Melbourne-based consulting and advisory firm Revenite has played a central role in the development of a ‘digital twin’ of the Melbourne Cricket Ground (MCG) using data, artificial intelligence and more than 6,000 connected sensors.
Developed for the Melbourne Cricket Club (MCC), the MCG ‘digital twin’ will act as a central command centre for the playing field, giving managers the ability to take action in real-time to reduce customer queues for food and beverages, deliver instant security updates and manage maintenance and route foot traffic through gates and turnstiles.
The project is part of a broader engagement of Revenite, a Microsoft partner, by the MCC, with the two organisations having worked together for the past three years. As part of the project, more than 40 data sets are currently being migrated to Microsoft Azure Data Warehouse by the MCC, which manages the MCG.
It is expected that all of the MCC’s data sets will be loaded into its Azure Data Warehouse over the next one to three years, with catering, ticket sales, attendances, champion data, weather and event details, along with part of the customer relationship management system, already migrated into the warehouse.
Revenite provisioned the warehouse for the MCC and set up Stream Analytics to allow data to be collected in near real time. Meanwhile, Power BI supports data visualisation and Power Apps is being used to develop innovative data rich solutions for employees.
In partnership with Microsoft, Revenite has so far worked with the MCC to develop an attendance prediction model. Using turnstile data from the 6,000-plus internet of things (IoT) sensors, the MCC collects data on patrons’ arrivals and attendance.
This data and other variables are included in a predictive regression model to predict attendance, arrival patterns and members’ turn up rates.
According to Microsoft, the delta between the model’s estimate and actual attendance at the MCG is just 4.8 per cent on average. In the past, forecasts based on gut feel tended to be out by around 14 per cent, the software vendor claimed.
It is claimed that the increased accuracy has the potential to help the MCC with pre-event planning, during event decision-making and post-event analysis.
“What we are trying to do is bring factual information into running our events,” MCC business intelligence manager James Aiken said. “And then improve the customer experience.
“So, we’re looking at ways that we can use all types of data sets to estimate the crowd better, not just estimate the crowd size, but also when they’re going to turn up. So we have the right resources in the right place at the right time.
“We want to try to be a bit more dynamic, have this sort of real time information flowing through so we can make decisions in real time. Rather than waiting for something to actually happen, we can start to predict that something’s about to happen,” he added.