AI as a parcel processing aid

Early problem recognition AI as a parcel processing aid

Published on 16.06.2021 by Thomas Berweger, Digital Analyst

Parcel volumes, which have been increasing for years, are boosting Swiss Post’s sales but pose an increasing challenge for parcel sorting centers. Artificial intelligence creates the opportunity to counteract problems in parcel processing at an early stage by detecting anomalies.

Swiss Post’s parcel centers play a central role in processing parcels on their journey from the sender to the recipient. Nearly half a million parcels are processed every night by Swiss Post for delivery to their recipients the following day. This number doubles over the Christmas period. The increasing complexity of parcel logistics and the associated increase in data volumes pose a major challenge when it comes to monitoring the data generated at the parcel centers. The rapid detection and handling of problems is therefore of fundamental importance.

The sooner a problem is identified, the faster it can be responded to, keeping delays, outages and additional workloads to a minimum. For example, a sudden increase in parcels that are not sorted in time. This requires a rapid response.

If certain parameters suddenly deviate significantly from the norm, this may indicate a possible problem with parcel sorting. Intelligent algorithms can be used here. Obvious events can be identified in the data, enabling Swiss Post to correct issues in the sorting centers as early as possible. Algorithms from the field of data science can detect anomalies in large data quantities, including many attributes that cannot be detected by manual monitoring. Machine learning is also being used in other sectors to identify outliers (e.g. in fraud detection when using credit cards).

A proof-of-concept by Swiss Post has demonstrated that outliers can be identified in parcel sorting data using various machine learning techniques. The use of an automated alarm system based on these kinds of algorithms that will sound an alarm in the event of relevant data outliers is therefore conceivable in the future. This would make it possible to rectify problems early on, benefiting recipients, senders and Swiss Post as the carrier.

Thomas Berweger

Thomas Berweger is a digital analyst at Swiss Post. He uses data to generate added value for Swiss Post and its customers. His field of competence ranges from market and business development to efficiency improvements and the development of new products and services.

((commentsAmount)) Comments

There was an error during request.
  • (( comment.firstname )) (( comment.lastname )) (( comment.published )) (( comment.content ))

Contact us

Do you have questions for our experts, or do you need advice? We will be only too happy to help!

Contact us