indexed by Council, date and time, and orchestrated using Data Factory and Data Bricks. According to Ibrahim; “A lot of this is very plug and play within the Azure environment.”
The system was completed in June 2020 – so much of the recent sensor data reflects how parks or assets have been used during Victoria’s lockdown. As the State emerges from lockdown it will be possible to use data to assess how people’s activities are changing.
According to Wolf; “We’ll have the trends of what things were happening during the COVID period and what it was like outside of a COVID period.
“You can see people activity in some of our shopping strips. And then you can see the people activity completely fall away. It’s nice to be able to share that with the local traders, so they can get an understanding of when things are coming back up again from a pure data perspective. “
COVID’s impact will vary council by council; another benefit of using Azure to host the data platform is the ability to tag information in the shared resource, so that the costs of using the system can be fairly apportioned between the different councils.
Having a data platform also helps to reduce IT risk says Wolf. Because the CoW has moved to a cloud-first approach to computing, having an independent data collection to store both IoT data but also data from the Council’s operational systems – means that should CoW ever want to change cloud providers it has a copy of its data to hand.
The transparency that the IoT network and data platform are providing is leading to rising demand across council to deploy new sorts of sensors – for example to monitor feral animals, to check whether newly installed bird-boxes are actually being used by birds, and to track noise levels in specific areas where there have been resident complaints.
The IoT LoRaWAN network has also been made available for public use and Wolf says about 75 per cent of current traffic is non-Council related. “We are completely shocked there’s been such a big uptake of IoT within our municipality. We would never have anticipated that.”
Anyone is able to enrol their sensor onto the LoRaWAN network and direct how often data is sent to which location. All the sensor data on the network is kept separate and private and according to Wolf because the small amounts of data transmitted by sensors there are no capacity concerns.
One early non-council application for the network has emerged from the Tech Schools in Whittlesea, Banyule and Nillumbik where an after-school program for students in Years 7-10 is being planned to start in 2021 which will allow school students to run projects to track and monitor the health of waterways using a variety of sensors.
Inside Council the opportunity to investigate the data using Power BI has been widely embraced says Ibrahim, with around 40 people now trained to use the system.
“Half of them would probably be using it often. And they’re producing reports, and these reports are being published and scheduled,” he says. Previously a report might take two or three days to find the data and generate a report. Now reporting is increasingly self-service and based on up to the minute data.
It’s not just reporting that is being streamlined – Wolf is working toward increased process automation across Council.
“The ultimate goal, my end game for this is, we genuinely want a sensor to trigger a job. So we don’t want the community to tell us there’s a pothole on a road, we want a sensor to know it before the community even notices.
“Or if a waterway is about to burst its banks – we want to get our staff out there blocking off the bike paths and things like that so that the water doesn’t affect safety. We want to know all that stuff before the community tells us. We think that’s genuinely the best customer service we can provide, knowing about things and fixing things before anyone notices.”
City of Whittlesea (CoW) recently won the Smart City Achievement award at the MAV Technology Awards.