Australia Post trials machine learning to estimate parcel delivery times

By on
Australia Post trials machine learning to estimate parcel delivery times

Down to a two-hour window.

Australia Post is using machine learning to calculate the arrival time of parcels down to a two-hour window based on data from a new route planner that has just been rolled out to delivery drivers.

Executive general manager for transformation and enablement John Cox on Tuesday said the new feature is currently being trialled to give customers a more accurate estimate on delivery times.

“What we’re trialling – and this is not out in the public yet – is what we call an estimated time of arrival,” Cox told the Digital Transformation Agency’s 2020 Digital Summit.

“So based off when the postie scans the parcel in the morning to put in their van, we’ll be able to notify the consumer that it will be delivered within that window of time.”

The estimate is calculated using the “single scanning platform” that all deliver drivers use to scan their parcels before their run, which also determines the “optimal” route for each delivery run.

The “Android-based technology” platform has been deployed in recent years in place of “multiple devices using different technologies that were really difficult to deploy changes on”.

Cox said the postal service had already introduced routing functionality on the scanning platform in response to the pandemic to give delivery drivers “turn by turn instructions” about their journey.

The instructions were deemed necessary to help 2000-odd bike posties retrain as van drivers earlier this year to respond to the increased level of e-commerce.

“Going into a van and delivering parcels actually requires a much more dynamic route that they need to select because there might be four or five houses that they don’t stop at,” he said.

“And they also need to be trained differently about what is safe in a van versus what is safe when you’re travelling on the footpath.

“So what we did was we took our scanner app that we had built and gave very clear, turn by turn instructions that were built off machine learning models of what is the optimal way to go.

“So the posties who are in vans are now able to very accurately traverse the streets, deliver a parcel, and what it’s given us is a very accurate view of when they’re going to pass your house.”

Cox said that a two-hour delivery window would be published initially, even though AusPost has “got it down to a much, much smaller window”.

“The issue is that if I told you that I could get it to you within plus or minus 15 minutes, the expectation is that we’ll get it [there] in plus or minus 15 minutes,” he said.

“And we know that things do happen when posties are out on the road. There might be congestion …or there might be a need to stop and talk to a customer about something that has happened.

“So we want to build them a little flexibility there.”

AusPost is also using machine learning to predict parcel volumes by “plus or minus five percent on a daily basis”, and also for forcasting the Christmas demand.

“What we’ve been able to do is take the machine learning models, been able to take the scanner platform, we’ve been able to take the fact that in real-time that is now feeding up into the cloud, and then provide all of that insight and make sure that it’s accurate throughout the day,” Cox said.

“So we’ve taken this crisis, taken the investment that we’ve put in place, and then looked at how do we train people to be as productive as they can be using that data insight to drive through what we hope will be a really positive customer experience.

“It also has the benefits of enabling us to more accurately work out the sizes, the rounds in advance, the timeframes, staff appropriately, which is also a real challenge give the demand.”

Got a news tip for our journalists? Share it with us anonymously here.
Copyright © iTnews.com.au . All rights reserved.
Tags:

Most Read Articles

You must be a registered member of iTnews to post a comment.
| Register

Log In

Username / Email:
Password:
  |  Forgot your password?