NOTE: In the end I was unable to move the bot to a public domain for some reason even after following the instructions here (https://aai-covid19-botathon.devpost.com/updates/12930-enterprise-a2019-guidelines). I have included set of images that show every line in the bot and their setting that can be found in the repository linked to this project.
This submission is for the Travel, Supply Chain & Logistics category.
I experienced a lot of problems with delivery during Covid-19. Many inexperienced delivery people are hired to offset the huge increase in orders. They often lose packages, deliver them incorrectly, or skip your house. The bootstrap shipping method also means I rarely get notified when my package arrives. I wanted a way to help fix this.
What it does
It allows delivery orders to be tracked by just sending an email. The delivery driver takes a picture of where they left the order and sends it by email to the company they deliver for (This means that delivery drivers don't need to install an additional app). From there, an automation anywhere bot gets triggered and downloads the image and the metadata from the email. It is able to extract the gps location from the image metadata, and the order number using OCR. If the product is shipped in the correct location, it sends an email to the customer saying their order is there for pickup. If the location is incorrect, it sends an email to the delivery person that they made a mistake. All of this is updated in a customer order spreadsheet.
How I built it
Automation Anywhere with python scripts to handle the metadata scrape and address to gps coordinate conversion.
Challenges I ran into
Learning the automation anywhere tools and debugging some errors
Accomplishments that I'm proud of
Getting the entire project completed and working at 100%
What's next for Bot-O-Ship
Creating a map to follow the delivery route of the driver to determine if they are doing their job efficiently. This can be done using the gps metadata from the images they take for every order.